JSDuck: Generated code documentation!
See CODING.md for how to run it.
Mistakes fixed:
* Warning: Unknown type function
-> Function
* Warning: Unknown type DOMElement
-> HTMLElement
* Warning: Unknown type DOM Node
-> HTMLElement
* Warning: Unknown type Integer
-> Mixed
* Warning: Unknown type Command
-> ve.Command
* Warning: Unknown type any
-> number
* Warning: Unknown type ve.Transaction
-> ve.dm.Transaction
* Warning: Unknown type ve.dm.AnnotationSet
-> ve.AnnotationSet
* Warning: Unknown type false
-> boolean
* Warning: Unknown type ve.dm.AlienNode
ve.dm doesn't have a generic AlienNode like ve.ce
-> Unknown type ve.dm.AlienInlineNode|ve.dm.AlienBlockNode
* Warning: Unknown type ve.ve.Surface
-> ve.ce.Surface
* ve.example.lookupNode:
-> Last @param should be @return
* ve.dm.Transaction.prototype.pushReplace:
-> @param {Array] should be @param {Array}
* Warning: ve.BranchNode.js:27: {@link ve.Node#hasChildren} links to non-existing member
-> (removed)
* Warning: ve.LeafNode.js:21: {@link ve.Node#hasChildren} links to non-existing member
-> (removed)
Differences fixed:
* Variadic arguments are like @param {Type...} [name]
instead of @param {Type} [name...]
* Convert all file headers from /** to /*! because JSDuck tries
to parse all /** blocks and fails to parse with all sorts of
errors for "Global property", "Unnamed property", and
"Duplicate property".
Find: \/\*\*([^@]+)(@copyright)
Replace: /*!$1$2
* Indented blocks are considered code examples.
A few methods had documentation with numbered lists that were
indented, which have now been updated to not be intended.
* The free-form text descriptions are parsed with Markdown,
which requires lists to be separated from paragraphs by an
empty line.
And we should use `backticks` instead of {braces} for inline
code in text paragraphs.
* Doc blocks for classes and their constructor have to be
in the correct order (@constructor, @param, @return must be
before @class, @abstract, @extends etc.)
* `@extends Class` must not have Class {wrapped}
* @throws must start with a {Type}
* @example means something else. It is used for an inline demo
iframe, not code block. For that simply indent with spaces.
* @member means something else.
Non-function properties are marked with @property, not @member.
* To create a link to a class or member, in most cases the name
is enough to create a link. E.g. Foo, Foo.bar, Foo.bar#quux,
where a hash stands for "instance member", so Foo.bar#quux,
links to Foo.bar.prototype.quux (the is not supported, as
"prototype" is considered an implementation detail, it only
indexes class name and method name).
If the magic linker doesn't work for some case, the
verbose syntax is {@link #target label}.
* @property can't have sub-properties (nested @param and @return
values are supported, only @static @property can't be nested).
We only have one case of this, which can be worked around by
moving those in a new virtual class. The code is unaltered
(only moved down so that it isn't with the scope of the main
@class block). ve.dm.TransactionProcessor.processors.
New:
* @mixins: Classes mixed into the current class.
* @event: Events that can be emitted by a class. These are also
inherited by subclasses. (+ @param, @return and @preventable).
So ve.Node#event-attach is inherited to ve.dm.BreakNode,
just like @method is.
* @singleton: Plain objects such as ve, ve.dm, ve.ce were missing
documentation causing a tree error. Documented those as a
JSDuck singleton, which they but just weren't documented yet.
NB: Members of @singleton don't need @static (if present,
triggers a compiler warning).
* @chainable: Shorthand for "@return this". We were using
"@return {classname}" which is ambiguous (returns the same
instance or another instance?), @chainable is specifically
for "@return this". Creates proper labels in the generated
HTML pages.
Removed:
* @mixin: (not to be confused with @mixins). Not supported by
JSDuck. Every class is standalone anyway. Where needed marked
them @class + @abstract instead.
Change-Id: I6a7c9e8ee8f995731bc205d666167874eb2ebe23
2013-01-04 08:54:17 +00:00
|
|
|
/*!
|
2013-01-15 23:38:49 +00:00
|
|
|
* VisualEditor DataModel Converter class.
|
2012-07-19 21:25:16 +00:00
|
|
|
*
|
2013-02-19 23:37:34 +00:00
|
|
|
* @copyright 2011-2013 VisualEditor Team and others; see AUTHORS.txt
|
2012-07-19 00:11:26 +00:00
|
|
|
* @license The MIT License (MIT); see LICENSE.txt
|
|
|
|
*/
|
|
|
|
|
2012-05-31 23:50:16 +00:00
|
|
|
/**
|
2013-01-15 23:38:49 +00:00
|
|
|
* DataModel converter.
|
|
|
|
*
|
|
|
|
* Converts between HTML DOM and VisualEditor linear data.
|
2012-05-31 23:50:16 +00:00
|
|
|
*
|
|
|
|
* @class
|
|
|
|
* @constructor
|
2013-02-16 02:37:50 +00:00
|
|
|
* @param {ve.dm.ModelRegistry} modelRegistry
|
2013-01-18 23:52:41 +00:00
|
|
|
* @param {ve.dm.NodeFactory} nodeFactory
|
|
|
|
* @param {ve.dm.AnnotationFactory} annotationFactory
|
2012-05-31 23:50:16 +00:00
|
|
|
*/
|
2013-02-21 23:01:04 +00:00
|
|
|
ve.dm.Converter = function VeDmConverter( modelRegistry, nodeFactory, annotationFactory, metaItemFactory ) {
|
2012-05-31 23:50:16 +00:00
|
|
|
// Properties
|
2013-02-16 02:37:50 +00:00
|
|
|
this.modelRegistry = modelRegistry;
|
2012-05-31 23:50:16 +00:00
|
|
|
this.nodeFactory = nodeFactory;
|
|
|
|
this.annotationFactory = annotationFactory;
|
2013-02-21 23:01:04 +00:00
|
|
|
this.metaItemFactory = metaItemFactory;
|
2013-04-10 20:32:33 +00:00
|
|
|
this.doc = null;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.documentData = null;
|
2013-04-10 20:32:33 +00:00
|
|
|
this.store = null;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.internalList = null;
|
2013-04-10 20:32:33 +00:00
|
|
|
this.contextStack = null;
|
2012-05-31 23:50:16 +00:00
|
|
|
};
|
|
|
|
|
2012-06-07 00:47:27 +00:00
|
|
|
/* Static Methods */
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get linear model data from a string optionally applying annotations
|
|
|
|
*
|
JSDuck: Generated code documentation!
See CODING.md for how to run it.
Mistakes fixed:
* Warning: Unknown type function
-> Function
* Warning: Unknown type DOMElement
-> HTMLElement
* Warning: Unknown type DOM Node
-> HTMLElement
* Warning: Unknown type Integer
-> Mixed
* Warning: Unknown type Command
-> ve.Command
* Warning: Unknown type any
-> number
* Warning: Unknown type ve.Transaction
-> ve.dm.Transaction
* Warning: Unknown type ve.dm.AnnotationSet
-> ve.AnnotationSet
* Warning: Unknown type false
-> boolean
* Warning: Unknown type ve.dm.AlienNode
ve.dm doesn't have a generic AlienNode like ve.ce
-> Unknown type ve.dm.AlienInlineNode|ve.dm.AlienBlockNode
* Warning: Unknown type ve.ve.Surface
-> ve.ce.Surface
* ve.example.lookupNode:
-> Last @param should be @return
* ve.dm.Transaction.prototype.pushReplace:
-> @param {Array] should be @param {Array}
* Warning: ve.BranchNode.js:27: {@link ve.Node#hasChildren} links to non-existing member
-> (removed)
* Warning: ve.LeafNode.js:21: {@link ve.Node#hasChildren} links to non-existing member
-> (removed)
Differences fixed:
* Variadic arguments are like @param {Type...} [name]
instead of @param {Type} [name...]
* Convert all file headers from /** to /*! because JSDuck tries
to parse all /** blocks and fails to parse with all sorts of
errors for "Global property", "Unnamed property", and
"Duplicate property".
Find: \/\*\*([^@]+)(@copyright)
Replace: /*!$1$2
* Indented blocks are considered code examples.
A few methods had documentation with numbered lists that were
indented, which have now been updated to not be intended.
* The free-form text descriptions are parsed with Markdown,
which requires lists to be separated from paragraphs by an
empty line.
And we should use `backticks` instead of {braces} for inline
code in text paragraphs.
* Doc blocks for classes and their constructor have to be
in the correct order (@constructor, @param, @return must be
before @class, @abstract, @extends etc.)
* `@extends Class` must not have Class {wrapped}
* @throws must start with a {Type}
* @example means something else. It is used for an inline demo
iframe, not code block. For that simply indent with spaces.
* @member means something else.
Non-function properties are marked with @property, not @member.
* To create a link to a class or member, in most cases the name
is enough to create a link. E.g. Foo, Foo.bar, Foo.bar#quux,
where a hash stands for "instance member", so Foo.bar#quux,
links to Foo.bar.prototype.quux (the is not supported, as
"prototype" is considered an implementation detail, it only
indexes class name and method name).
If the magic linker doesn't work for some case, the
verbose syntax is {@link #target label}.
* @property can't have sub-properties (nested @param and @return
values are supported, only @static @property can't be nested).
We only have one case of this, which can be worked around by
moving those in a new virtual class. The code is unaltered
(only moved down so that it isn't with the scope of the main
@class block). ve.dm.TransactionProcessor.processors.
New:
* @mixins: Classes mixed into the current class.
* @event: Events that can be emitted by a class. These are also
inherited by subclasses. (+ @param, @return and @preventable).
So ve.Node#event-attach is inherited to ve.dm.BreakNode,
just like @method is.
* @singleton: Plain objects such as ve, ve.dm, ve.ce were missing
documentation causing a tree error. Documented those as a
JSDuck singleton, which they but just weren't documented yet.
NB: Members of @singleton don't need @static (if present,
triggers a compiler warning).
* @chainable: Shorthand for "@return this". We were using
"@return {classname}" which is ambiguous (returns the same
instance or another instance?), @chainable is specifically
for "@return this". Creates proper labels in the generated
HTML pages.
Removed:
* @mixin: (not to be confused with @mixins). Not supported by
JSDuck. Every class is standalone anyway. Where needed marked
them @class + @abstract instead.
Change-Id: I6a7c9e8ee8f995731bc205d666167874eb2ebe23
2013-01-04 08:54:17 +00:00
|
|
|
* @static
|
|
|
|
* @param {string} text Plain text to convert
|
2013-03-20 22:35:05 +00:00
|
|
|
* @param {ve.dm.AnnotationSet} [annotations] Annotations to apply
|
2012-06-07 00:47:27 +00:00
|
|
|
* @returns {Array} Linear model data, one element per character
|
|
|
|
*/
|
2012-08-07 01:50:44 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText = function ( text, annotations ) {
|
2013-05-08 04:25:55 +00:00
|
|
|
var i, len,
|
2013-06-05 16:30:28 +00:00
|
|
|
characters = ve.splitClusters( text );
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2013-01-25 22:32:09 +00:00
|
|
|
if ( !annotations || annotations.isEmpty() ) {
|
2012-06-07 00:47:27 +00:00
|
|
|
return characters;
|
|
|
|
}
|
|
|
|
// Apply annotations to characters
|
2013-04-02 19:11:41 +00:00
|
|
|
for ( i = 0, len = characters.length; i < len; i++ ) {
|
2013-03-20 22:35:05 +00:00
|
|
|
// Just store the annotations' indexes from the index-value store
|
|
|
|
characters[i] = [characters[i], annotations.getIndexes().slice()];
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
|
|
|
return characters;
|
|
|
|
};
|
|
|
|
|
2013-04-02 19:11:41 +00:00
|
|
|
/**
|
|
|
|
* Utility function for annotation rendering. Transforms one set of annotations into another
|
|
|
|
* by opening and closing annotations. Each time an annotation is opened or closed, the associated
|
|
|
|
* callback is called with the annotation passed as a parameter.
|
|
|
|
*
|
|
|
|
* Note that currentSet will be modified, and will be equal to targetSet once this function returns.
|
|
|
|
*
|
2013-04-10 20:32:33 +00:00
|
|
|
* @static
|
2013-04-02 19:11:41 +00:00
|
|
|
* @param {ve.dm.AnnotationSet} currentSet The set of annotations currently opened. Will be modified.
|
|
|
|
* @param {ve.dm.AnnotationSet} targetSet The set of annotations we want to have.
|
|
|
|
* @param {Function} open Callback called when an annotation is opened. Passed a ve.dm.Annotation.
|
2013-07-16 19:32:28 +00:00
|
|
|
* @param {Function} close Callback called when an annotation is closed. Passed a ve.dm.Annotation.
|
2013-04-02 19:11:41 +00:00
|
|
|
*/
|
|
|
|
ve.dm.Converter.openAndCloseAnnotations = function ( currentSet, targetSet, open, close ) {
|
2013-07-26 15:41:29 +00:00
|
|
|
var i, len, index, startClosingAt, currentSetOpen, targetSetOpen;
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2013-04-02 19:11:41 +00:00
|
|
|
// Close annotations as needed
|
|
|
|
// Go through annotationStack from bottom to top (low to high),
|
|
|
|
// and find the first annotation that's not in annotations.
