Changeset 55667

Timestamp:

04/20/2023 05:08:47 PM (2 years ago)

Author:

Bernhard Reiter

Message:

HTML API: Add support for a few invalid HTML comment forms.

• Comments created by means of a tag closer with an invalid tag name, e.g. </3>.
• Comments closed with the invalid --!> closer. (Comments should be closed by --> but if the ! appears it will also close it, in error.)
• Empty tag name elements, which are technically skipped over and aren't comments, e.g. </>.

Props dmsnell, costdev.
Fixes #58007.

Location:

trunk

Files:

• src/wp-includes/html-api/class-wp-html-tag-processor.php (modified) (3 diffs)
• tests/phpunit/tests/html-api/wpHtmlTagProcessor.php (modified) (2 diffs)

trunk/src/wp-includes/html-api/class-wp-html-tag-processor.php

@@ -972 +972 @@
      *
      * @since 6.2.0
+     * @since 6.2.1 Support abruptly-closed comments, invalid-tag-closer-comments, and empty elements.
      *
      * @return bool Whether a tag was found before the end of the document.
@@ -1040 +1041 @@
                     '-' === $html[ $at + 3 ]
                 ) {
-                    $closer_at = strpos( $html, '-->', $at + 4 );
-                    if ( false === $closer_at ) {
+                    $closer_at = $at + 4;
+                    // If it's not possible to close the comment then there is nothing more to scan.
+                    if ( strlen( $html ) <= $closer_at ) {
                         return false;
                     }
 
-                    $at = $closer_at + 3;
-                    continue;
+                    // Abruptly-closed empty comments are a sequence of dashes followed by `>`.
+                    $span_of_dashes = strspn( $html, '-', $closer_at );
+                    if ( '>' === $html[ $closer_at + $span_of_dashes ] ) {
+                        $at = $closer_at + $span_of_dashes + 1;
+                        continue;
+                    }
+
+                    /*
+                     * Comments may be closed by either a --> or an invalid --!>.
+                     * The first occurrence closes the comment.
+                     *
+                     * See https://html.spec.whatwg.org/#parse-error-incorrectly-closed-comment
+                     */
+                    $closer_at--; // Pre-increment inside condition below reduces risk of accidental infinite looping.
+                    while ( ++$closer_at < strlen( $html ) ) {
+                        $closer_at = strpos( $html, '--', $closer_at );
+                        if ( false === $closer_at ) {
+                            return false;
+                        }
+
+                        if ( $closer_at + 2 < strlen( $html ) && '>' === $html[ $closer_at + 2 ] ) {
+                            $at = $closer_at + 3;
+                            continue 2;
+                        }
+
+                        if ( $closer_at + 3 < strlen( $html ) && '!' === $html[ $closer_at + 2 ] && '>' === $html[ $closer_at + 3 ] ) {
+                            $at = $closer_at + 4;
+                            continue 2;
+                        }
+                    }
                 }
 
@@ -1106 +1136 @@
 
             /*
+             * </> is a missing end tag name, which is ignored.
+             *
+             * See https://html.spec.whatwg.org/#parse-error-missing-end-tag-name
+             */
+            if ( '>' === $html[ $at + 1 ] ) {
+                $at++;
+                continue;
+            }
+
+            /*
              * <? transitions to a bogus comment state – skip to the nearest >
-             * https://html.spec.whatwg.org/multipage/parsing.html#tag-open-state
+             * See https://html.spec.whatwg.org/multipage/parsing.html#tag-open-state
              */
             if ( '?' === $html[ $at + 1 ] ) {
                 $closer_at = strpos( $html, '>', $at + 2 );
+                if ( false === $closer_at ) {
+                    return false;
+                }
+
+                $at = $closer_at + 1;
+                continue;
+            }
+
+            /*
+             * If a non-alpha starts the tag name in a tag closer it's a comment.
+             * Find the first `>`, which closes the comment.
+             *
+             * See https://html.spec.whatwg.org/#parse-error-invalid-first-character-of-tag-name
+             */
+            if ( $this->is_closing_tag ) {
+                $closer_at = strpos( $html, '>', $at + 3 );
                 if ( false === $closer_at ) {
                     return false;

