<a> and <img> tags in HTML Backend Lacks Information

[alex-james-bit]

It would be interesting to see a feature in HTML Backend which captures - "src" attribute value of tag and "href" attribute of tag and mention the link while parsing.
This will allow users to navigate to links if navigation is supported to those links.

I can explain with an example -

<a href="https://www.google.com"> Click Here to go to Google </a>
<img src ="https://images.google.com">

This kind of output will ensure proper coverage of related information -

[Click Here to go to Google] (https://www.google.com)
[Image Found] (https://images.google.com)

[lightsofapollo]

This would be incredibly helpful. We have use cases where we want an LLM to traverse the document and having the links is critical.

[geoHeil]

#1085 might be related though not directly from the HTML backend - but from when serializing the parsed docling object to HTML that the context of the image in the table gets lost