|
2013-07-02 20:22:09 +00:00
|
|
|
targetSetOpen = targetSet.clone();
|
2013-05-05 19:35:34 +00:00
|
|
|
for ( i = 0, len = currentSet.getLength(); i < len; i++ ) {
|
2013-07-26 15:41:29 +00:00
|
|
|
index = currentSet.getIndex( i );
|
2013-07-26 14:39:47 +00:00
|
|
|
// containsComparableForSerialization is expensive,
|
|
|
|
// so do a simple contains check first
|
|
|
|
if (
|
2013-07-26 15:41:29 +00:00
|
|
|
targetSetOpen.containsIndex( index ) ||
|
|
|
|
targetSetOpen.containsComparableForSerialization( currentSet.get( i ) )
|
2013-07-26 14:39:47 +00:00
|
|
|
) {
|
2013-07-26 15:41:29 +00:00
|
|
|
targetSetOpen.removeIndex( index );
|
|
|
|
} else {
|
2013-04-02 19:11:41 +00:00
|
|
|
startClosingAt = i;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if ( startClosingAt !== undefined ) {
|
|
|
|
// Close all annotations from top to bottom (high to low)
|
|
|
|
// until we reach startClosingAt
|
|
|
|
for ( i = currentSet.getLength() - 1; i >= startClosingAt; i-- ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
close( currentSet.get( i ) );
|
2013-04-02 19:11:41 +00:00
|
|
|
// Remove from currentClone
|
|
|
|
currentSet.removeAt( i );
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-07-02 20:22:09 +00:00
|
|
|
currentSetOpen = currentSet.clone();
|
2013-04-02 19:11:41 +00:00
|
|
|
// Open annotations as needed
|
2013-05-05 19:35:34 +00:00
|
|
|
for ( i = 0, len = targetSet.getLength(); i < len; i++ ) {
|
2013-07-26 15:41:29 +00:00
|
|
|
index = targetSet.getIndex( i );
|
2013-07-26 14:39:47 +00:00
|
|
|
// containsComparableForSerialization is expensive,
|
|
|
|
// so do a simple contains check first
|
|
|
|
if (
|
2013-07-26 15:41:29 +00:00
|
|
|
currentSetOpen.containsIndex( index ) ||
|
|
|
|
currentSetOpen.containsComparableForSerialization( targetSet.get( i ) )
|
2013-07-26 14:39:47 +00:00
|
|
|
) {
|
2013-07-02 20:22:09 +00:00
|
|
|
// If an annotation is already open remove it from the currentSetOpen list
|
|
|
|
// as it may exist multiple times in the targetSet, and so may need to be
|
|
|
|
// opened again
|
2013-07-26 15:41:29 +00:00
|
|
|
currentSetOpen.removeIndex( index );
|
|
|
|
} else {
|
|
|
|
open( targetSet.get( i ) );
|
|
|
|
// Add to currentClone
|
|
|
|
currentSet.pushIndex( index );
|
2013-04-02 19:11:41 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
2013-05-08 07:38:50 +00:00
|
|
|
/**
|
2013-05-18 03:49:25 +00:00
|
|
|
* Build an HTML attribute list for attribute preservation.
|
2013-05-08 07:38:50 +00:00
|
|
|
*
|
2013-05-18 03:49:25 +00:00
|
|
|
* The attribute list is an array of objects, one for each DOM element. Each object contains a
|
|
|
|
* map with attribute keys and values in .values, an (ordered) array of attribute keys in .keys,
|
|
|
|
* and an array of attribute lists for the child nodes in .children .
|
|
|
|
*
|
|
|
|
* @static
|
|
|
|
* @param {HTMLElement[]} domElements Array of DOM elements to build attribute list for
|
|
|
|
* @param {boolean|string|RegExp|Array|Object} spec Attribute specification, see ve.dm.Model
|
|
|
|
* @param {boolean} [deep=false] If true, recurse into children. If false, .children will be empty
|
|
|
|
* @param {Object[]} [attributeList] Existing attribute list to populate; used for recursion
|
|
|
|
* @returns {Object[]|undefined} Attribute list, or undefined if empty
|
2013-05-08 07:38:50 +00:00
|
|
|
*/
|
2013-05-18 03:49:25 +00:00
|
|
|
ve.dm.Converter.buildHtmlAttributeList = function ( domElements, spec, deep, attributeList ) {
|
|
|
|
var i, ilen, j, jlen, domAttributes, childList, empty = true;
|
|
|
|
attributeList = attributeList || [];
|
|
|
|
for ( i = 0, ilen = domElements.length; i < ilen; i++ ) {
|
|
|
|
domAttributes = domElements[i].attributes || [];
|
2013-06-14 06:17:08 +00:00
|
|
|
attributeList[i] = { 'values': {} };
|
2013-05-18 03:49:25 +00:00
|
|
|
for ( j = 0, jlen = domAttributes.length; j < jlen; j++ ) {
|
|
|
|
if ( ve.dm.Model.matchesAttributeSpec( domAttributes[j].name, spec ) ) {
|
|
|
|
attributeList[i].values[domAttributes[j].name] = domAttributes[j].value;
|
|
|
|
empty = false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if ( deep ) {
|
|
|
|
attributeList[i].children = [];
|
|
|
|
childList = ve.dm.Converter.buildHtmlAttributeList(
|
2013-06-17 22:41:22 +00:00
|
|
|
// Use .children rather than .childNodes so we don't mess around with things that
|
|
|
|
// can't have attributes anyway. Unfortunately, non-element nodes have .children
|
|
|
|
// set to undefined so we have to coerce it to an array in that case.
|
|
|
|
domElements[i].children || [], spec, deep, attributeList[i].children
|
2013-05-18 03:49:25 +00:00
|
|
|
);
|
|
|
|
if ( childList ) {
|
|
|
|
empty = false;
|
|
|
|
} else {
|
|
|
|
delete attributeList[i].children;
|
|
|
|
}
|
|
|
|
}
|
2013-05-08 07:38:50 +00:00
|
|
|
}
|
2013-05-18 03:49:25 +00:00
|
|
|
return empty ? undefined : attributeList;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Render an attribute list onto a set of DOM elements.
|
|
|
|
*
|
|
|
|
* Attributes set to undefined will be removed. The attribute specification restricts which
|
|
|
|
* attributes are rendered.
|
|
|
|
*
|
|
|
|
* @static
|
|
|
|
* @param {Object[]} attributeList Attribute list, see buildHtmlAttributeList()
|
|
|
|
* @param {HTMLElement[]} domElements Array of DOM elements to render onto
|
|
|
|
* @param {boolean|string|RegExp|Array|Object} [spec=true] Attribute specification, see ve.dm.Model
|
|
|
|
* @param {boolean} [overwrite=false] If true, overwrite attributes that are already set
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.renderHtmlAttributeList = function ( attributeList, domElements, spec, overwrite ) {
|
2013-06-14 06:17:08 +00:00
|
|
|
var i, ilen, key, values;
|
2013-05-18 03:49:25 +00:00
|
|
|
if ( spec === undefined ) {
|
|
|
|
spec = true;
|
2013-05-08 07:38:50 +00:00
|
|
|
}
|
2013-05-18 03:49:25 +00:00
|
|
|
if ( spec === false ) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
for ( i = 0, ilen = attributeList.length; i < ilen; i++ ) {
|
|
|
|
if ( !domElements[i] ) {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
values = attributeList[i].values;
|
2013-06-14 06:17:08 +00:00
|
|
|
for ( key in values ) {
|
|
|
|
if ( ve.dm.Model.matchesAttributeSpec( key, spec ) ) {
|
|
|
|
if ( values[key] === undefined ) {
|
|
|
|
domElements[i].removeAttribute( key );
|
|
|
|
} else if ( overwrite || !domElements[i].hasAttribute( key ) ) {
|
|
|
|
domElements[i].setAttribute( key, values[key] );
|
2013-05-18 03:49:25 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if ( attributeList[i].children ) {
|
|
|
|
ve.dm.Converter.renderHtmlAttributeList(
|
2013-06-17 22:41:22 +00:00
|
|
|
attributeList[i].children, domElements[i].children, spec
|
2013-05-18 03:49:25 +00:00
|
|
|
);
|
|
|
|
}
|
2013-05-08 07:38:50 +00:00
|
|
|
}
|
|
|
|
};
|
|
|
|
|
2012-05-31 23:50:16 +00:00
|
|
|
/* Methods */
|
|
|
|
|
2013-04-10 20:32:33 +00:00
|
|
|
/**
|
|
|
|
* Check whether this converter instance is currently inside a getDataFromDom() conversion.
|
|
|
|
*
|
|
|
|
* @method
|
2013-04-13 21:22:26 +00:00
|
|
|
* @returns {boolean} Whether we're converting
|
2013-04-10 20:32:33 +00:00
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.isConverting = function () {
|
|
|
|
return this.contextStack !== null;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get the IndexValueStore used for the current conversion.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {ve.dm.IndexValueStore|null} Current store, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.getStore = function () {
|
|
|
|
return this.store;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get the HTML document currently being converted
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {HTMLDocument|null} HTML document being converted, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.getHtmlDocument = function () {
|
|
|
|
return this.doc;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get the current conversion context. This is the recursion state of getDataFromDomRecursion().
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {Object|null} Context object, or null if not converting
|
|
|
|
*/
|
2013-05-06 11:34:32 +00:00
|
|
|
ve.dm.Converter.prototype.getCurrentContext = function () {
|
2013-04-10 20:32:33 +00:00
|
|
|
return this.contextStack === null ? null : this.contextStack[this.contextStack.length - 1];
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Get the annotations currently being applied by the converter. Note that this is specific to
|
|
|
|
* the current recursion level.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {ve.dm.AnnotationSet|null} Annotation set, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.getActiveAnnotations = function () {
|
|
|
|
var context = this.getCurrentContext();
|
|
|
|
return context ? context.annotations : null;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Whether the converter is currently expecting content. Note that this is specific to the current
|
|
|
|
* recursion level.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {boolean|null} Boolean indicating whether content is expected, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.isExpectingContent = function () {
|
|
|
|
var context = this.getCurrentContext();
|
|
|
|
return context ? context.expectingContent : null;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Whether the conversion is currently inside a wrapper paragraph generated by the converter.
|
|
|
|
* Note that this is specific to the current recursion level.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {boolean|null} Boolean indicating whether we're wrapping, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.isInWrapper = function () {
|
|
|
|
var context = this.getCurrentContext();
|
|
|
|
return context ? context.inWrapper : null;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Whether the active wrapper can be closed. Note that this is specific to the current recursion
|
|
|
|
* level. If there is no active wrapper, this returns false.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @returns {boolean|null} Boolean indicating whether the wrapper can be closed, or null if not converting
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.canCloseWrapper = function () {
|
|
|
|
var context = this.getCurrentContext();
|
|
|
|
return context ? context.canCloseWrapper : null;
|
|
|
|
};
|
|
|
|
|
2012-06-06 17:17:30 +00:00
|
|
|
/**
|
2012-06-08 04:58:56 +00:00
|
|
|
* Get the DOM element for a given linear model element.
|
|
|
|
*
|
2013-02-16 02:37:50 +00:00
|
|
|
* This invokes the toDomElements function registered for the element type.
|
2012-06-06 17:17:30 +00:00
|
|
|
*
|
|
|
|
* @method
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
* @param {Object|Array} dataElement Linear model element or data slice
|
2013-02-11 19:46:58 +00:00
|
|
|
* @param {HTMLDocument} doc Document to create DOM elements in
|
2013-07-16 21:40:05 +00:00
|
|
|
* @param {HTMLElement[]} [childDomElements] Array of child DOM elements to pass in (annotations only)
|
2013-01-15 23:38:49 +00:00
|
|
|
* @returns {HTMLElement|boolean} DOM element, or false if the element cannot be converted
|
2012-06-06 17:17:30 +00:00
|
|
|
*/
|
2013-07-16 21:40:05 +00:00
|
|
|
ve.dm.Converter.prototype.getDomElementsFromDataElement = function ( dataElements, doc, childDomElements ) {
|
2013-05-18 03:49:25 +00:00
|
|
|
var domElements,
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
dataElement = ve.isArray( dataElements ) ? dataElements[0] : dataElements,
|
2013-02-21 23:01:04 +00:00
|
|
|
nodeClass = this.modelRegistry.lookup( dataElement.type );
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2013-02-16 02:37:50 +00:00
|
|
|
if ( !nodeClass ) {
|
|
|
|
throw new Error( 'Attempting to convert unknown data element type ' + dataElement.type );
|
2012-11-24 02:44:54 +00:00
|
|
|
}
|
2013-04-17 17:53:26 +00:00
|
|
|
if ( nodeClass.static.isInternal ) {
|
|
|
|
return false;
|
|
|
|
}
|
2013-07-16 21:40:05 +00:00
|
|
|
domElements = nodeClass.static.toDomElements( dataElements, doc, this, childDomElements );
|
2013-07-16 21:53:06 +00:00
|
|
|
if ( ( !domElements || !domElements.length ) && !( nodeClass.prototype instanceof ve.dm.Annotation ) ) {
|
2013-02-16 02:37:50 +00:00
|
|
|
throw new Error( 'toDomElements() failed to return an array when converting element of type ' + dataElement.type );
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
2013-05-18 03:49:25 +00:00
|
|
|
if ( dataElement.htmlAttributes ) {
|
|
|
|
ve.dm.Converter.renderHtmlAttributeList( dataElement.htmlAttributes, domElements );
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
return domElements;
|
2012-06-07 00:47:27 +00:00
|
|
|
};
|
|
|
|
|
2013-02-21 23:01:04 +00:00
|
|
|
/**
|
|
|
|
* Create a data element from a DOM element.
|
2013-04-02 18:28:42 +00:00
|
|
|
* @param {ve.dm.Model} modelClass Model class to use for conversion
|
2013-02-21 23:01:04 +00:00
|
|
|
* @param {HTMLElement[]} domElements DOM elements to convert
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
* @returns {Object|Array|null} Data element or array of linear model data, or null to alienate
|
2013-02-21 23:01:04 +00:00
|
|
|
*/
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
ve.dm.Converter.prototype.createDataElements = function ( modelClass, domElements ) {
|
2013-05-08 04:25:55 +00:00
|
|
|
var dataElements = modelClass.static.toDataElement( domElements, this );
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
if ( !dataElements ) {
|
|
|
|
return null;
|
|
|
|
}
|
|
|
|
if ( !ve.isArray( dataElements ) ) {
|
|
|
|
dataElements = [ dataElements ];
|
|
|
|
}
|
2013-05-08 04:25:55 +00:00
|
|
|
return dataElements;
|
|
|
|
};
|
|
|
|
|
2012-06-06 17:17:30 +00:00
|
|
|
/**
|
2012-06-08 04:58:56 +00:00
|
|
|
* Build an HTML DOM node for a linear model annotation.
|
2012-06-06 17:17:30 +00:00
|
|
|
*
|
|
|
|
* @method
|
2012-06-08 04:58:56 +00:00
|
|
|
* @param {Object} dataAnnotation Annotation object
|
2012-10-06 00:25:57 +00:00
|
|
|
* @returns {HTMLElement} HTML DOM node
|
2012-06-06 17:17:30 +00:00
|
|
|
*/
|
2013-02-11 19:46:58 +00:00
|
|
|
ve.dm.Converter.prototype.getDomElementFromDataAnnotation = function ( dataAnnotation, doc ) {
|
2013-05-28 11:49:35 +00:00
|
|
|
var htmlData = dataAnnotation.toHtml(),
|
2013-02-11 19:46:58 +00:00
|
|
|
domElement = doc.createElement( htmlData.tag );
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2012-11-16 18:13:21 +00:00
|
|
|
ve.setDomAttributes( domElement, htmlData.attributes );
|
2012-10-06 00:25:57 +00:00
|
|
|
return domElement;
|
2012-06-06 17:17:30 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
2013-02-11 19:46:58 +00:00
|
|
|
* Convert an HTML document to a linear model.
