mediawiki-extensions-Visual.../modules/parser/mediawiki.HTML5TreeBuilder.node.js

151 lines
4.3 KiB
JavaScript
Raw Normal View History

/* Front-end/Wrapper for a particular tree builder, in this case the
* parser/tree builder from the node 'html5' module. Feed it tokens using
* processToken, and it will build you a DOM tree retrievable using .document
* or .body(). */
2011-11-18 14:00:14 +00:00
var events = require('events'),
HTML5 = require('./html5/index');
2011-11-18 14:00:14 +00:00
FauxHTML5 = {};
2011-11-18 14:00:14 +00:00
FauxHTML5.TreeBuilder = function ( env ) {
// The parser we are going to emit our tokens to
2011-11-18 14:00:14 +00:00
this.parser = new HTML5.Parser();
// Sets up the parser
2011-11-18 14:00:14 +00:00
this.parser.parse(this);
// implicitly start a new document
2012-02-01 18:37:48 +00:00
this.processToken(new TagTk( 'body' ));
this.env = env;
};
2011-11-18 14:00:14 +00:00
// Inherit from EventEmitter
FauxHTML5.TreeBuilder.prototype = new events.EventEmitter();
FauxHTML5.TreeBuilder.prototype.constructor = FauxHTML5.TreeBuilder;
2011-11-18 14:00:14 +00:00
/**
* Register for (token) 'chunk' and 'end' events from a token emitter,
* normally the TokenTransformDispatcher.
*/
Biggish token transform system refactoring * All parser pipelines including tokenizer and DOM stuff are now constructed from a 'recipe' data structure in a ParserPipelineFactory. * All sub-pipelines of these can now be cached * Event registrations to a pipeline are directly forwarded to the last pipeline member to save relatively expensive event forwarding. * Some APIs for on-demand expansion / format conversion of parameters from parser functions are added: param.to('tokens/expanded', cb) param.to('text/wiki', cb) (this does not work yet) All parameters are additionally wrapped into a Param object that provides method for positional parameter naming (.named() or conversion to a dict (.dict()). * The async token transform manager is now separated from a frame object, with the frame holding arguments, an on-demand expansion method and loop checks. * Only keys of template parameters are now expanded. Parser functions or template arguments trigger an expansion on-demand. This (unsurprisingly) makes a big performance difference with typical switch-heavy template systems. * Return values from async transforms are no longer used in favor of plain callbacks. This saves the complication of having to maintain two code paths. A trick in transformTokens still avoids the construction of unneeded TokenAccumulators. * The results of template expansions are no longer buffered. * 301 parser tests are passing Known issues: * Cosmetic cleanup remains to do * Some parser functions do not support async expansions yet, and need to be modified. Change-Id: I1a7690baffbe8141cadf67270904a1b2e1df879a
2012-04-25 14:35:59 +00:00
FauxHTML5.TreeBuilder.prototype.addListenersOn = function ( emitter ) {
emitter.addListener('chunk', this.onChunk.bind( this ) );
emitter.addListener('end', this.onEnd.bind( this ) );
};
FauxHTML5.TreeBuilder.prototype.onChunk = function ( tokens ) {
this.env.dp( 'chunk: ' + JSON.stringify( tokens, null, 2 ) );
for (var i = 0, length = tokens.length; i < length; i++) {
this.processToken(tokens[i]);
}
};
FauxHTML5.TreeBuilder.prototype.onEnd = function ( ) {
//console.warn('Fauxhtml5 onEnd');
// FIXME HACK: For some reason the end token is not processed sometimes,
// which normally fixes the body reference up.
var document = this.parser.document;
document.body = document.getElementsByTagName('body')[0];
//console.warn( 'onEnd: ' + document.body.innerHTML );
this.emit( 'document', document );
// XXX: more clean up to allow reuse.
this.parser.setup();
2012-02-01 18:37:48 +00:00
this.processToken(new TagTk( 'body' ));
};
FauxHTML5.TreeBuilder.prototype._att = function (maybeAttribs) {
var atts = [];
if ( maybeAttribs && $.isArray( maybeAttribs ) ) {
for(var i = 0, length = maybeAttribs.length; i < length; i++) {
var att = maybeAttribs[i];
atts.push({nodeName: att.k, nodeValue: att.v});
}
}
return atts;
};
// Adapt the token format to internal HTML tree builder format, call the actual
// html tree builder by emitting the token.