trunk/tests/phpunit/tests/html-api/wpHtmlTagProcessor.php

@@ -1734 +1734 @@
 
     /**
+     * Invalid tag names are comments on tag closers.
+     *
+     * @ticket 58007
+     *
+     * @link https://html.spec.whatwg.org/#parse-error-invalid-first-character-of-tag-name
+     *
+     * @dataProvider data_next_tag_ignores_invalid_first_character_of_tag_name_comments
+     *
+     * @param string $html_with_markers HTML containing an invalid tag closer whose element before and
+     *                                  element after contain the "start" and "end" CSS classes.
+     */
+    public function test_next_tag_ignores_invalid_first_character_of_tag_name_comments( $html_with_markers ) {
+        $p = new WP_HTML_Tag_Processor( $html_with_markers );
+        $p->next_tag( array( 'class_name' => 'start' ) );
+        $p->next_tag();
+
+        $this->assertSame( 'end', $p->get_attribute( 'class' ) );
+    }
+
+    /**
+     * Data provider.
+     *
+     * @return array[]
+     */
+    public function data_next_tag_ignores_invalid_first_character_of_tag_name_comments() {
+        return array(
+            'Invalid tag openers as normal text'           => array(
+                '<ul><li><div class=start>I <3 when outflow > inflow</div><img class=end></li></ul>',
+            ),
+
+            'Invalid tag closers as comments'              => array(
+                '<ul><li><div class=start>I </3 when <img> outflow <br class=end> inflow</div></li></ul>',
+            ),
+
+            'Unexpected question mark instead of tag name' => array(
+                '<div class=start><?xml-stylesheet type="text/css" href="style.css"?><hr class=end>',
+            ),
+        );
+    }
+
+    /**
      * @ticket 56299
      *
@@ -1782 +1823 @@
                 'rcdata_tag'      => 'TEXTAREA',
             ),
+        );
+    }
+
+    /**
+     * Ensures that the invalid comment closing syntax "--!>" properly closes a comment.
+     *
+     * @ticket 58007
+     *
+     * @covers WP_HTML_Tag_Processor::next_tag
+     *
+     */
+    public function test_allows_incorrectly_closed_comments() {
+        $p = new WP_HTML_Tag_Processor( '<img id=before><!-- <img id=inside> --!><img id=after>--><img id=final>' );
+
+        $p->next_tag();
+        $this->assertSame( 'before', $p->get_attribute( 'id' ), 'Did not find starting tag.' );
+
+        $p->next_tag();
+        $this->assertSame( 'after', $p->get_attribute( 'id' ), 'Did not properly close improperly-closed comment.' );
+
+        $p->next_tag();
+        $this->assertSame( 'final', $p->get_attribute( 'id' ), 'Did not skip over unopened comment-closer.' );
+    }
+
+    /**
+     * Ensures that unclosed and invalid comments don't trigger warnings or errors.
+     *
+     * @ticket 58007
+     *
+     * @covers WP_HTML_Tag_Processor::next_tag
+     *
+     * @dataProvider data_html_with_unclosed_comments
+     *
+     * @param string $html_ending_before_comment_close HTML with opened comments that aren't closed
+     */
+    public function test_documents_may_end_with_unclosed_comment( $html_ending_before_comment_close ) {
+        $p = new WP_HTML_Tag_Processor( $html_ending_before_comment_close );
+
+        $this->assertFalse( $p->next_tag() );
+    }
+
+    /**
+     * Data provider.
+     *
+     * @return array[]
+     */
+    public function data_html_with_unclosed_comments() {
+        return array(
+            'Shortest open valid comment'      => array( '<!--' ),
+            'Basic truncated comment'          => array( '<!-- this ends --' ),
+            'Comment with closer look-alike'   => array( '<!-- this ends --x' ),
+            'Comment with closer look-alike 2' => array( '<!-- this ends --!x' ),
+            'Invalid tag-closer comment'       => array( '</(when will this madness end?)' ),
+            'Invalid tag-closer comment 2'     => array( '</(when will this madness end?)--' ),
+        );
+    }
+
+    /**
+     * Ensures that abruptly-closed empty comments are properly closed.
+     *
+     * @ticket 58007
+     *
+     * @covers WP_HTML_Tag_Processor::next_tag
+     *
+     * @dataProvider data_abruptly_closed_empty_comments
+     *
+     * @param string $html_with_after_marker HTML to test with "id=after" on element immediately following an abruptly closed comment.
+     */
+    public function test_closes_abrupt_closing_of_empty_comment( $html_with_after_marker ) {
+        $p = new WP_HTML_Tag_Processor( $html_with_after_marker );
+        $p->next_tag();
+        $p->next_tag();
+
+        $this->assertSame( 'after', $p->get_attribute( 'id' ), 'Did not find tag after closing abruptly-closed comment' );
+    }
+
+    /**
+     * Data provider.
+     *
+     * @return array[]
+     */
+    public function data_abruptly_closed_empty_comments() {
+        return array(
+            'Empty comment with two dashes only' => array( '<hr><!--><hr id=after>' ),
+            'Empty comment with two dashes only, improperly closed' => array( '<hr><!--!><hr id=inside>--><hr id=after>' ),
+            'Comment with two dashes only, improperly closed twice' => array( '<hr><!--!><hr id=inside>--!><hr id=after>' ),
+            'Empty comment with three dashes'    => array( '<hr><!---><hr id=after>' ),
+            'Empty comment with three dashes, improperly closed' => array( '<hr><!---!><hr id=inside>--><hr id=after>' ),
+            'Comment with three dashes, improperly closed twice' => array( '<hr><!---!><hr id=inside>--!><hr id=after>' ),
+            'Empty comment with four dashes'     => array( '<hr><!----><hr id=after>' ),
+            'Empty comment with four dashes, improperly closed' => array( '<hr><!----!><hr id=after>--><hr id=final>' ),
+            'Comment with four dashes, improperly closed twice' => array( '<hr><!----!><hr id=after>--!><hr id=final>' ),
+            'Comment with almost-closer inside'  => array( '<hr><!-- ---!><hr id=after>--!><hr id=final>' ),
         );
     }