|
|
|
|
* @param {HTMLDocument} doc HTML document to convert
|
2013-04-17 17:53:26 +00:00
|
|
|
* @param {ve.dm.IndexValueStore} store Index-value store
|
|
|
|
* @param {ve.dm.InternalList} internalList Internal list
|
2013-03-20 22:35:05 +00:00
|
|
|
* @returns {ve.dm.ElementLinearData} Linear model data
|
2013-02-11 19:46:58 +00:00
|
|
|
*/
|
2013-04-17 17:53:26 +00:00
|
|
|
ve.dm.Converter.prototype.getDataFromDom = function ( doc, store, internalList ) {
|
|
|
|
var linearData, refData;
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2013-04-10 20:32:33 +00:00
|
|
|
// Set up the converter state
|
|
|
|
this.doc = doc;
|
|
|
|
this.store = store;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.internalList = internalList;
|
2013-04-10 20:32:33 +00:00
|
|
|
this.contextStack = [];
|
2013-02-11 19:46:58 +00:00
|
|
|
// Possibly do things with doc and the head in the future
|
2013-04-17 17:53:26 +00:00
|
|
|
|
|
|
|
linearData = new ve.dm.ElementLinearData(
|
2013-03-20 22:35:05 +00:00
|
|
|
store,
|
2013-04-10 20:32:33 +00:00
|
|
|
this.getDataFromDomRecursion( doc.body )
|
2013-03-20 22:35:05 +00:00
|
|
|
);
|
2013-05-28 13:07:46 +00:00
|
|
|
refData = this.internalList.convertToData( this );
|
2013-04-17 17:53:26 +00:00
|
|
|
linearData.batchSplice( linearData.getLength(), 0, refData );
|
|
|
|
|
2013-04-10 20:32:33 +00:00
|
|
|
// Clear the state
|
|
|
|
this.doc = null;
|
|
|
|
this.store = null;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.internalList = null;
|
2013-04-10 20:32:33 +00:00
|
|
|
this.contextStack = null;
|
2013-04-17 17:53:26 +00:00
|
|
|
return linearData;
|
2013-02-11 19:46:58 +00:00
|
|
|
};
|
|
|
|
|
2013-05-15 17:00:10 +00:00
|
|
|
/**
|
|
|
|
* Wrapper for getDataFromDomRecursion which resets contextStack before the call
|
|
|
|
* and then set it back after the call.
|
|
|
|
*
|
|
|
|
* TODO: This is kind of a hack, better implementation would be more appropriate in near future.
|
|
|
|
*
|
|
|
|
* @method
|
|
|
|
* @param {HTMLElement} domElement HTML element to convert
|
|
|
|
* @param {Object} [wrapperElement] Data element to wrap the returned data in
|
|
|
|
* @param {ve.dm.AnnotationSet} [annotationSet] Override the set of annotations to use
|
|
|
|
* @returns {Array} Linear model data
|
|
|
|
*/
|
2013-07-02 20:22:09 +00:00
|
|
|
ve.dm.Converter.prototype.getDataFromDomRecursionClean = function ( domElement, wrapperElement, annotationSet ) {
|
2013-05-15 17:00:10 +00:00
|
|
|
var result, contextStack = this.contextStack;
|
|
|
|
this.contextStack = [];
|
|
|
|
result = this.getDataFromDomRecursion( domElement, wrapperElement, annotationSet );
|
|
|
|
this.contextStack = contextStack;
|
|
|
|
return result;
|
|
|
|
};
|
|
|
|
|
2013-02-11 19:46:58 +00:00
|
|
|
/**
|
2013-04-10 20:32:33 +00:00
|
|
|
* Recursive implementation of getDataFromDom(). For internal use, and for use in
|
|
|
|
* ve.dm.Model.static.toDataElement() implementations.
|
2012-06-06 17:17:30 +00:00
|
|
|
*
|
|
|
|
* @method
|
2013-02-11 19:46:58 +00:00
|
|
|
* @param {HTMLElement} domElement HTML element to convert
|
2013-04-10 20:32:33 +00:00
|
|
|
* @param {Object} [wrapperElement] Data element to wrap the returned data in
|
|
|
|
* @param {ve.dm.AnnotationSet} [annotationSet] Override the set of annotations to use
|
2012-06-08 04:58:56 +00:00
|
|
|
* @returns {Array} Linear model data
|
2012-06-06 17:17:30 +00:00
|
|
|
*/
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.prototype.getDataFromDomRecursion = function ( domElement, wrapperElement, annotationSet ) {
|
2013-05-01 23:00:33 +00:00
|
|
|
/**
|
|
|
|
* Add whitespace to an element at a specific offset.
|
|
|
|
*
|
|
|
|
* @param {Array} element Data element
|
|
|
|
* @param {number} index Whitespace index, 0-3
|
|
|
|
* @param {string} whitespace Whitespace content
|
|
|
|
*/
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
function addWhitespace( element, index, whitespace ) {
|
2013-05-01 23:00:33 +00:00
|
|
|
if ( !whitespace ) {
|
|
|
|
return;
|
|
|
|
}
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
if ( !element.internal ) {
|
|
|
|
element.internal = {};
|
|
|
|
}
|
|
|
|
// whitespace = [ outerPre, innerPre, innerPost, outerPost ]
|
2013-03-04 16:24:09 +00:00
|
|
|
// <tag> text </tag> <nextTag>
|
|
|
|
// ^^^^^^^^ ^^^^^^^^ ^^^^^^^^^ ^^^^^^^^^
|
|
|
|
// outerPre innerPre innerPost outerPost
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
if ( !element.internal.whitespace ) {
|
|
|
|
element.internal.whitespace = [];
|
|
|
|
}
|
|
|
|
element.internal.whitespace[index] = whitespace;
|
|
|
|
}
|
|
|
|
function processNextWhitespace( element ) {
|
|
|
|
// This function uses and changes nextWhitespace in the outer function's scope,
|
|
|
|
// which means it's not really a function but more of a shortcut.
|
|
|
|
if ( nextWhitespace !== '' ) {
|
|
|
|
addWhitespace( element, 0, nextWhitespace );
|
|
|
|
nextWhitespace = '';
|
|
|
|
}
|
|
|
|
}
|
2013-07-26 23:24:27 +00:00
|
|
|
// FIXME rewrite this horrible meta item / whitespace queueing/wrapping business
|
2013-04-18 23:06:58 +00:00
|
|
|
function outputWrappedMetaItems( whitespaceTreatment ) {
|
2013-05-08 04:25:55 +00:00
|
|
|
var i, len,
|
2013-07-26 23:24:27 +00:00
|
|
|
toInsert = [],
|
2013-05-08 04:25:55 +00:00
|
|
|
prev = wrappingParagraph;
|
|
|
|
|
2013-04-18 23:06:58 +00:00
|
|
|
for ( i = 0, len = wrappedMetaItems.length; i < len; i++ ) {
|
|
|
|
if ( wrappedMetaItems[i].type && wrappedMetaItems[i].type.charAt( 0 ) !== '/' ) {
|
|
|
|
if ( wrappedMetaItems[i].internal && wrappedMetaItems[i].internal.whitespace ) {
|
|
|
|
if ( whitespaceTreatment === 'restore' ) {
|
2013-07-26 23:24:27 +00:00
|
|
|
toInsert = toInsert.concat( ve.dm.Converter.getDataContentFromText(
|
2013-04-18 23:06:58 +00:00
|
|
|
wrappedMetaItems[i].internal.whitespace[0], context.annotations
|
|
|
|
) );
|
|
|
|
delete wrappedMetaItems[i].internal;
|
|
|
|
} else if ( whitespaceTreatment === 'fixup' ) {
|
|
|
|
addWhitespace( prev, 3, wrappedMetaItems[i].internal.whitespace[0] );
|
|
|
|
}
|
|
|
|
}
|
|
|
|
prev = wrappedMetaItems[i];
|
|
|
|
}
|
2013-07-26 23:24:27 +00:00
|
|
|
toInsert.push( wrappedMetaItems[i] );
|
|
|
|
}
|
|
|
|
if ( wrappedWhitespace !== '' && whitespaceTreatment === 'restore' ) {
|
|
|
|
// If we have wrapped whitespace, insert the wrapped meta items before it
|
|
|
|
// This is horrible and this whole system desperately needs to be rewritten
|
|
|
|
ve.batchSplice( data, wrappedWhitespaceIndex, 0, toInsert );
|
|
|
|
} else {
|
|
|
|
data = data.concat( toInsert );
|
2013-04-18 23:06:58 +00:00
|
|
|
}
|
|
|
|
wrappedMetaItems = [];
|
|
|
|
}
|
(bug 43056) Inline tags like <span> are block-alienated sometimes
This happens when the <span> is the start of unwrapped content. The
converter logic to look at the tag name in wrapping mode doesn't kick in
because we're not yet in wrapping mode at that point.
The core issue was that previously, we relied on the document
structure/state to choose between alienBlock and alienInline, and only
used the tag name where the document structure was ambiguous (wrapping).
Changed this to be the other way around: we now rely primarily on the
tag name, and if that doesn't match what we expect based on the document
structure, we work around that if possible. Specifically:
* inline tag in our wrapper --> inline alien
* block tag in our wrapper --> close wrapper, block alien
* inline tag in wrapper that's not ours --> inline alien
* block tag in wrapper that's not ours --> *inline* alien
* inline tag in structural location --> open wrapper, inline alien
* block tag in structural location --> block alien
* inline tag in content location --> inline alien
* block tag in content location --> *inline* alien
only in the fourth and the last case do we need to use the "wrong" alien type to
preserve document validity, and it will always be inline where block was
expected, which should reduce UI issues.
The condensed version of the above, which is used in the code, is:
* If in a non-wrapper content location, use inline
* If in a wrapper that's not ours, use inline
* Otherwise, decide based on tag name
* Open or close wrapper if needed
ve.dm.Converter:
* Replace isInline logic in createAlien() with the above
* Factor out code to start wrapping (was duplicated) into startWrapping()
* Call startWrapping() if createAlien() returns an alienInline and we're
in a structural location
Tests:
* Add test cases with aliens at the start and end of unwrapped content
** The first one failed prior to these changes and now passes, the
second one was already passing
* Fix about group test case, was exhibiting the bug that this commit fixes
Change-Id: I657aa0ff5bc2b57cd48ef8a99c8ca930936c03b8
2012-12-20 00:59:58 +00:00
|
|
|
function startWrapping() {
|
|
|
|
// Mark this paragraph as having been generated by
|
|
|
|
// us, so we can strip it on the way out
|
|
|
|
wrappingParagraph = {
|
|
|
|
'type': 'paragraph',
|
|
|
|
'internal': { 'generated': 'wrapper' }
|
|
|
|
};
|
|
|
|
data.push( wrappingParagraph );
|
2013-02-16 02:16:21 +00:00
|
|
|
context.inWrapper = true;
|
2013-01-31 20:07:56 +00:00
|
|
|
context.canCloseWrapper = true;
|
|
|
|
context.expectingContent = true;
|
(bug 43056) Inline tags like <span> are block-alienated sometimes
This happens when the <span> is the start of unwrapped content. The
converter logic to look at the tag name in wrapping mode doesn't kick in
because we're not yet in wrapping mode at that point.
The core issue was that previously, we relied on the document
structure/state to choose between alienBlock and alienInline, and only
used the tag name where the document structure was ambiguous (wrapping).
Changed this to be the other way around: we now rely primarily on the
tag name, and if that doesn't match what we expect based on the document
structure, we work around that if possible. Specifically:
* inline tag in our wrapper --> inline alien
* block tag in our wrapper --> close wrapper, block alien
* inline tag in wrapper that's not ours --> inline alien
* block tag in wrapper that's not ours --> *inline* alien
* inline tag in structural location --> open wrapper, inline alien
* block tag in structural location --> block alien
* inline tag in content location --> inline alien
* block tag in content location --> *inline* alien
only in the fourth and the last case do we need to use the "wrong" alien type to
preserve document validity, and it will always be inline where block was
expected, which should reduce UI issues.
