Upgrade StorageFeature Request: Improve outline/heading recognition for HTML snapshots (e.g. UpToDate)
Hi Zotero Team,
I would like to suggest enhancing Zotero’s ability to detect document outlines and hierarchical headings in saved HTML snapshots, especially for content from UpToDate and similar medical knowledge platforms.
These websites do not use standard heading tags. Instead, they apply custom CSS classes and ARIA attributes (e.g. aria-level="2", aria-level="3") to define title hierarchy.
Currently, Zotero fails to recognize this structure, so no collapsible outline is generated after importing the HTML file saved via SingleFile or the built-in snapshot tool.
It would be very helpful if Zotero could:
Parse ARIA heading levels (aria-level) to build document outlines;
Recognize common custom heading class names used on mainstream academic/medical sites;
Improve compatibility with HTML snapshots from medical databases like UpToDate.
This improvement will greatly boost the reading and note-taking experience for researchers and medical professionals who rely heavily on these resources.
Thank you for your consideration!
I would like to suggest enhancing Zotero’s ability to detect document outlines and hierarchical headings in saved HTML snapshots, especially for content from UpToDate and similar medical knowledge platforms.
These websites do not use standard heading tags. Instead, they apply custom CSS classes and ARIA attributes (e.g. aria-level="2", aria-level="3") to define title hierarchy.
Currently, Zotero fails to recognize this structure, so no collapsible outline is generated after importing the HTML file saved via SingleFile or the built-in snapshot tool.
It would be very helpful if Zotero could:
Parse ARIA heading levels (aria-level) to build document outlines;
Recognize common custom heading class names used on mainstream academic/medical sites;
Improve compatibility with HTML snapshots from medical databases like UpToDate.
This improvement will greatly boost the reading and note-taking experience for researchers and medical professionals who rely heavily on these resources.
Thank you for your consideration!
(If this is just ChatGPT padding out a one-sentence request with fluff, please don't do that next time - it confuses maintainers and wastes everyone's time. If your request is for aria-level elements to be recognized as headings, just say that.)
https://s3.amazonaws.com/zotero.org/images/forums/u6640029/7ks5w394oq3fi9s7xezx.png
The outline in the Zotero reader:
https://s3.amazonaws.com/zotero.org/images/forums/u6640029/ij7ajeidtcxiefnhgaal.png
I hope Zotero can recognize the outline of snapshots more flexibly.
I have sent the snapshot.
Zotero truly helps me a lot.
Love you all~