FauxHTML5.TreeBuilder.prototype.processToken = function (token) {
//console.warn( 'processToken: ' + JSON.stringify( token ));
Big token transform framework overhaul part 2 * Tokens are now immutable. The progress of transformations is tracked on chunks instead of tokens. Tokenizer output is cached and can be directly returned without a need for cloning. Transforms are required to clone or newly create tokens they are modifying. * Expansions per chunk are now shared between equivalent frames via a cache stored on the chunk itself. Equivalence of frames is not yet ideal though, as right now a hash tree of *unexpanded* arguments is used. This should be switched to a hash of the fully expanded local parameters instead. * There is now a vastly improved maybeSyncReturn wrapper for async transforms that either forwards processing to the iterative transformTokens if the current transform is still ongoing, or manages a recursive transformation if needed. * Parameters for parser functions are now wrapped in abstract Params and ParserValue objects, which support some handy on-demand *value* expansions. Keys are always expanded. Parser functions are converted to use these interfaces, and now properly expand their values in the correct frame. Making this expansion lazier is certainly possible, but would complicate transformTokens and other token-handling machinery. Need to investigate if it would really be worth it. Dead branch elimination is certainly a bigger win overall. * Complex recursive asynchronous expansions should now be closer to correct for both the iterative (transformTokens) and recursive (maybeSyncReturn after transformTokens has returned) code paths. * Performance degraded slightly. There are no micro-optimizations done yet and the shared expansion cache still has a low hit rate. The progress tracking on chunks is not yet perfect, so there are likely a lot of unneeded re-expansions that can be easily eliminated. There is also more debug tracing right now. Obama currently expands in 54 seconds on my laptop. Change-Id: I4a603f3d3c70ca657ebda9fbb8570269f943d6b6
2012-05-10 08:04:24 +00:00
var attribs = token.attribs || [];
if ( token.dataAttribs ) {
var dataMW = JSON.stringify( token.dataAttribs );
if ( dataMW !== '{}' ) {
attribs = attribs.concat([
{
// Mediawiki-specific round-trip / non-semantic information
k: 'data-parsoid',
v: dataMW
} ] );
}
}
2011-11-18 14:00:14 +00:00
switch( token.constructor ) {
case String:
if ( token.match(/^[ \t\r\n\f]+$/) ) {
// Treat space characters specially so that the tree builder
// doesn't apply the foster parenting algorithm
this.emit('token', {type: 'SpaceCharacters', data: token});
} else {
this.emit('token', {type: 'Characters', data: token});
}
break;
case NlTk:
this.emit('token', {type: 'SpaceCharacters', data: '\n'});
break;
2012-02-01 18:37:48 +00:00
case TagTk:
this.emit('token', {type: 'StartTag',
name: token.name,
Big token transform framework overhaul part 2 * Tokens are now immutable. The progress of transformations is tracked on chunks instead of tokens. Tokenizer output is cached and can be directly returned without a need for cloning. Transforms are required to clone or newly create tokens they are modifying. * Expansions per chunk are now shared between equivalent frames via a cache stored on the chunk itself. Equivalence of frames is not yet ideal though, as right now a hash tree of *unexpanded* arguments is used. This should be switched to a hash of the fully expanded local parameters instead. * There is now a vastly improved maybeSyncReturn wrapper for async transforms that either forwards processing to the iterative transformTokens if the current transform is still ongoing, or manages a recursive transformation if needed. * Parameters for parser functions are now wrapped in abstract Params and ParserValue objects, which support some handy on-demand *value* expansions. Keys are always expanded. Parser functions are converted to use these interfaces, and now properly expand their values in the correct frame. Making this expansion lazier is certainly possible, but would complicate transformTokens and other token-handling machinery. Need to investigate if it would really be worth it. Dead branch elimination is certainly a bigger win overall. * Complex recursive asynchronous expansions should now be closer to correct for both the iterative (transformTokens) and recursive (maybeSyncReturn after transformTokens has returned) code paths. * Performance degraded slightly. There are no micro-optimizations done yet and the shared expansion cache still has a low hit rate. The progress tracking on chunks is not yet perfect, so there are likely a lot of unneeded re-expansions that can be easily eliminated. There is also more debug tracing right now. Obama currently expands in 54 seconds on my laptop. Change-Id: I4a603f3d3c70ca657ebda9fbb8570269f943d6b6
2012-05-10 08:04:24 +00:00
data: this._att(attribs)});
2012-02-01 18:37:48 +00:00
break;
case SelfclosingTagTk:
this.emit('token', {type: 'StartTag',
name: token.name,
Big token transform framework overhaul part 2 * Tokens are now immutable. The progress of transformations is tracked on chunks instead of tokens. Tokenizer output is cached and can be directly returned without a need for cloning. Transforms are required to clone or newly create tokens they are modifying. * Expansions per chunk are now shared between equivalent frames via a cache stored on the chunk itself. Equivalence of frames is not yet ideal though, as right now a hash tree of *unexpanded* arguments is used. This should be switched to a hash of the fully expanded local parameters instead. * There is now a vastly improved maybeSyncReturn wrapper for async transforms that either forwards processing to the iterative transformTokens if the current transform is still ongoing, or manages a recursive transformation if needed. * Parameters for parser functions are now wrapped in abstract Params and ParserValue objects, which support some handy on-demand *value* expansions. Keys are always expanded. Parser