The condensed version of the above, which is used in the code, is:
* If in a non-wrapper content location, use inline
* If in a wrapper that's not ours, use inline
* Otherwise, decide based on tag name
* Open or close wrapper if needed
ve.dm.Converter:
* Replace isInline logic in createAlien() with the above
* Factor out code to start wrapping (was duplicated) into startWrapping()
* Call startWrapping() if createAlien() returns an alienInline and we're
in a structural location
Tests:
* Add test cases with aliens at the start and end of unwrapped content
** The first one failed prior to these changes and now passes, the
second one was already passing
* Fix about group test case, was exhibiting the bug that this commit fixes
Change-Id: I657aa0ff5bc2b57cd48ef8a99c8ca930936c03b8
2012-12-20 00:59:58 +00:00
|
|
|
processNextWhitespace( wrappingParagraph );
|
|
|
|
}
|
2012-11-20 04:26:54 +00:00
|
|
|
function stopWrapping() {
|
|
|
|
if ( wrappedWhitespace !== '' ) {
|
|
|
|
// Remove wrappedWhitespace from data
|
2012-12-11 18:23:33 +00:00
|
|
|
data.splice( wrappedWhitespaceIndex, wrappedWhitespace.length );
|
2013-05-01 23:00:33 +00:00
|
|
|
// Add whitespace to the last sibling: either the last meta item or the wrapper paragraph
|
|
|
|
addWhitespace( wrappedMetaItems.length > 0 ? wrappedMetaItems[wrappedMetaItems.length - 2] : wrappingParagraph, 3, wrappedWhitespace );
|
2012-11-20 04:26:54 +00:00
|
|
|
nextWhitespace = wrappedWhitespace;
|
|
|
|
}
|
|
|
|
data.push( { 'type': '/paragraph' } );
|
2013-04-18 23:06:58 +00:00
|
|
|
outputWrappedMetaItems( 'fixup' );
|
2012-11-20 04:26:54 +00:00
|
|
|
wrappingParagraph = undefined;
|
2013-02-16 02:16:21 +00:00
|
|
|
context.inWrapper = false;
|
2013-01-31 20:07:56 +00:00
|
|
|
context.canCloseWrapper = false;
|
2013-04-10 20:32:33 +00:00
|
|
|
context.expectingContent = context.originallyExpectingContent;
|
2012-11-20 04:26:54 +00:00
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
function getAboutGroup( el ) {
|
2013-05-08 04:25:55 +00:00
|
|
|
var elAbout, node,
|
|
|
|
textNodes = [],
|
|
|
|
aboutGroup = [ el ];
|
|
|
|
|
2013-02-16 02:37:50 +00:00
|
|
|
if ( !el.getAttribute || el.getAttribute( 'about' ) === null ) {
|
|
|
|
return aboutGroup;
|
|
|
|
}
|
|
|
|
elAbout = el.getAttribute( 'about' );
|
|
|
|
for ( node = el.nextSibling; node; node = node.nextSibling ) {
|
|
|
|
if ( !node.getAttribute ) {
|
2012-11-08 02:03:05 +00:00
|
|
|
// Text nodes don't have a getAttribute() method. Thanks HTML DOM,
|
|
|
|
// that's really helpful ^^
|
2013-02-16 02:37:50 +00:00
|
|
|
textNodes.push( node );
|
2012-11-08 02:03:05 +00:00
|
|
|
continue;
|
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
if ( node.getAttribute( 'about' ) === elAbout ) {
|
|
|
|
aboutGroup = aboutGroup.concat( textNodes );
|
|
|
|
textNodes = [];
|
|
|
|
aboutGroup.push( node );
|
|
|
|
} else {
|
|
|
|
break;
|
2012-11-08 02:03:05 +00:00
|
|
|
}
|
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
return aboutGroup;
|
2012-11-08 02:03:05 +00:00
|
|
|
}
|
2013-07-15 11:06:31 +00:00
|
|
|
function isAllInstanceOf( data, targetClass ) {
|
|
|
|
var i, type, itemClass;
|
2013-06-18 19:00:49 +00:00
|
|
|
for ( i = data.length - 1; i >= 0; i-- ) {
|
2013-07-15 11:06:31 +00:00
|
|
|
type = ve.dm.LinearData.static.getType( data[i] );
|
|
|
|
if ( type ) {
|
|
|
|
itemClass = modelRegistry.lookup( type ) || ve.dm.AlienNode;
|
|
|
|
if ( !( itemClass.prototype === targetClass.prototype || itemClass.prototype instanceof targetClass ) ) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
} else {
|
2013-06-18 19:00:49 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
2012-11-08 02:03:05 +00:00
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
var i, childDomElement, childDomElements, childDataElements, text, childTypes, matches,
|
2013-02-16 02:37:50 +00:00
|
|
|
wrappingParagraph, prevElement, childAnnotations, modelName, modelClass,
|
2013-05-18 03:49:25 +00:00
|
|
|
annotation, childIsContent, aboutGroup, htmlAttributes,
|
2013-07-15 11:06:31 +00:00
|
|
|
modelRegistry = this.modelRegistry,
|
2012-08-02 18:46:13 +00:00
|
|
|
data = [],
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
nextWhitespace = '',
|
|
|
|
wrappedWhitespace = '',
|
2012-12-11 18:23:33 +00:00
|
|
|
wrappedWhitespaceIndex,
|
2013-04-18 23:06:58 +00:00
|
|
|
wrappedMetaItems = [],
|
2013-04-10 20:32:33 +00:00
|
|
|
context = {},
|
2013-05-08 04:25:55 +00:00
|
|
|
prevContext = this.contextStack.length ?
|
|
|
|
this.contextStack[this.contextStack.length - 1] : null;
|
2013-04-10 20:32:33 +00:00
|
|
|
|
|
|
|
context.annotations = annotationSet || (
|
|
|
|
prevContext ? prevContext.annotations.clone() : new ve.dm.AnnotationSet( this.store )
|
|
|
|
);
|
|
|
|
context.branchType = wrapperElement ? wrapperElement.type : (
|
|
|
|
prevContext ? prevContext.branchType : 'document'
|
|
|
|
);
|
|
|
|
context.branchHasContent = this.nodeFactory.canNodeContainContent( context.branchType );
|
|
|
|
context.originallyExpectingContent = context.branchHasContent || !context.annotations.isEmpty();
|
|
|
|
context.expectingContent = context.originallyExpectingContent;
|
|
|
|
context.inWrapper = prevContext ? prevContext.inWrapper : false;
|
|
|
|
context.canCloseWrapper = false;
|
|
|
|
this.contextStack.push( context );
|
|
|
|
|
2012-06-07 00:47:27 +00:00
|
|
|
// Open element
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( wrapperElement ) {
|
|
|
|
data.push( wrapperElement );
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
|
|
|
// Add contents
|
2012-08-02 18:46:13 +00:00
|
|
|
for ( i = 0; i < domElement.childNodes.length; i++ ) {
|
|
|
|
childDomElement = domElement.childNodes[i];
|
2012-06-07 00:47:27 +00:00
|
|
|
switch ( childDomElement.nodeType ) {
|
|
|
|
case Node.ELEMENT_NODE:
|
2013-05-03 22:01:06 +00:00
|
|
|
aboutGroup = getAboutGroup( childDomElement );
|
|
|
|
modelName = this.modelRegistry.matchElement( childDomElement, aboutGroup.length > 1 );
|
2013-02-16 02:37:50 +00:00
|
|
|
modelClass = this.modelRegistry.lookup( modelName ) || ve.dm.AlienNode;
|
|
|
|
if ( modelClass.prototype instanceof ve.dm.Annotation ) {
|
2013-05-08 04:25:55 +00:00
|
|
|
childDomElements = [ childDomElement ];
|
2013-04-10 22:13:50 +00:00
|
|
|
} else {
|
|
|
|
// Node or meta item
|
|
|
|
childDomElements = modelClass.static.enableAboutGrouping ?
|
|
|
|
aboutGroup : [ childDomElement ];
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
}
|
2013-05-08 04:25:55 +00:00
|
|
|
childDataElements = this.createDataElements( modelClass, childDomElements );
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
|
|
|
|
if ( !childDataElements ) {
|
|
|
|
// Alienate
|
|
|
|
modelClass = ve.dm.AlienNode;
|
|
|
|
childDomElements = modelClass.static.enableAboutGrouping ?
|
|
|
|
aboutGroup : [ childDomElement ];
|
|
|
|
childDataElements = this.createDataElement( modelClass, childDomElements );
|
|
|
|
} else {
|
|
|
|
// Update modelClass to reflect the type we got back
|
|
|
|
modelClass = this.modelRegistry.lookup( childDataElements[0].type );
|
Great Annotation Refactor of 2013
This changes the annotation API to be the same as the node API, sans
a few boolean flags that don't apply. The APIs were different, but
there was really no good reason why, so this makes things simpler for
API users. It also means we'll be able to factor a bunch of things out
because they're now duplicated between nodes, meta items and annotations.
Linear model annotations are now objects with 'type' and 'attributes'
properties (rather than 'name' and 'data'), for consistency with elements.
They now also contain html/0/* attributes for HTML attribute preservation,
which obsoletes the htmlTagName and htmlAttributes properties.
dm.Annotation subclasses take a reference to such an object and implement
conversion using .static.toDataElement and .static.toDomElements just
like nodes do. The custom .getHash() functions are no longer necessary
because of the way HTML attribute preservation was reimplemented.
CE rendering has been moved out of dm.Annotation (it never made sense to
have CE rendering functions in DM classes, this was bothering me) and into
separate ce.Annotation subclasses. These are very similar to CE nodes in
that they have a this.$ generated based on something in the DM; the main
difference is that nodes listen to events and update themselves, whereas
annotations are static and are simply destroyed and rebuilt when they
change. This change also adds whitelisted HTML attribute rendering for
annotations, as well as class="ve-ce-FooAnnotation" attributes.
Now that annotation classes produce real DOM nodes rather than weird
objects describing HTML tags, we can't generate HTML as a string in
ce.ContentBranchNode anymore. getRenderedContents() has been rewritten
to be much more similar to the way the converter renders annotations;
in fact, significant parts of it were copied from the converter, so that
should be factored out in the future. This change actually fixes an
annotation rendering discrepancy between ce.ContentBranchNode and
dm.Converter; see the diff of ve.ce.ContentBranchNode.test.js.
ve.ce.MWEntityNode.js:
* Remove stray property
ve.dm.MWExternalLinkAnnotation.js:
* Store 'rel' attribute
ve.dm.TextStyleAnnotation.js:
* Put all the conversion logic in the abstract base class
ve.dm.Converter.js:
* Also feed annotations through getDomElementsFromDataElement() and
createDataElement()
ve.dm.Node.js:
* Fix undocumented property
ve.ce.ContentBranchNode.test.js:
* Add descriptive messages for each test case
* Compare DOM trees, not HTML strings
* Compare without all the class="ve-ce-WhateverAnnotation" clutter
ve.ui.LinkInspector.js:
* Replace direct .getHash() calls (evil!) with ve.getHash()
Bug: 46464
Bug: 44808
Change-Id: I31991488579b8cce6d98ed8b29b486ba5ec38cdc
2013-04-02 17:23:33 +00:00
|
|
|
}
|
2013-04-10 22:13:50 +00:00
|
|
|
|
|
|
|
// Now take the appropriate action based on that
|
|
|
|
if ( modelClass.prototype instanceof ve.dm.Annotation ) {
|
2013-05-18 03:49:25 +00:00
|
|
|
htmlAttributes = ve.dm.Converter.buildHtmlAttributeList(
|
|
|
|
childDomElements, modelClass.static.storeHtmlAttributes
|
|
|
|
);
|
|
|
|
if ( htmlAttributes ) {
|
|
|
|
childDataElements[0].htmlAttributes = htmlAttributes;
|
|
|
|
}
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
annotation = this.annotationFactory.create( modelName, childDataElements[0] );
|
2013-02-16 02:37:50 +00:00
|
|
|
// Start wrapping if needed
|
2013-02-16 02:16:21 +00:00
|
|
|
if ( !context.inWrapper && !context.expectingContent ) {
|
(bug 43056) Inline tags like <span> are block-alienated sometimes
This happens when the <span> is the start of unwrapped content. The
converter logic to look at the tag name in wrapping mode doesn't kick in
because we're not yet in wrapping mode at that point.
The core issue was that previously, we relied on the document
structure/state to choose between alienBlock and alienInline, and only
used the tag name where the document structure was ambiguous (wrapping).
Changed this to be the other way around: we now rely primarily on the
tag name, and if that doesn't match what we expect based on the document
structure, we work around that if possible. Specifically:
* inline tag in our wrapper --> inline alien
* block tag in our wrapper --> close wrapper, block alien
* inline tag in wrapper that's not ours --> inline alien
* block tag in wrapper that's not ours --> *inline* alien
* inline tag in structural location --> open wrapper, inline alien
* block tag in structural location --> block alien
* inline tag in content location --> inline alien
* block tag in content location --> *inline* alien
only in the fourth and the last case do we need to use the "wrong" alien type to
preserve document validity, and it will always be inline where block was
expected, which should reduce UI issues.
The condensed version of the above, which is used in the code, is:
* If in a non-wrapper content location, use inline
* If in a wrapper that's not ours, use inline
* Otherwise, decide based on tag name
* Open or close wrapper if needed
ve.dm.Converter:
* Replace isInline logic in createAlien() with the above
* Factor out code to start wrapping (was duplicated) into startWrapping()
* Call startWrapping() if createAlien() returns an alienInline and we're
in a structural location
Tests:
* Add test cases with aliens at the start and end of unwrapped content
** The first one failed prior to these changes and now passes, the
second one was already passing
* Fix about group test case, was exhibiting the bug that this commit fixes
Change-Id: I657aa0ff5bc2b57cd48ef8a99c8ca930936c03b8
2012-12-20 00:59:58 +00:00
|
|
|
startWrapping();
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
prevElement = wrappingParagraph;
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
|
|
|
// Append child element data
|
2013-04-10 20:32:33 +00:00
|
|
|
childAnnotations = context.annotations.clone();
|
2013-01-25 22:32:09 +00:00
|
|
|
childAnnotations.push( annotation );
|
2013-06-12 18:39:24 +00:00
|
|
|
|
|
|
|
childDataElements = this.getDataFromDomRecursion( childDomElement, undefined, childAnnotations );
|
2013-07-15 11:06:31 +00:00
|
|
|
if ( !childDataElements.length || isAllInstanceOf( childDataElements, ve.dm.AlienMetaItem ) ) {
|
2013-06-12 18:39:24 +00:00
|
|
|
// Empty annotation, create a meta item
|
|
|
|
childDataElements = this.createDataElements( ve.dm.AlienMetaItem, childDomElements );
|
|
|
|
childDataElements.push( { 'type': '/' + childDataElements[0].type } );
|
2013-06-30 06:51:34 +00:00
|
|
|
// Annotate meta item
|
|
|
|
if ( !context.annotations.isEmpty() ) {
|
|
|
|
childDataElements[0].annotations = context.annotations.getIndexes().slice();
|
|
|
|
}
|
2013-06-12 18:39:24 +00:00
|
|
|
}
|
2013-06-30 05:35:21 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
2013-06-12 18:39:24 +00:00
|
|
|
data = data.concat( childDataElements );
|
2013-04-18 00:08:47 +00:00
|
|
|
// Clear wrapped whitespace
|
|
|
|
wrappedWhitespace = '';
|
2013-02-16 02:37:50 +00:00
|
|
|
} else {
|
2013-02-21 23:01:04 +00:00
|
|
|
// Node or meta item
|
|
|
|
if ( modelClass.prototype instanceof ve.dm.MetaItem ) {
|
2013-05-18 03:49:25 +00:00
|
|
|
htmlAttributes = ve.dm.Converter.buildHtmlAttributeList(
|
|
|
|
childDomElements, modelClass.static.storeHtmlAttributes, true
|
|
|
|
);
|
|
|
|
if ( htmlAttributes ) {
|
|
|
|
childDataElements[0].htmlAttributes = htmlAttributes;
|
|
|
|
}
|
2013-02-21 23:01:04 +00:00
|
|
|
// No additional processing needed
|
|
|
|
// Write to data and continue
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
if ( childDataElements.length === 1 ) {
|
|
|
|
childDataElements.push( { 'type': '/' + childDataElements[0].type } );
|
|
|
|
}
|
2013-07-12 14:03:52 +00:00
|
|
|
// Annotate meta item
|
|
|
|
if ( !context.annotations.isEmpty() ) {
|
|
|
|
childDataElements[0].annotations = context.annotations.getIndexes().slice();
|
|
|
|
}
|
|
|
|
// Queue wrapped meta items only if it's actually possible for us to move them out
|
|
|
|
// of the wrapper
|
|
|
|
if ( context.inWrapper && context.canCloseWrapper ) {
|
2013-04-18 23:06:58 +00:00
|
|
|
wrappedMetaItems = wrappedMetaItems.concat( childDataElements );
|
|
|
|
if ( wrappedWhitespace !== '' ) {
|
|
|
|
data.splice( wrappedWhitespaceIndex, wrappedWhitespace.length );
|
|
|
|
addWhitespace( childDataElements[0], 0, wrappedWhitespace );
|
|
|
|
nextWhitespace = wrappedWhitespace;
|
|
|
|
wrappedWhitespace = '';
|
|
|
|
}
|
|
|
|
} else {
|
2013-06-30 05:35:21 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
2013-04-18 23:06:58 +00:00
|
|
|
data = data.concat( childDataElements );
|
|
|
|
processNextWhitespace( childDataElements[0] );
|
|
|
|
prevElement = childDataElements[0];
|
|
|
|
}
|
2013-07-23 13:35:26 +00:00
|
|
|
// In case we consumed multiple childDomElements, adjust i accordingly
|
|
|
|
i += childDomElements.length - 1;
|
2013-02-21 23:01:04 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
childIsContent = this.nodeFactory.isNodeContent( childDataElements[0].type );
|
2013-02-16 02:37:50 +00:00
|
|
|
|
|
|
|
// If childIsContent isn't what we expect, adjust
|
|
|
|
if ( !context.expectingContent && childIsContent ) {
|
|
|
|
startWrapping();
|
|
|
|
prevElement = wrappingParagraph;
|
|
|
|
} else if ( context.expectingContent && !childIsContent ) {
|
|
|
|
if ( context.inWrapper && context.canCloseWrapper ) {
|
2012-12-05 22:35:10 +00:00
|
|
|
stopWrapping();
|
|
|
|
} else {
|
2013-02-16 02:37:50 +00:00
|
|
|
// Alienate
|
|
|
|
modelClass = ve.dm.AlienNode;
|
|
|
|
childDomElements = modelClass.static.enableAboutGrouping ?
