Release 0.1.0.

Edward Z. Yang ext:(%22) · Edward Z. Yang ext:(%22) · commit d4df251ea534 · 2009-06-01T21:11:21.000-04:00
diff --git a/README b/README
@@ -1,15 +1,47 @@
-This implementation doesn't work yet. Bits and pieces are getting written
-at a time.
+html5lib - php flavour
 
-This is an implementation of the HTML5 specification for PHP. More friendly
-details forthcoming, but here are some notes:
+This is an implementation of the tokenization and tree-building parts
+of the HTML5 specification in PHP.  Potential uses of this library
+can be found in web-scrapers and HTML filters.
+
+Warning: This is a pre-alpha release, and as such, certain parts of
+this code are not up-to-snuff (e.g. error reporting and performance).
+However, the code is very close to spec and passes 100% of tests
+not related to parse errors.  Nevertheless, expect to have to update
+your code on the next upgrade.
+
+
+Usage notes:
+
+    <?php
+    require_once '/path/to/HTML5/Parser.php';
+    $dom = HTML5_Parser::parse('<html><body>...');
+    $nodelist = HTML5_Parser::parseFragment('<b>Boo</b><br>');
+    $nodelist = HTML5_Parser::parseFragment('<td>Bar</td>', 'table');
+
+
+Documentation:
+
+HTML5_Parser::parse($text)
+    $text  : HTML to parse
+    return : DOMDocument of parsed document
+
+HTML5_Parser::parseFragment($text, $context)
+    $text    : HTML to parse
+    $context : String name of context element
+    return   : DOMDocument of parsed document
+
+
+Developer notes:
 
 * To setup unit tests, you need to add a small stub file test-settings.php
   that contains $simpletest_location = 'path/to/simpletest/'; This needs to
   be version 1.1 (or, until that is released, SVN trunk) of SimpleTest.
 
 * We don't want to ultimately use PHP's DOM because it is not tolerant
   of certain types of errors that HTML 5 allows (for example, an element
-  "foo@bar"). But for now, we will, since it's much easier.
+  "foo@bar"). But the current implementation uses it, since it's easy.
+  Eventually, this html5lib implementation will get a version of SimpleTree;
+  and may possibly start using that by default.
 
     vim: et sw=4 sts=4
diff --git a/VERSION b/VERSION
@@ -0,0 +1 @@
+0.1.0
diff --git a/library/HTML5/Parser.php b/library/HTML5/Parser.php
@@ -1,6 +1,36 @@
 <?php
 
+require_once dirname(__FILE__) . '/Data.php';
+require_once dirname(__FILE__) . '/InputStream.php';
+require_once dirname(__FILE__) . '/TreeBuilder.php';
+require_once dirname(__FILE__) . '/Tokenizer.php';
+
+/**
+ * Outwards facing interface for HTML5.
+ */
 class HTML5_Parser
 {
-
+    /**
+     * Parses a full HTML document.
+     * @param $text HTML text to parse
+     * @param $builder Custom builder implementation
+     * @return Parsed HTML as DOMDocument
+     */
+    static public function parse($text, $builder = null) {
+        $tokenizer = new HTML5_Tokenizer($text, $builder);
+        $tokenizer->parse();
+        return $tokenizer->save();
+    }
+    /**
+     * Parses an HTML fragment.
+     * @param $text HTML text to parse
+     * @param $context String name of context element to pretend parsing is in.
+     * @param $builder Custom builder implementation
+     * @return Parsed HTML as DOMDocument
+     */
+    static public function parseFragment($text, $context = null, $builder = null) {
+        $tokenizer = new HTML5_Tokenizer($text, $builder);
+        $tokenizer->parseFragment($context);
+        return $tokenizer->save();
+    }
 }
diff --git a/library/HTML5/Tokenizer.php b/library/HTML5/Tokenizer.php
@@ -84,9 +84,9 @@ class HTML5_Tokenizer {
     /**
      * @param $data Data to parse
      */
-    public function __construct($data) {
+    public function __construct($data, $builder = null) {
         $this->stream = new HTML5_InputStream($data);
-        $this->tree = new HTML5_TreeBuilder;
+        if (!$builder) $this->tree = new HTML5_TreeBuilder;
         $this->content_model = self::PCDATA;
     }
 
diff --git a/library/HTML5/TreeBuilder.php b/library/HTML5/TreeBuilder.php
@@ -121,7 +121,7 @@ private function strConst($number) {
     const MARKER     = 300;
 
     // Namespaces for foreign content
-    const NS_HTML   = 'http://www.w3.org/1999/xhtml';
+    const NS_HTML   = null; // to prevent DOM from requiring NS on everything
     const NS_MATHML = 'http://www.w3.org/1998/Math/MathML';
     const NS_SVG    = 'http://www.w3.org/2000/svg';
     const NS_XLINK  = 'http://www.w3.org/1999/xlink';
@@ -3516,8 +3516,8 @@ public function currentTableIsTainted() {
      */
     public function setupContext($context = null) {
         $this->fragment = true;
-        $context = $this->dom->createElementNS(self::NS_HTML, $context);
         if ($context) {
+            $context = $this->dom->createElementNS(self::NS_HTML, $context);
             /* 4.1. Set the HTML parser's tokenization  stage's content model
              * flag according to the context element, as follows: */
             switch ($context->tagName) {
diff --git a/tests/HTML5/ParserTest.php b/tests/HTML5/ParserTest.php
@@ -0,0 +1,15 @@
+<?php
+
+require_once dirname(__FILE__) . '/../autorun.php';
+
+class HTML5_ParserTest extends UnitTestCase
+{
+    public function testParse() {
+        $result = HTML5_Parser::parse('<html><body></body></html>');
+        $this->assertIsA($result, 'DOMDocument');
+    }
+    public function testParseFragment() {
+        $result = HTML5_Parser::parseFragment('<b>asdf</b> foo');
+        $this->assertIsA($result, 'DOMNodeList');
+    }
+}