functions are converted to use these interfaces, and now properly expand their values in the correct frame. Making this expansion lazier is certainly possible, but would complicate transformTokens and other token-handling machinery. Need to investigate if it would really be worth it. Dead branch elimination is certainly a bigger win overall. * Complex recursive asynchronous expansions should now be closer to correct for both the iterative (transformTokens) and recursive (maybeSyncReturn after transformTokens has returned) code paths. * Performance degraded slightly. There are no micro-optimizations done yet and the shared expansion cache still has a low hit rate. The progress tracking on chunks is not yet perfect, so there are likely a lot of unneeded re-expansions that can be easily eliminated. There is also more debug tracing right now. Obama currently expands in 54 seconds on my laptop. Change-Id: I4a603f3d3c70ca657ebda9fbb8570269f943d6b6
2012-05-10 08:04:24 +00:00
data: this._att(attribs)});
2012-02-01 18:37:48 +00:00
if ( HTML5.VOID_ELEMENTS.indexOf( token.name.toLowerCase() ) < 0 ) {
// VOID_ELEMENTS are automagically treated as self-closing by
// the tree builder
this.emit('token', {type: 'EndTag',
name: token.name,
Big token transform framework overhaul part 2 * Tokens are now immutable. The progress of transformations is tracked on chunks instead of tokens. Tokenizer output is cached and can be directly returned without a need for cloning. Transforms are required to clone or newly create tokens they are modifying. * Expansions per chunk are now shared between equivalent frames via a cache stored on the chunk itself. Equivalence of frames is not yet ideal though, as right now a hash tree of *unexpanded* arguments is used. This should be switched to a hash of the fully expanded local parameters instead. * There is now a vastly improved maybeSyncReturn wrapper for async transforms that either forwards processing to the iterative transformTokens if the current transform is still ongoing, or manages a recursive transformation if needed. * Parameters for parser functions are now wrapped in abstract Params and ParserValue objects, which support some handy on-demand *value* expansions. Keys are always expanded. Parser functions are converted to use these interfaces, and now properly expand their values in the correct frame. Making this expansion lazier is certainly possible, but would complicate transformTokens and other token-handling machinery. Need to investigate if it would really be worth it. Dead branch elimination is certainly a bigger win overall. * Complex recursive asynchronous expansions should now be closer to correct for both the iterative (transformTokens) and recursive (maybeSyncReturn after transformTokens has returned) code paths. * Performance degraded slightly. There are no micro-optimizations done yet and the shared expansion cache still has a low hit rate. The progress tracking on chunks is not yet perfect, so there are likely a lot of unneeded re-expansions that can be easily eliminated. There is also more debug tracing right now. Obama currently expands in 54 seconds on my laptop. Change-Id: I4a603f3d3c70ca657ebda9fbb8570269f943d6b6
2012-05-10 08:04:24 +00:00
data: this._att(attribs)});
2012-02-01 18:37:48 +00:00
}
break;
case EndTagTk:
this.emit('token', {type: 'EndTag',
name: token.name,
Big token transform framework overhaul part 2 * Tokens are now immutable. The progress of transformations is tracked on chunks instead of tokens. Tokenizer output is cached and can be directly returned without a need for cloning. Transforms are required to clone or newly create tokens they are modifying. * Expansions per chunk are now shared between equivalent frames via a cache stored on the chunk itself. Equivalence of frames is not yet ideal though, as right now a hash tree of *unexpanded* arguments is used. This should be switched to a hash of the fully expanded local parameters instead. * There is now a vastly improved maybeSyncReturn wrapper for async transforms that either forwards processing to the iterative transformTokens if the current transform is still ongoing, or manages a recursive transformation if needed. * Parameters for parser functions are now wrapped in abstract Params and ParserValue objects, which support some handy on-demand *value* expansions. Keys are always expanded. Parser functions are converted to use these interfaces, and now properly expand their values in the correct frame. Making this expansion lazier is certainly possible, but would complicate transformTokens and other token-handling machinery. Need to investigate if it would really be worth it. Dead branch elimination is certainly a bigger win overall. * Complex recursive asynchronous expansions should now be closer to correct for both the iterative (transformTokens) and recursive (maybeSyncReturn after transformTokens has returned) code paths. * Performance degraded slightly. There are no micro-optimizations done yet and the shared expansion cache still has a low hit rate. The progress tracking on chunks is not yet perfect, so there are likely a lot of unneeded re-expansions that can be easily eliminated. There is also more debug tracing right now. Obama currently expands in 54 seconds on my laptop. Change-Id: I4a603f3d3c70ca657ebda9fbb8570269f943d6b6
2012-05-10 08:04:24 +00:00
data: this._att(attribs)});
2012-02-01 18:37:48 +00:00
break;
case CommentTk:
this.emit('token', {type: 'Comment',
data: token.value});
break;
case EOFTk:
this.emit('end');
this.emit('token', { type: 'EOF' } );
this.document = this.parser.document;
if ( ! this.document.body ) {
// HACK: This should not be needed really.
this.document.body = this.parser.document.getElementsByTagName('body')[0];
}
// Emit the document to consumers
//this.emit('document', this.document);
break;
default:
console.warn("Unhandled token: " + JSON.stringify(token));
break;
2011-11-18 14:00:14 +00:00
}
};
if (typeof module == "object") {
module.exports.FauxHTML5 = FauxHTML5;
2011-11-18 14:00:14 +00:00
}