|
|
|
|
aboutGroup : [ childDomElement ];
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
childDataElements = this.createDataElements( modelClass, childDomElements );
|
|
|
|
childIsContent = this.nodeFactory.isNodeContent( childDataElements[0].type );
|
2012-12-05 22:35:10 +00:00
|
|
|
}
|
2012-11-20 23:37:06 +00:00
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
|
2013-05-09 19:22:12 +00:00
|
|
|
// If we're inserting content into a wrapper, any wrappedWhitespace
|
|
|
|
// up to this point can be considered dealt with
|
|
|
|
if ( context.inWrapper && childIsContent ) {
|
|
|
|
wrappedWhitespace = '';
|
|
|
|
}
|
|
|
|
|
2013-02-16 02:37:50 +00:00
|
|
|
// Annotate child
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( childIsContent && !context.annotations.isEmpty() ) {
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
childDataElements[0].annotations = context.annotations.getIndexes().slice();
|
2013-02-16 02:37:50 +00:00
|
|
|
}
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
// Output child and process children if needed
|
2013-02-16 02:37:50 +00:00
|
|
|
if (
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
childDataElements.length === 1 &&
|
2013-02-16 02:37:50 +00:00
|
|
|
childDomElements.length === 1 &&
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
this.nodeFactory.canNodeHaveChildren( childDataElements[0].type ) &&
|
|
|
|
!this.nodeFactory.doesNodeHandleOwnChildren( childDataElements[0].type )
|
2013-02-16 02:37:50 +00:00
|
|
|
) {
|
2013-05-18 03:49:25 +00:00
|
|
|
htmlAttributes = ve.dm.Converter.buildHtmlAttributeList(
|
|
|
|
childDomElements, modelClass.static.storeHtmlAttributes
|
|
|
|
);
|
|
|
|
if ( htmlAttributes ) {
|
|
|
|
childDataElements[0].htmlAttributes = htmlAttributes;
|
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
// Recursion
|
|
|
|
// Opening and closing elements are added by the recursion too
|
2013-06-30 05:35:21 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
2013-02-16 02:37:50 +00:00
|
|
|
data = data.concat(
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
this.getDataFromDomRecursion( childDomElement, childDataElements[0],
|
2013-04-10 20:32:33 +00:00
|
|
|
new ve.dm.AnnotationSet( this.store )
|
2013-02-16 02:37:50 +00:00
|
|
|
)
|
|
|
|
);
|
|
|
|
} else {
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
if ( childDataElements.length === 1 ) {
|
|
|
|
childDataElements.push( { 'type': '/' + childDataElements[0].type } );
|
|
|
|
}
|
2013-05-18 03:49:25 +00:00
|
|
|
htmlAttributes = ve.dm.Converter.buildHtmlAttributeList(
|
|
|
|
childDomElements, modelClass.static.storeHtmlAttributes, true
|
|
|
|
);
|
|
|
|
if ( htmlAttributes ) {
|
|
|
|
childDataElements[0].htmlAttributes = htmlAttributes;
|
|
|
|
}
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
// Write childDataElements directly
|
2013-06-30 05:35:21 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
data = data.concat( childDataElements );
|
2013-02-16 02:37:50 +00:00
|
|
|
}
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
processNextWhitespace( childDataElements[0] );
|
|
|
|
prevElement = childDataElements[0];
|
2013-02-16 02:37:50 +00:00
|
|
|
|
|
|
|
// In case we consumed multiple childDomElements, adjust i accordingly
|
|
|
|
i += childDomElements.length - 1;
|
2012-11-20 04:26:54 +00:00
|
|
|
}
|
2012-06-07 00:47:27 +00:00
|
|
|
break;
|
|
|
|
case Node.TEXT_NODE:
|
2012-08-02 18:46:13 +00:00
|
|
|
text = childDomElement.data;
|
2012-08-03 19:12:09 +00:00
|
|
|
if ( text === '' ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Empty text node?!?
|
2012-08-03 19:12:09 +00:00
|
|
|
break;
|
2012-06-15 05:46:04 +00:00
|
|
|
}
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( !context.originallyExpectingContent ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Strip and store outer whitespace
|
|
|
|
if ( text.match( /^\s+$/ ) ) {
|
|
|
|
// This text node is whitespace only
|
2013-02-16 02:16:21 +00:00
|
|
|
if ( context.inWrapper ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// We're already wrapping, so output this whitespace
|
|
|
|
// and store it in wrappedWhitespace (see
|
|
|
|
// comment about wrappedWhitespace below)
|
|
|
|
wrappedWhitespace = text;
|
2012-12-11 18:23:33 +00:00
|
|
|
wrappedWhitespaceIndex = data.length;
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
data = data.concat(
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText( wrappedWhitespace, context.annotations )
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
);
|
|
|
|
} else {
|
|
|
|
// We're not in wrapping mode, store this whitespace
|
|
|
|
if ( !prevElement ) {
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( wrapperElement ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// First child, store as inner
|
|
|
|
// whitespace in the parent
|
2013-04-10 20:32:33 +00:00
|
|
|
addWhitespace( wrapperElement, 1, text );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
}
|
Remainder JSHint fixes on modules/ve/*
[jshint]
ce/ve.ce.Surface.js: line 670, col 9, Too many var statements.
ce/ve.ce.Surface.js: line 695, col 6, Missing semicolon.
ce/ve.ce.Surface.js: line 726, col 22, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 726, col 41, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 733, col 13, Too many var statements.
ce/ve.ce.Surface.js: line 734, col 24, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 1013, col 13, Too many var statements.
ce/ve.ce.Surface.js: line 1019, col 17, Too many var statements.
ce/ve.ce.Surface.js: line 1023, col 18, Too many ar statements.
ce/ve.ce.Surface.js: line 1027, col 13, Too many var statements.
dm/annotations/ve.dm.LinkAnnotation.js: line 70, col 52, Insecure '.'.
dm/ve.dm.Converter.js: line 383, col 29, Empty block.
dm/ve.dm.Converter.js: line 423, col 33, Empty block.
Commands:
* jshint .
* ack '(if|else|function|switch|for|while)\('
* Sublime Text 2:
Find(*): (if|else|function|switch|for|while)\(
Replace: $1 (
* ack ' ' -Q # double spaces, except in certain comments
Change-Id: I8e34bf2924bc8688fdf8acef08bbc4f6707e93be
2012-09-02 21:45:01 +00:00
|
|
|
// Else, WTF?!? This is not supposed to
|
|
|
|
// happen, but it's not worth
|
|
|
|
// throwing an exception over.
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
} else {
|
|
|
|
addWhitespace( prevElement, 3, text );
|
|
|
|
}
|
|
|
|
nextWhitespace = text;
|
|
|
|
wrappedWhitespace = '';
|
2013-04-18 23:06:58 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
}
|
|
|
|
// We're done, no actual text left to process
|
|
|
|
break;
|
|
|
|
} else {
|
|
|
|
// This text node contains actual text
|
|
|
|
// Separate the real text from the whitespace
|
|
|
|
// HACK: . doesn't match newlines in JS, so use
|
|
|
|
// [\s\S] to match any character
|
|
|
|
matches = text.match( /^(\s*)([\s\S]*?)(\s*)$/ );
|
2013-02-16 02:16:21 +00:00
|
|
|
if ( !context.inWrapper ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Wrap the text in a paragraph and output it
|
(bug 43056) Inline tags like <span> are block-alienated sometimes
This happens when the <span> is the start of unwrapped content. The
converter logic to look at the tag name in wrapping mode doesn't kick in
because we're not yet in wrapping mode at that point.
The core issue was that previously, we relied on the document
structure/state to choose between alienBlock and alienInline, and only
used the tag name where the document structure was ambiguous (wrapping).
Changed this to be the other way around: we now rely primarily on the
tag name, and if that doesn't match what we expect based on the document
structure, we work around that if possible. Specifically:
* inline tag in our wrapper --> inline alien
* block tag in our wrapper --> close wrapper, block alien
* inline tag in wrapper that's not ours --> inline alien
* block tag in wrapper that's not ours --> *inline* alien
* inline tag in structural location --> open wrapper, inline alien
* block tag in structural location --> block alien
* inline tag in content location --> inline alien
* block tag in content location --> *inline* alien
only in the fourth and the last case do we need to use the "wrong" alien type to
preserve document validity, and it will always be inline where block was
expected, which should reduce UI issues.
The condensed version of the above, which is used in the code, is:
* If in a non-wrapper content location, use inline
* If in a wrapper that's not ours, use inline
* Otherwise, decide based on tag name
* Open or close wrapper if needed
ve.dm.Converter:
* Replace isInline logic in createAlien() with the above
* Factor out code to start wrapping (was duplicated) into startWrapping()
* Call startWrapping() if createAlien() returns an alienInline and we're
in a structural location
Tests:
* Add test cases with aliens at the start and end of unwrapped content
** The first one failed prior to these changes and now passes, the
second one was already passing
* Fix about group test case, was exhibiting the bug that this commit fixes
Change-Id: I657aa0ff5bc2b57cd48ef8a99c8ca930936c03b8
2012-12-20 00:59:58 +00:00
|
|
|
startWrapping();
|
2012-06-15 05:46:04 +00:00
|
|
|
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Only store leading whitespace if we just
|
|
|
|
// started wrapping
|
|
|
|
if ( matches[1] !== '' ) {
|
|
|
|
if ( !prevElement ) {
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( wrapperElement ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// First child, store as inner
|
|
|
|
// whitespace in the parent
|
2013-04-10 20:32:33 +00:00
|
|
|
addWhitespace( wrapperElement, 1, matches[1] );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
}
|
Remainder JSHint fixes on modules/ve/*
[jshint]
ce/ve.ce.Surface.js: line 670, col 9, Too many var statements.
ce/ve.ce.Surface.js: line 695, col 6, Missing semicolon.
ce/ve.ce.Surface.js: line 726, col 22, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 726, col 41, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 733, col 13, Too many var statements.
ce/ve.ce.Surface.js: line 734, col 24, Expected '===' and instead saw '=='.
ce/ve.ce.Surface.js: line 1013, col 13, Too many var statements.
ce/ve.ce.Surface.js: line 1019, col 17, Too many var statements.
ce/ve.ce.Surface.js: line 1023, col 18, Too many ar statements.
ce/ve.ce.Surface.js: line 1027, col 13, Too many var statements.
dm/annotations/ve.dm.LinkAnnotation.js: line 70, col 52, Insecure '.'.
dm/ve.dm.Converter.js: line 383, col 29, Empty block.
dm/ve.dm.Converter.js: line 423, col 33, Empty block.
Commands:
* jshint .
* ack '(if|else|function|switch|for|while)\('
* Sublime Text 2:
Find(*): (if|else|function|switch|for|while)\(
Replace: $1 (
* ack ' ' -Q # double spaces, except in certain comments
Change-Id: I8e34bf2924bc8688fdf8acef08bbc4f6707e93be
2012-09-02 21:45:01 +00:00
|
|
|
// Else, WTF?!? This is not supposed to
|
|
|
|
// happen, but it's not worth
|
|
|
|
// throwing an exception over.
