https://bugs.kde.org/show_bug.cgi?id=522026
--- Comment #9 from Sune Vuorela <[email protected]> --- (In reply to Rasmus Enevoldsen from comment #6) > I looked into the Poppler side before coming back to you. > Poppler has had a tagged PDF structure tree parser since 2013 > (https://perezdecastro.org/2013/poppler-tagged-pdf.html) and there is an > open meta-issue tracking it > (https://gitlab.freedesktop.org/poppler/poppler/-/work_items/26). More > importantly, Poppler exposes a complete, stable glib API for consuming the > structure tree — headings, paragraphs, lists, tables, figures, alt text, > reading order — via PopplerStructureElement: > https://poppler.freedesktop.org/api/glib/PopplerStructureElement.html > This API has been available since Poppler 0.26. So as far as I can tell, the > data is there for Okular to consume and pass through to Qt Accessibility / > AT-SPI. > Is there a known blocker on Okular's side that prevents this from being > implemented? The first blocker is that the Qt api in Poppler needs to expose the structured information maybe similar to the glib api, maybe differently. We also have a open bug about alt text in images needing other parts of the structured information. I don't think there is anything complex in doing it, it is just a time consuming task that no one yet have taken up. -- You are receiving this mail because: You are watching all bug changes.
