mediawiki-extensions-Visual.../tests/parser/parserTests.js
2011-11-21 09:22:30 +00:00

189 lines
4.6 KiB
JavaScript

/**
* Initial parser tests runner for experimental JS parser
*
* This pulls all the parserTests.txt items and runs them through the JS
* parser and JS HTML renderer. Currently no comparison is done on output,
* as a direct string comparison won't be very encouraging. :)
*
* Needs smarter compare, as well as search-y helpers.
*
* 2011-07-20 <brion@pobox.com>
*/
"use strict";
var fs = require('fs'),
path = require('path');
// @fixme wrap more or this setup in a common module
// Fetch up some of our wacky parser bits...
var basePath = path.join(path.dirname(path.dirname(process.cwd())), 'modules');
function _require(filename) {
return require(path.join(basePath, filename));
}
function _import(filename, symbols) {
var module = _require(filename);
symbols.forEach(function(symbol) {
global[symbol] = module[symbol];
})
}
// needed for html5 parser adapter
//var events = require('events');
// For now most modules only need this for $.extend and $.each :)
global.$ = require('jquery');
// hack for renderer
global.document = $('<div>')[0].ownerDocument;
var pj = path.join;
// Local CommonJS-friendly libs
global.PEG = _require(pj('parser', 'lib.pegjs.js'));
// Our code...
_import(pj('parser', 'mediawiki.parser.peg.js'), ['PegParser']);
_import(pj('parser', 'mediawiki.parser.environment.js'), ['MWParserEnvironment']);
_import(pj('parser', 'ext.cite.taghook.ref.js'), ['MWRefTagHook']);
_require(pj('parser', 'mediawiki.html5TokenEmitter.js'));
// WikiDom and serializers
_require(pj('es', 'es.js'));
_require(pj('es', 'es.Html.js'));
_require(pj('es', 'serializers', 'es.AnnotationSerializer.js'));
_require(pj('es', 'serializers', 'es.HtmlSerializer.js'));
_require(pj('es', 'serializers', 'es.WikitextSerializer.js'));
_require(pj('es', 'serializers', 'es.JsonSerializer.js'));
// Preload the grammar file...
PegParser.src = fs.readFileSync(path.join(basePath, 'parser', 'pegParser.pegjs.txt'), 'utf8');
var parser = new PegParser();
var testFileName = '../../../../../tests/parser/parserTests.txt'; // default
if (process.argv.length > 2) {
// hack :D
testFileName = process.argv[2];
console.log(testFileName);
}
try {
var testParser = PEG.buildParser(fs.readFileSync('parserTests.pegjs', 'utf8'));
} catch (e) {
console.log(e);
}
var testFile = fs.readFileSync(testFileName, 'utf8');
try {
var cases = testParser.parse(testFile);
} catch (e) {
console.log(e);
}
var articles = {};
function normalizeTitle(name) {
if (typeof name !== 'string') {
throw new Error('nooooooooo not a string');
}
name = name.replace(/[\s_]+/g, '_');
name = name.substr(0, 1).toUpperCase() + name.substr(1);
if (name == '') {
throw new Error('Invalid/empty title');
}
return name;
}
function fetchArticle(name) {
var norm = normalizeTitle(name);
if (norm in articles) {
return articles[norm];
}
}
function processArticle(item) {
var norm = normalizeTitle(item.title);
articles[norm] = item.text;
}
function nodeToHtml(node) {
return $('<div>').append(node).html();
}
function processTest(item) {
var tokenizer = new FauxHTML5.Tokenizer();
if (!('title' in item)) {
console.log(item);
throw new Error('Missing title from test case.');
}
if (!('input' in item)) {
console.log(item);
throw new Error('Missing input from test case ' + item.title);
}
if (!('result' in item)) {
console.log(item);
throw new Error('Missing input from test case ' + item.title);
}
console.log(item.title);
console.log("INPUT:");
console.log(item.input + "\n");
parser.parseToTree(item.input + "\n", function(tree, err) {
if (err) {
console.log('PARSE FAIL', err);
} else {
var environment = new MWParserEnvironment({
tagHooks: {
'ref': MWRefTagHook,
'references': MWReferencesTagHook
}
});
//var res = es.HtmlSerializer.stringify(tree,environment);
processTokens(tree, tokenizer);
if (err) {
console.log('RENDER FAIL', err);
} else {
console.log('EXPECTED:');
console.log(item.result + "\n");
console.log('RENDERED:');
//console.log(JSON.stringify(tree, null, 2));
console.log(tokenizer.parser.document
.getElementsByTagName('body')[0]
.innerHTML);
}
}
});
}
function processTokens ( tokens, tokenizer ) {
tokenizer.processToken({type: 'TAG', name: 'body'});
// Process all tokens
for (var i = 0, length = tokens.length; i < length; i++) {
tokenizer.processToken(tokens[i]);
}
// And signal the end
tokenizer.processToken({type: 'END'});
}
cases.forEach(function(item) {
if (typeof item == 'object') {
if (item.type == 'article') {
processArticle(item);
} else if (item.type == 'test') {
processTest(item);
}
}
});