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
} else {
|
|
|
|
addWhitespace( prevElement, 3, matches[1] );
|
|
|
|
}
|
|
|
|
addWhitespace( wrappingParagraph, 0, matches[1] );
|
|
|
|
}
|
|
|
|
} else {
|
2013-04-18 23:06:58 +00:00
|
|
|
outputWrappedMetaItems( 'restore' );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// We were already wrapping in a paragraph,
|
|
|
|
// so the leading whitespace must be output
|
|
|
|
data = data.concat(
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText( matches[1], context.annotations )
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
);
|
|
|
|
}
|
|
|
|
// Output the text sans whitespace
|
|
|
|
data = data.concat(
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText( matches[2], context.annotations )
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
);
|
|
|
|
|
|
|
|
// Don't store this in wrappingParagraph.internal.whitespace[3]
|
|
|
|
// and nextWhitespace just yet. Instead, store it
|
|
|
|
// in wrappedWhitespace. There might be more text
|
|
|
|
// nodes after this one, so we output wrappedWhitespace
|
|
|
|
// for now and undo that if it turns out this was
|
|
|
|
// the last text node. We can't output it later
|
|
|
|
// because we have to apply the correct annotations.
|
|
|
|
wrappedWhitespace = matches[3];
|
2012-12-11 18:23:33 +00:00
|
|
|
wrappedWhitespaceIndex = data.length;
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
data = data.concat(
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText( wrappedWhitespace, context.annotations )
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
);
|
|
|
|
prevElement = wrappingParagraph;
|
|
|
|
break;
|
|
|
|
}
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
}
|
|
|
|
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Strip leading and trailing inner whitespace
|
2012-08-16 17:53:33 +00:00
|
|
|
// (but only in non-annotation nodes)
|
|
|
|
// and store it so it can be restored later.
|
2012-11-07 20:03:58 +00:00
|
|
|
if (
|
2013-04-10 20:32:33 +00:00
|
|
|
context.annotations.isEmpty() && i === 0 && wrapperElement &&
|
|
|
|
!this.nodeFactory.doesNodeHaveSignificantWhitespace( wrapperElement.type )
|
2012-11-07 20:03:58 +00:00
|
|
|
) {
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
// Strip leading whitespace from the first child
|
|
|
|
matches = text.match( /^\s+/ );
|
|
|
|
if ( matches && matches[0] !== '' ) {
|
2013-04-10 20:32:33 +00:00
|
|
|
addWhitespace( wrapperElement, 1, matches[0] );
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
text = text.substring( matches[0].length );
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (
|
2013-04-10 20:32:33 +00:00
|
|
|
context.annotations.isEmpty() &&
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
i === domElement.childNodes.length - 1 &&
|
2013-04-10 20:32:33 +00:00
|
|
|
wrapperElement &&
|
|
|
|
!this.nodeFactory.doesNodeHaveSignificantWhitespace( wrapperElement.type )
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
) {
|
|
|
|
// Strip trailing whitespace from the last child
|
|
|
|
matches = text.match( /\s+$/ );
|
|
|
|
if ( matches && matches[0] !== '' ) {
|
2013-04-10 20:32:33 +00:00
|
|
|
addWhitespace( wrapperElement, 2, matches[0] );
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
text = text.substring( 0,
|
|
|
|
text.length - matches[0].length );
|
|
|
|
}
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
2012-06-12 00:32:28 +00:00
|
|
|
|
2012-06-07 00:47:27 +00:00
|
|
|
// Annotate the text and output it
|
|
|
|
data = data.concat(
|
2013-04-10 20:32:33 +00:00
|
|
|
ve.dm.Converter.getDataContentFromText( text, context.annotations )
|
2012-06-07 00:47:27 +00:00
|
|
|
);
|
|
|
|
break;
|
|
|
|
case Node.COMMENT_NODE:
|
2013-04-18 23:06:58 +00:00
|
|
|
// TODO treat this as a node with nodeName #comment, removes code duplication
|
2013-06-05 17:22:01 +00:00
|
|
|
childDataElements = this.createDataElements( ve.dm.AlienMetaItem, [ childDomElement ] );
|
|
|
|
childDataElements.push( { 'type': '/' + childDataElements[0].type } );
|
2013-06-23 17:48:32 +00:00
|
|
|
|
|
|
|
// Annotate
|
|
|
|
if ( !context.annotations.isEmpty() ) {
|
|
|
|
childDataElements[0].annotations = context.annotations.getIndexes().slice();
|
|
|
|
}
|
|
|
|
|
2013-06-25 19:29:58 +00:00
|
|
|
// Queue wrapped meta items only if it's actually possible for us to move them out
|
|
|
|
// of the wrapper
|
|
|
|
if ( context.inWrapper && context.canCloseWrapper ) {
|
2013-04-18 23:06:58 +00:00
|
|
|
wrappedMetaItems = wrappedMetaItems.concat( childDataElements );
|
|
|
|
if ( wrappedWhitespace !== '' ) {
|
|
|
|
data.splice( wrappedWhitespaceIndex, wrappedWhitespace.length );
|
|
|
|
addWhitespace( childDataElements[0], 0, wrappedWhitespace );
|
|
|
|
nextWhitespace = wrappedWhitespace;
|
|
|
|
wrappedWhitespace = '';
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
data = data.concat( childDataElements );
|
|
|
|
processNextWhitespace( childDataElements[0] );
|
|
|
|
prevElement = childDataElements[0];
|
|
|
|
}
|
2012-06-07 00:47:27 +00:00
|
|
|
break;
|
2012-06-06 17:17:30 +00:00
|
|
|
}
|
|
|
|
}
|
2012-06-07 00:47:27 +00:00
|
|
|
// End auto-wrapping of bare content
|
2013-02-16 02:16:21 +00:00
|
|
|
if ( context.inWrapper && context.canCloseWrapper ) {
|
2012-11-20 04:26:54 +00:00
|
|
|
stopWrapping();
|
2013-02-16 02:16:21 +00:00
|
|
|
// HACK: don't set context.inWrapper = false here because it's checked below
|
|
|
|
context.inWrapper = true;
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
2012-06-19 21:36:35 +00:00
|
|
|
|
|
|
|
// If we're closing a node that doesn't have any children, but could contain a paragraph,
|
|
|
|
// add a paragraph. This prevents things like empty list items
|
2013-04-10 20:32:33 +00:00
|
|
|
childTypes = this.nodeFactory.getChildNodeTypes( context.branchType );
|
|
|
|
if ( context.branchType !== 'paragraph' && wrapperElement && data[data.length - 1] === wrapperElement &&
|
|
|
|
!context.inWrapper && !this.nodeFactory.canNodeContainContent( context.branchType ) &&
|
|
|
|
!this.nodeFactory.isNodeContent( context.branchType ) &&
|
2012-08-11 08:14:56 +00:00
|
|
|
( childTypes === null || ve.indexOf( 'paragraph', childTypes ) !== -1 )
|
2012-06-19 21:36:35 +00:00
|
|
|
) {
|
2012-11-17 03:37:38 +00:00
|
|
|
data.push( { 'type': 'paragraph', 'internal': { 'generated': 'empty' } } );
|
2012-06-19 21:36:35 +00:00
|
|
|
data.push( { 'type': '/paragraph' } );
|
|
|
|
}
|
|
|
|
|
2012-06-07 00:47:27 +00:00
|
|
|
// Close element
|
2013-04-10 20:32:33 +00:00
|
|
|
if ( wrapperElement ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Add the whitespace after the last child to the parent as innerPost
|
2013-06-30 07:29:38 +00:00
|
|
|
// But don't do this if the parent is empty, because in that case we've already put that
|
|
|
|
// whitespace in innerPre
|
|
|
|
if ( nextWhitespace !== '' && data[data.length - 1] !== wrapperElement ) {
|
2013-04-10 20:32:33 +00:00
|
|
|
addWhitespace( wrapperElement, 2, nextWhitespace );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
nextWhitespace = '';
|
|
|
|
}
|
2013-06-30 07:29:38 +00:00
|
|
|
data.push( { 'type': '/' + wrapperElement.type } );
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
2012-06-19 00:38:42 +00:00
|
|
|
// Don't return an empty document
|
2013-07-15 11:06:31 +00:00
|
|
|
if ( context.branchType === 'document' && isAllInstanceOf( data, ve.dm.MetaItem ) && !annotationSet ) {
|
|
|
|
return data.concat( [
|
2012-11-17 03:37:38 +00:00
|
|
|
{ 'type': 'paragraph', 'internal': { 'generated': 'empty' } },
|
2012-08-16 20:06:18 +00:00
|
|
|
{ 'type': '/paragraph' }
|
2013-07-15 11:06:31 +00:00
|
|
|
] );
|
2012-06-19 00:38:42 +00:00
|
|
|
}
|
2013-04-10 20:32:33 +00:00
|
|
|
|
|
|
|
this.contextStack.pop();
|
2012-06-07 00:47:27 +00:00
|
|
|
return data;
|
2012-06-06 17:17:30 +00:00
|
|
|
};
|
|
|
|
|
2013-07-23 13:35:26 +00:00
|
|
|
/**
|
|
|
|
* Check if all the domElements provided are metadata or whitespace.
|
|
|
|
*
|
|
|
|
* A list of model names to exclude when matching can optionally be passed.
|
|
|
|
*
|
|
|
|
* @param {HTMLElement[]} domElements DOM elements to check
|
|
|
|
* @param {string[]} [excludeTypes] Model names to exclude when matching DOM elements
|
|
|
|
* @returns {boolean} All the elements are metadata or whitespace
|
|
|
|
*/
|
|
|
|
ve.dm.Converter.prototype.isDomAllMetaOrWhitespace = function ( domElements, excludeTypes ) {
|
|
|
|
var i, childDomElement, modelName, modelClass;
|
|
|
|
|
|
|
|
for ( i = 0; i < domElements.length; i++ ) {
|
|
|
|
childDomElement = domElements[i];
|
|
|
|
switch ( childDomElement.nodeType ) {
|
|
|
|
case Node.ELEMENT_NODE:
|
|
|
|
modelName = this.modelRegistry.matchElement( childDomElement, false, excludeTypes );
|
|
|
|
modelClass = this.modelRegistry.lookup( modelName ) || ve.dm.AlienNode;
|
|
|
|
if (
|
|
|
|
!( modelClass.prototype instanceof ve.dm.Annotation ) &&
|
|
|
|
!( modelClass.prototype instanceof ve.dm.MetaItem )
|
|
|
|
) {
|
|
|
|
// If the element not meta or an annotation, then we must have content
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
// Recursively check children
|
|
|
|
if (
|
|
|
|
childDomElement.childNodes.length &&
|
|
|
|
!this.isDomAllMetaOrWhitespace( childDomElement.childNodes, excludeTypes )
|
|
|
|
) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
continue;
|
|
|
|
case Node.TEXT_NODE:
|
|
|
|
// Check for whitespace-only
|
|
|
|
if ( !childDomElement.data.match( /\S/ ) ) {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case Node.COMMENT_NODE:
|
|
|
|
// Comments are always meta
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
};
|
|
|
|
|
2012-06-06 17:17:30 +00:00
|
|
|
/**
|
2012-06-08 04:58:56 +00:00
|
|
|
* Convert linear model data to an HTML DOM
|
2012-06-06 17:17:30 +00:00
|
|
|
*
|
|
|
|
* @method
|
2013-04-17 17:53:26 +00:00
|
|
|
* @param {Array} documentData Linear model data
|
2013-03-20 22:35:05 +00:00
|
|
|
* @param {ve.dm.IndexValueStore} store Index-value store
|
2013-04-17 17:53:26 +00:00
|
|
|
* @param {ve.dm.InternalList} internalList Internal list
|
2013-02-11 19:46:58 +00:00
|
|
|
* @returns {HTMLDocument} Document containing the resulting HTML
|
2012-06-06 17:17:30 +00:00
|
|
|
*/
|
2013-04-17 17:53:26 +00:00
|
|
|
ve.dm.Converter.prototype.getDomFromData = function ( documentData, store, internalList ) {
|
2013-05-28 11:49:35 +00:00
|
|
|
var doc = ve.createDocumentFromHtml( '' );
|
2013-05-08 04:25:55 +00:00
|
|
|
|
2013-04-17 17:53:26 +00:00
|
|
|
// Set up the converter state
|
|
|
|
this.documentData = documentData;
|
2013-04-20 16:08:23 +00:00
|
|
|
this.store = store;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.internalList = internalList;
|
|
|
|
|
|
|
|
this.getDomSubtreeFromData( documentData, doc.body );
|
|
|
|
|
|
|
|
// Clear the state
|
|
|
|
this.documentData = null;
|
2013-04-20 16:08:23 +00:00
|
|
|
this.store = null;
|
2013-04-17 17:53:26 +00:00
|
|
|
this.internalList = null;
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
return doc;
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Convert linear model data to an HTML DOM subtree and add it to a container element.
|
|
|
|
*
|
|
|
|
* @param {Array} data Linear model data
|
|
|
|
* @param {HTMLElement} container DOM element to add the generated elements to. Should be empty.
|
|
|
|
* @throws Unbalanced data: looking for closing /type
|
|
|
|
*/
|
2013-04-20 16:08:23 +00:00
|
|
|
ve.dm.Converter.prototype.getDomSubtreeFromData = function ( data, container ) {
|
2013-08-02 11:03:27 +00:00
|
|
|
var text, i, j, isStart, annotations, dataElement, dataElementOrSlice,
|
2013-05-08 04:25:55 +00:00
|
|
|
childDomElements, pre, ours, theirs, parentDomElement, lastChild, isContentNode, sibling,
|
2013-07-16 19:32:28 +00:00
|
|
|
previousSiblings, doUnwrap, textNode, type, annotatedDomElementStack, annotatedDomElements,
|
2013-06-10 23:05:42 +00:00
|
|
|
dataLen = data.length,
|
2013-05-01 23:00:33 +00:00
|
|
|
canContainContentStack = [],
|
2013-04-02 19:11:41 +00:00
|
|
|
conv = this,
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
doc = container.ownerDocument,
|
2012-08-02 18:46:13 +00:00
|
|
|
domElement = container,
|
2013-04-20 16:08:23 +00:00
|
|
|
annotationStack = new ve.dm.AnnotationSet( this.store );
|
2012-11-24 02:44:54 +00:00
|
|
|
|
2013-07-16 19:32:28 +00:00
|
|
|
// TODO this whole function should be rewritten with a domElementStack and ascend() and
|
|
|
|
// descend() functions, to build the whole DOM bottom-up rather than top-down. That would make
|
|
|
|
// unwrapping easier and will hopefully result in fewer DOM operations.
