HTML_Processor gets wrong breadcrumbs for elements in <head>

[vicobot]

If you visit elements that typically appear in the <head>, like META or LINK, the HTML_Processor returns a breadcrumbs array with BODY instead of HEAD, e.g.

Array
(
[0] => HTML
[1] => BODY
[2] => META
)

How to reproduce:

• call $processor = \WP_HTML_Processor::create_fragment($html) on a full HTML page, e.g. by hooking into wp_template_enhancement_output_buffer
• get $processor->next_tag('META') or some other tag that lives in <head>
• get $processor->get_breadcrumbs() and check the resulting array, e.g. with print_r() or by doing an array_diff() with the expected result ['HTML', 'HEAD', 'META']

Wordpress: 6.9 (php8.2-apache docker)

[westonruter]

@vicobot this is because you're parsing in BODY mode. You need to use WP_HTML_Processor::create_full_parser() instead of WP_HTML_Processor::create_fragment().

[dmsnell]

oops, I was prepping my response while @westonruter was responding. to add to what he wrote, the difference between those methods is that create_fragment() is specifically designed to operate within the context of inner HTML inside a specified element, the default being BODY.

Use this for cases where you are processing chunks of HTML that will be found within a bigger HTML document, such as rendered block output that exists within a post, the_content inside a rendered site layout.

If you use create_full_parser() it will assume that you are providing the full HTML for a page from start to finish and the initial META elements will appear within a HEAD. However, as ​per the HTML spec, when parsing a META tag, an element is to be created in the current element regardless of the parser’s current insertion mode.

You can see this ​demonstrated using your browser’s interpretation of the HTML, where META remains inside the BODY