|
|
|
|
|
|
|
|
function openAnnotation() {
|
2013-04-02 19:11:41 +00:00
|
|
|
// Add text if needed
|
|
|
|
if ( text.length > 0 ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements.push( doc.createTextNode( text ) );
|
2013-04-02 19:11:41 +00:00
|
|
|
text = '';
|
|
|
|
}
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements = [];
|
|
|
|
annotatedDomElementStack.push( annotatedDomElements );
|
2013-04-02 19:11:41 +00:00
|
|
|
}
|
|
|
|
|
2013-07-16 19:32:28 +00:00
|
|
|
function closeAnnotation( annotation ) {
|
|
|
|
var i, len, annotationElement, annotatedChildDomElements;
|
|
|
|
|
2013-04-02 19:11:41 +00:00
|
|
|
// Add text if needed
|
|
|
|
if ( text.length > 0 ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements.push( doc.createTextNode( text ) );
|
2013-04-02 19:11:41 +00:00
|
|
|
text = '';
|
|
|
|
}
|
2013-07-16 19:32:28 +00:00
|
|
|
|
|
|
|
annotatedChildDomElements = annotatedDomElementStack.pop();
|
|
|
|
annotatedDomElements = annotatedDomElementStack[annotatedDomElementStack.length - 1];
|
2013-07-16 21:40:05 +00:00
|
|
|
annotationElement = conv.getDomElementsFromDataElement(
|
|
|
|
annotation.getElement(), doc, annotatedChildDomElements
|
|
|
|
)[0];
|
2013-07-16 21:53:06 +00:00
|
|
|
if ( annotationElement ) {
|
|
|
|
for ( i = 0, len = annotatedChildDomElements.length; i < len; i++ ) {
|
|
|
|
annotationElement.appendChild( annotatedChildDomElements[i] );
|
|
|
|
}
|
|
|
|
annotatedDomElements.push( annotationElement );
|
|
|
|
} else {
|
|
|
|
for ( i = 0, len = annotatedChildDomElements.length; i < len; i++ ) {
|
|
|
|
annotatedDomElements.push( annotatedChildDomElements[i] );
|
|
|
|
}
|
2013-07-16 19:32:28 +00:00
|
|
|
}
|
2013-04-02 19:11:41 +00:00
|
|
|
}
|
|
|
|
|
2013-06-10 23:05:42 +00:00
|
|
|
function findEndOfNode( i ) {
|
|
|
|
var j = i + 1, depth = 1;
|
|
|
|
while ( j < dataLen && depth > 0 ) {
|
|
|
|
if ( data[j].type ) {
|
|
|
|
depth += data[j].type.charAt( 0 ) === '/' ? -1 : 1;
|
|
|
|
}
|
|
|
|
j++;
|
|
|
|
}
|
|
|
|
if ( depth !== 0 ) {
|
|
|
|
throw new Error( 'Unbalanced data: looking for closing /' +
|
|
|
|
dataElement.type );
|
|
|
|
}
|
|
|
|
return j;
|
|
|
|
}
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
function getDataElementOrSlice() {
|
2013-06-10 23:05:42 +00:00
|
|
|
var dataSlice;
|
2013-05-22 19:51:51 +00:00
|
|
|
if (
|
|
|
|
ve.dm.nodeFactory.lookup( data[i].type ) &&
|
|
|
|
ve.dm.nodeFactory.doesNodeHandleOwnChildren( data[i].type )
|
|
|
|
) {
|
2013-06-10 23:05:42 +00:00
|
|
|
dataSlice = data.slice( i, findEndOfNode( i ) );
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
} else {
|
|
|
|
dataSlice = data[i];
|
|
|
|
}
|
|
|
|
return dataSlice;
|
|
|
|
}
|
|
|
|
|
2013-06-10 23:05:42 +00:00
|
|
|
function removeInternalNodes() {
|
|
|
|
var dataCopy, endOffset;
|
|
|
|
// See if there is an internalList in the data, and if there is one, remove it
|
|
|
|
// Removing it here prevents unwanted interactions with whitespace preservation
|
|
|
|
for ( i = 0; i < dataLen; i++ ) {
|
|
|
|
if (
|
|
|
|
data[i].type && data[i].type.charAt( 0 ) !== '/' &&
|
|
|
|
ve.dm.nodeFactory.lookup( data[i].type ) &&
|
|
|
|
ve.dm.nodeFactory.isNodeInternal( data[i].type )
|
|
|
|
) {
|
|
|
|
// Copy data if we haven't already done so
|
|
|
|
if ( !dataCopy ) {
|
|
|
|
dataCopy = data.slice();
|
|
|
|
}
|
|
|
|
endOffset = findEndOfNode( i );
|
|
|
|
// Remove this node's data from dataCopy
|
|
|
|
dataCopy.splice( i - ( dataLen - dataCopy.length ), endOffset - i );
|
|
|
|
// Move i such that it will be at endOffset in the next iteration
|
|
|
|
i = endOffset - 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if ( dataCopy ) {
|
|
|
|
data = dataCopy;
|
|
|
|
dataLen = data.length;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
removeInternalNodes();
|
|
|
|
|
|
|
|
for ( i = 0; i < dataLen; i++ ) {
|
2012-06-08 04:58:56 +00:00
|
|
|
if ( typeof data[i] === 'string' ) {
|
2012-06-07 22:02:25 +00:00
|
|
|
// Text
|
2012-06-08 04:58:56 +00:00
|
|
|
text = '';
|
2013-08-02 11:03:27 +00:00
|
|
|
isStart = i > 0 &&
|
|
|
|
ve.dm.LinearData.static.isOpenElementData( data[i-1] ) &&
|
2013-09-16 15:41:45 +00:00
|
|
|
!ve.dm.nodeFactory.doesNodeHaveSignificantWhitespace(
|
|
|
|
ve.dm.LinearData.static.getType( data[i-1] )
|
|
|
|
);
|
2012-06-08 04:58:56 +00:00
|
|
|
// Continue forward as far as the plain text goes
|
|
|
|
while ( typeof data[i] === 'string' ) {
|
2013-09-16 15:41:45 +00:00
|
|
|
// HACK: Skip over leading whitespace (bug 51462) in non-whitespace-preserving tags
|
2013-08-02 11:03:27 +00:00
|
|
|
if ( !( isStart && data[i].match( /\s/ ) ) ) {
|
|
|
|
text += data[i];
|
|
|
|
isStart = false;
|
|
|
|
}
|
2012-06-08 04:58:56 +00:00
|
|
|
i++;
|
|
|
|
}
|
|
|
|
// i points to the first non-text thing, go back one so we don't skip this later
|
|
|
|
i--;
|
|
|
|
// Add text
|
2013-02-11 19:46:58 +00:00
|
|
|
domElement.appendChild( doc.createTextNode( text ) );
|
2012-06-08 04:58:56 +00:00
|
|
|
} else if (
|
|
|
|
ve.isArray( data[i] ) ||
|
|
|
|
(
|
2013-02-21 23:01:04 +00:00
|
|
|
data[i].annotations !== undefined && (
|
|
|
|
this.metaItemFactory.lookup( data[i].type ) ||
|
|
|
|
this.nodeFactory.isNodeContent( data[i].type )
|
|
|
|
)
|
2012-06-08 04:58:56 +00:00
|
|
|
)
|
|
|
|
) {
|
2013-06-23 17:48:32 +00:00
|
|
|
// Annotated text, nodes or meta
|
2012-06-08 04:58:56 +00:00
|
|
|
text = '';
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements = [];
|
|
|
|
annotatedDomElementStack = [ annotatedDomElements ];
|
2012-06-08 04:58:56 +00:00
|
|
|
while (
|
2013-09-20 14:22:06 +00:00
|
|
|
data[i] !== undefined && (
|
|
|
|
ve.isArray( data[i] ) ||
|
|
|
|
(
|
|
|
|
data[i].annotations !== undefined && (
|
|
|
|
this.metaItemFactory.lookup( data[i].type ) ||
|
|
|
|
this.nodeFactory.isNodeContent( data[i].type )
|
|
|
|
)
|
2013-06-23 17:48:32 +00:00
|
|
|
)
|
2012-06-08 04:58:56 +00:00
|
|
|
)
|
|
|
|
) {
|
2013-03-20 22:35:05 +00:00
|
|
|
annotations = new ve.dm.AnnotationSet(
|
2013-04-20 16:08:23 +00:00
|
|
|
this.store, data[i].annotations || data[i][1]
|
2013-03-20 22:35:05 +00:00
|
|
|
);
|
2013-04-02 19:11:41 +00:00
|
|
|
ve.dm.Converter.openAndCloseAnnotations( annotationStack, annotations,
|
|
|
|
openAnnotation, closeAnnotation
|
|
|
|
);
|
2012-08-24 02:06:36 +00:00
|
|
|
|
2012-06-08 04:58:56 +00:00
|
|
|
if ( data[i].annotations === undefined ) {
|
2012-08-24 02:06:36 +00:00
|
|
|
// Annotated text
|
2012-06-08 04:58:56 +00:00
|
|
|
text += data[i][0];
|
|
|
|
} else {
|
2012-08-24 02:06:36 +00:00
|
|
|
// Annotated node
|
2012-06-08 04:58:56 +00:00
|
|
|
// Add text if needed
|
|
|
|
if ( text.length > 0 ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements.push( doc.createTextNode( text ) );
|
2012-06-08 04:58:56 +00:00
|
|
|
text = '';
|
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
// Insert the elements
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
dataElementOrSlice = getDataElementOrSlice();
|
|
|
|
childDomElements = this.getDomElementsFromDataElement( dataElementOrSlice, doc );
|
2013-02-16 02:37:50 +00:00
|
|
|
for ( j = 0; j < childDomElements.length; j++ ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements.push( childDomElements[j] );
|
2013-02-16 02:37:50 +00:00
|
|
|
}
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
if ( ve.isArray( dataElementOrSlice ) ) {
|
|
|
|
i += dataElementOrSlice.length - 1;
|
|
|
|
} else {
|
|
|
|
i++; // Skip the closing
|
|
|
|
}
|
2012-06-08 04:58:56 +00:00
|
|
|
}
|
|
|
|
i++;
|
|
|
|
}
|
|
|
|
// We're now at the first non-annotated thing, go back one so we don't skip this later
|
|
|
|
i--;
|
|
|
|
|
|
|
|
// Add any gathered text
|
|
|
|
if ( text.length > 0 ) {
|
2013-07-16 19:32:28 +00:00
|
|
|
annotatedDomElements.push( doc.createTextNode( text ) );
|
2012-06-08 04:58:56 +00:00
|
|
|
text = '';
|
|
|
|
}
|
2013-07-16 19:32:28 +00:00
|
|
|
// Close any remaining annotations
|
2013-07-26 14:39:47 +00:00
|
|
|
ve.dm.Converter.openAndCloseAnnotations( annotationStack, new ve.dm.AnnotationSet( this.store ),
|
2013-07-16 19:32:28 +00:00
|
|
|
openAnnotation, closeAnnotation
|
|
|
|
);
|
|
|
|
// Put the annotated nodes in the DOM
|
|
|
|
for ( j = 0; j < annotatedDomElements.length; j++ ) {
|
|
|
|
domElement.appendChild( annotatedDomElements[j] );
|
2012-06-08 04:58:56 +00:00
|
|
|
}
|
|
|
|
} else if ( data[i].type !== undefined ) {
|
2012-08-02 18:46:13 +00:00
|
|
|
dataElement = data[i];
|
2012-06-07 22:02:25 +00:00
|
|
|
// Element
|
2012-09-07 21:58:27 +00:00
|
|
|
if ( dataElement.type.charAt( 0 ) === '/' ) {
|
2013-05-01 23:00:33 +00:00
|
|
|
// Close element
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
parentDomElement = domElement.parentNode;
|
2013-05-01 23:00:33 +00:00
|
|
|
type = data[i].type.substr( 1 );
|
|
|
|
if ( this.metaItemFactory.lookup( type ) ) {
|
|
|
|
isContentNode = canContainContentStack[canContainContentStack.length - 1];
|
|
|
|
} else {
|
|
|
|
isContentNode = this.nodeFactory.isNodeContent( type );
|
|
|
|
canContainContentStack.pop();
|
|
|
|
}
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Process whitespace
|
2012-08-16 17:53:33 +00:00
|
|
|
// whitespace = [ outerPre, innerPre, innerPost, outerPost ]
|
2012-09-07 21:58:27 +00:00
|
|
|
if (
|
|
|
|
!isContentNode &&
|
|
|
|
domElement.veInternal &&
|
|
|
|
domElement.veInternal.whitespace
|
|
|
|
) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Process inner whitespace. innerPre is for sure legitimate
|
|
|
|
// whitespace that should be inserted; if it was a duplicate
|
|
|
|
// of our child's outerPre, we would have cleared it.
|
2012-08-16 17:53:33 +00:00
|
|
|
pre = domElement.veInternal.whitespace[1];
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
if ( pre ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
if (
|
|
|
|
domElement.firstChild &&
|
2013-03-04 16:24:09 +00:00
|
|
|
domElement.firstChild.nodeType === Node.TEXT_NODE
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
) {
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
// First child is a TextNode, prepend to it
|
|
|
|
domElement.firstChild.insertData( 0, pre );
|
|
|
|
} else {
|
|
|
|
// Prepend a TextNode
|
2013-03-04 16:24:09 +00:00
|
|
|
textNode = doc.createTextNode( pre );
|
|
|
|
textNode.veIsWhitespace = true;
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
domElement.insertBefore(
|
2013-03-04 16:24:09 +00:00
|
|
|
textNode,
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
domElement.firstChild
|
|
|
|
);
|
|
|
|
}
|
|
|
|
}
|
2013-02-16 02:37:50 +00:00
|
|
|
lastChild = domElement.veInternal.childDomElements ?
|
|
|
|
domElement.veInternal
|
|
|
|
.childDomElements[domElement.veInternal.childDomElements.length - 1]
|
|
|
|
.lastChild :
|
|
|
|
domElement.lastChild;
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
ours = domElement.veInternal.whitespace[2];
|
|
|
|
if ( domElement.lastOuterPost === undefined ) {
|
|
|
|
// This node didn't have any structural children
|
|
|
|
// (i.e. it's a content-containing node), so there's
|
|
|
|
// nothing to check innerPost against
|
|
|
|
theirs = ours;
|
|
|
|
} else {
|
|
|
|
theirs = domElement.lastOuterPost;
|
|
|
|
}
|
|
|
|
if ( ours && ours === theirs ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
if ( lastChild && lastChild.nodeType === Node.TEXT_NODE ) {
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
// Last child is a TextNode, append to it
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
domElement.lastChild.appendData( ours );
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
} else {
|
|
|
|
// Append a TextNode
|
2013-03-04 16:24:09 +00:00
|
|
|
textNode = doc.createTextNode( ours );
|
|
|
|
textNode.veIsWhitespace = true;
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
domElement.appendChild(
|
2013-03-04 16:24:09 +00:00
|
|
|
textNode
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
);
|
|
|
|
}
|
|
|
|
}
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Tell the parent about our outerPost
|
|
|
|
parentDomElement.lastOuterPost = domElement.veInternal.whitespace[3] || '';
|
2012-09-07 21:58:27 +00:00
|
|
|
} else if ( !isContentNode ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Use empty string, because undefined means there were no
|
|
|
|
// structural children
|
|
|
|
parentDomElement.lastOuterPost = '';
|
Strip and preserve inner leading&trailing whitespace in the linear model
This makes things like
== Foo ==
* Bar
render without the leading and trailing spaces, while still
round-tripping those spaces.
* Added a .fringeWhitespace property to the linear model and ve.dm.Node
** Object containing innerPre, innerPost, outerPre, outerPost
** Only inner* are used right now, outer* are planned for future use
** Like .attributes , it's suppressed if it's an empty object
* In getDataFromDom():
** Store the stripped whitespace in .fringeWhitespace
** Move emptiness check up: empty elements with .fringeWhitespace have
to be preserved
** Move paragraph wrapping up: .fringeWhitespace has to be applied to
the generated paragraph, not its parent
** Add wrapperElement to keep track of the element .fringeWhitespace has
to be added to; this is either dataElement or the generated paragraph
or nothing, but we can't modify dataElement because it's used later
* In getDomFromData():
** When processing an opening, store the fringeWhitespace data in the
generated DOM node
** When processing a closing, add the stored whitespace back in
* In the ve.dm.Document constructor, pass through .fringeWhitespace from
the linear model data to the generated nodes
Tests:
* Change one existing test case to account for this change
* Add three new test cases for this behavior
* Add normalizedHtml field so I can test behavior with bare content
Change-Id: I0411544652dd72b923c831c495d69ee4322a2c14
2012-08-10 21:09:04 +00:00
|
|
|
}
|
2012-09-07 21:58:27 +00:00
|
|
|
// else don't touch lastOuterPost
|
2012-08-16 20:06:18 +00:00
|
|
|
|
2013-03-04 16:24:09 +00:00
|
|
|
// Logic to unwrap empty & wrapper nodes.
|
2012-08-16 20:06:18 +00:00
|
|
|
// It would be nicer if we could avoid generating in the first
|
|
|
|
// place, but then remembering where we have to skip ascending
|
|
|
|
// to the parent would be tricky.
|
2013-03-04 16:24:09 +00:00
|
|
|
doUnwrap = false;
|
2013-03-20 22:35:05 +00:00
|
|
|
if ( domElement.veInternal ) {
|
|
|
|
switch ( domElement.veInternal.generated ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
case 'empty':
|
|
|
|
// 'empty' elements - first ensure they are actually empty
|
|
|
|
if ( domElement.childNodes.length === 0 && (
|
|
|
|
// then check that we are the last child
|
|
|
|
// before unwrapping (and therefore destroying)
|
|
|
|
i === data.length - 1 ||
|
2013-06-24 17:51:59 +00:00
|
|
|
data[i + 1].type.charAt( 0 ) === '/'
|
2013-03-04 16:24:09 +00:00
|
|
|
)
|
|
|
|
) {
|
|
|
|
doUnwrap = true;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case 'wrapper':
|
|
|
|
// 'wrapper' elements - ensure there is a block level
|
|
|
|
// element between this element and the previous sibling
|
|
|
|
// wrapper or parent node
|
|
|
|
doUnwrap = true;
|
|
|
|
previousSiblings = domElement.parentElement.childNodes;
|
|
|
|
// Note: previousSiblings includes the current element
|
|
|
|
// so we only go up to length - 2
|
2013-03-20 22:35:05 +00:00
|
|
|
for ( j = previousSiblings.length - 2; j >= 0; j-- ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
sibling = previousSiblings[j];
|
2013-03-20 22:35:05 +00:00
|
|
|
if ( sibling.nodeType === Node.TEXT_NODE && !sibling.veIsWhitespace ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
// we've found an unwrapped paragraph so don't unwrap
|
|
|
|
doUnwrap = false;
|
|
|
|
break;
|
|
|
|
}
|
2013-03-20 22:35:05 +00:00
|
|
|
if ( ve.isBlockElement( sibling ) ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
// there is a block element before the next unwrapped node
|
|
|
|
// so it's safe to unwrap
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2013-03-20 22:35:05 +00:00
|
|
|
if ( doUnwrap ) {
|
2012-08-20 20:13:07 +00:00
|
|
|
while ( domElement.firstChild ) {
|
2012-08-16 20:06:18 +00:00
|
|
|
parentDomElement.insertBefore(
|
2012-08-20 20:13:07 +00:00
|
|
|
domElement.firstChild,
|
2012-08-16 20:06:18 +00:00
|
|
|
domElement
|
|
|
|
);
|
|
|
|
}
|
|
|
|
parentDomElement.removeChild( domElement );
|
|
|
|
}
|
|
|
|
|
2012-08-16 17:53:33 +00:00
|
|
|
delete domElement.veInternal;
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
delete domElement.lastOuterPost;
|
2013-05-22 19:40:13 +00:00
|
|
|
// Ascend to parent node, except if this is an internal node
|
|
|
|
// TODO: It's not covered with unit tests.
|
|
|
|
if ( !ve.dm.nodeFactory.lookup( type ) || !ve.dm.nodeFactory.isNodeInternal( type ) ) {
|
|
|
|
domElement = parentDomElement;
|
|
|
|
}
|
2012-06-08 04:58:56 +00:00
|
|
|
} else {
|
2012-11-24 02:44:54 +00:00
|
|
|
// Create node from data
|
2013-05-01 23:00:33 +00:00
|
|
|
if ( !this.metaItemFactory.lookup( data[i].type ) ) {
|
|
|
|
canContainContentStack.push(
|
|
|
|
// if the last item was true then this item must inherit it
|
|
|
|
canContainContentStack[canContainContentStack.length - 1] ||
|
|
|
|
this.nodeFactory.canNodeContainContent( data[i].type )
|
|
|
|
);
|
|
|
|
}
|
|
|
|
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
dataElementOrSlice = getDataElementOrSlice();
|
|
|
|
childDomElements = this.getDomElementsFromDataElement( dataElementOrSlice, doc );
|
2013-04-17 17:53:26 +00:00
|
|
|
if ( childDomElements ) {
|
|
|
|
// Add clone of internal data; we use a clone rather than a reference because
|
|
|
|
// we modify .veInternal.whitespace[1] in some cases
|
|
|
|
childDomElements[0].veInternal = ve.extendObject(
|
|
|
|
{ 'childDomElements': childDomElements },
|
2013-07-28 20:51:32 +00:00
|
|
|
dataElement.internal ? ve.copy( dataElement.internal ) : {}
|
2013-04-17 17:53:26 +00:00
|
|
|
);
|
|
|
|
// Add elements
|
|
|
|
for ( j = 0; j < childDomElements.length; j++ ) {
|
|
|
|
domElement.appendChild( childDomElements[j] );
|
|
|
|
}
|
|
|
|
// Descend into the first child node
|
|
|
|
parentDomElement = domElement;
|
|
|
|
domElement = childDomElements[0];
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
|
2013-04-17 17:53:26 +00:00
|
|
|
// Process outer whitespace
|
|
|
|
// Every piece of outer whitespace is duplicated somewhere:
|
|
|
|
// each node's outerPost is duplicated as the next node's
|
|
|
|
// outerPre, the first node's outerPre is the parent's
|
|
|
|
// innerPre, and the last node's outerPost is the parent's
|
|
|
|
// innerPost. For each piece of whitespace, we verify that
|
|
|
|
// the duplicate matches. If it doesn't, we take that to
|
|
|
|
// mean the user has messed with it and don't output any
|
|
|
|
// whitespace.
|
|
|
|
if ( domElement.veInternal && domElement.veInternal.whitespace ) {
|
|
|
|
// Process this node's outerPre
|
|
|
|
ours = domElement.veInternal.whitespace[0];
|
|
|
|
theirs = undefined;
|
|
|
|
if ( domElement.previousSibling ) {
|
|
|
|
// Get previous sibling's outerPost
|
|
|
|
theirs = parentDomElement.lastOuterPost;
|
|
|
|
} else if ( parentDomElement === container ) {
|
|
|
|
// outerPre of the very first node in the document, this one
|
|
|
|
// has no duplicate
|
|
|
|
theirs = ours;
|
|
|
|
} else {
|
|
|
|
// First child, get parent's innerPre
|
|
|
|
if (
|
|
|
|
parentDomElement.veInternal &&
|
|
|
|
parentDomElement.veInternal.whitespace
|
|
|
|
) {
|
|
|
|
theirs = parentDomElement.veInternal.whitespace[1];
|
|
|
|
// Clear after use so it's not used twice
|
|
|
|
parentDomElement.veInternal.whitespace[1] = undefined;
|
|
|
|
}
|
|
|
|
// else theirs=undefined
|
|
|
|
}
|
|
|
|
if ( ours && ours === theirs ) {
|
|
|
|
// Matches the duplicate, insert a TextNode
|
|
|
|
textNode = doc.createTextNode( ours );
|
|
|
|
textNode.veIsWhitespace = true;
|
|
|
|
parentDomElement.insertBefore(
|
|
|
|
textNode,
|
|
|
|
domElement
|
|
|
|
);
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
Allow nodes to handle their own children
For data->DOM, this is easy: .toDataElements() can optionally return an
array instead of an object, and that will be treated as the data to
insert. If this happens, the converter won't descend. The node handler
can recursively invoke the converter if it needs to (although I suspect
the current implementation is broken when converting block content in an
inline context).
For DOM->data, this is a bit more complex. The node sets
.static.handlesOwnChildren = true; , which triggers the converter to
pass a data slice rather than a single data element, and not to
descend. The node handler can invoke the converter to recursively
convert DOM subtrees to data.
ve.dm.Converter (data->DOM):
* Renamed createDataElement() to createDataElements()
** .toDataElement() may return element or array, handle this
* Renamed childDataElement to childDataElements, is now an array
* Actually alienate if .toDataElement() returns null
** Shockingly, this claimed to be supported before but wasn't
* Rather than pushing to data, concat to it
** Add closing if needed
* Don't descend if .toDataElement() returned an array of length >1, or
if the node has .handlesOwnChildren = true
ve.dm.Converter (DOM->data):
* Split getDomSubtreeFromData() and getDomFromData()
* When converting a node that handles its own children, pass in a data
slice and skip over that data
Change-Id: I196cb4c0895cbf0b428a189adb61b56565573ab3
2013-04-11 00:31:09 +00:00
|
|
|
|
|
|
|
if ( ve.isArray( dataElementOrSlice ) ) {
|
|
|
|
i += dataElementOrSlice.length - 2;
|
|
|
|
}
|
2012-06-08 04:58:56 +00:00
|
|
|
}
|
2012-06-07 22:02:25 +00:00
|
|
|
}
|
2012-06-07 00:47:27 +00:00
|
|
|
}
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Process the outerPost whitespace of the very last node
|
|
|
|
if ( container.lastOuterPost !== undefined ) {
|
2013-03-04 16:24:09 +00:00
|
|
|
if ( container.lastChild && container.lastChild.nodeType === Node.TEXT_NODE ) {
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
// Last child is a TextNode, append to it
|
|
|
|
container.lastChild.appendData( container.lastOuterPost );
|
|
|
|
} else {
|
|
|
|
// Append a TextNode
|
2013-02-11 19:46:58 +00:00
|
|
|
container.appendChild( doc.createTextNode( container.lastOuterPost ) );
|
Preserve whitespace between elements
This commit fully utilizes all four positions in the internal.whitespace
array. Outer whitespace is now preserved as well, and is duplicated
either in the adjacent sibling (one node's outerPost is the next
sibling's outerPre) or in the parent (a branch node's innerPre is its
first child's outerPre, and its innerPost is its last child's
outerPost). Before restoring saved whitespace, we check that these two
agree with each other, and if they disagree we assume the user has been
moving stuff around and don't restore any whitespace in that spot. The
whitespace at the very beginning and the very end of the document (i.e.
the first node's outerPre and the last node's outerPost) isn't
duplicated anywhere, nor is inner whitespace in content nodes.
The basic outline of the implementation is:
* When we encounter whitespace, strip it and store it in the previous
node's outerPost. Also store it in nextWhitespace so we can put it in
the next node's outerPre once we encounter that node.
* When we encounter whitespace in wrapped bare text, we don't know in
advance if it's gonna be succeeded by more non-whitespace (in which
case it needs to be output verbatim), or not (in which case it's
leading whitespace and needs to be stripped and stored). The fact that
annotations are nodes in HTML makes this trickier. So we write the
whitespace to the temporary linmod and store it in wrappedWhitespace,
then if it turns out to be trailing whitespace we take it back out of
the data array and record it the usual way.
* Because text nodes can contain any combination of leading whitespace
actual text and trailing whitespace, and because we may or may not
already have opened a wrapping paragraph, there are a lot of different
combinations to handle. We handle all of them but the resulting code
is pretty dense and verbose.
More low-level list of changes:
In getDataFromDom():
* Added helper function addWhitespace() for storing whitespace for an
element
* Added helper function processNextWhitespace() for processing any
whitespace passed on from the previous node via the nextWhitespace var
* Rename paragraph to wrappingParagraph. Make wrapping default to
alreadyWrapped so we can simplify wrapping||alreadyWrapped and
!wrapping&&!alreadyWrapped. Add wrappingIsOurs to track whether the
wrapping originated in this recursion level (needed for deciding when
to close the wrapper).
* Add prevElement to track the previous element so we can propagate
whitespace to it, and nextWhitespace so we can propagate whitespace to
the next element.
* Remove previous newline stripping hacks
* Integrate the logic for wrapping bare content with the outer
whitespace preservation code
* Remove wrapperElement, no longer needed because we have a dedicated
variable for the wrapping paragraph now and what was previously inner
whitespace preservation for wrapper paragraphs is now covered by the
outer whitespace preservation code.
In getDomFromData():
* Reinsert whitespace where appropriate
** outerPre is inserted when opening the element
** This covers outerPost as well except for the last child's outerPost,
which is handled as the parent's innerPost when closing the parent.
** innerPre and innerPost are inserted when closing the element. Care is
taken not to insert these if they're duplicates of something else.
* Propagate each node's outerPost to the next node (either the next
sibling or the parent) using parentDomElement.lastOuterPost. We can't
get this using .lastChild because we will have destroyed that child's
.veInternal by then, and we can't tell whether a node will be its
parent's last child when we process it (all other processing,
including first child handling is done when processing the node itself,
but this cannot be).
* Special handling is needed for the last node's outerPost, which ends
up in the container's .lastOuterPost property.
Tests:
* Allow .html to be null in data<->DOM converter tests. This indicates
that the test is a one-way data->DOM test, not a DOM->data->DOM
round-trip test. The data will be converted to HTML and checked
against .normalizedHtml
* Update existing tests as needed
* Add tests for outer whitespace preservation and storage
* Add test for squashing of whitespace in case of disagreement (this
requires .html=null)
Change-Id: I4db4fe372a421182e80a2535657af7784ff15f95
2012-08-21 00:37:42 +00:00
|
|
|
}
|
|
|
|
delete container.lastOuterPost;
|
|
|
|
}
|
2013-06-26 02:18:59 +00:00
|
|
|
// Get rid of excess text nodes
|
|
|
|
container.normalize();
|
2012-06-06 17:17:30 +00:00
|
|
|
};
|
|
|
|
|
2012-05-31 23:50:16 +00:00
|
|
|
/* Initialization */
|
|
|
|
|
2013-02-21 23:01:04 +00:00
|
|
|
ve.dm.converter = new ve.dm.Converter( ve.dm.modelRegistry, ve.dm.nodeFactory, ve.dm.annotationFactory, ve.dm.metaItemFactory );
|