1723418 - [XFA] Image alt text not exposed

Status: VERIFIED FIXED

(Firefox :: PDF Viewer, defect, P1)

Description

[James Teh [:Jamie]]

STR (with the NVDA screen reader):

1. Open this PDF in Firefox: https://bug1671648.bmoattachments.org/attachment.cgi?id=9182033
2. Wait a few seconds for the first page to load.
3. Press control+home to ensure you are at the top of the document.
4. Press g to move to the next image.
   • Expected: NVDA should report "Canadian Flag image. Citizenship and Immigration Canada. graphic"
   • Actual: NVDA reports "page 2 region Loading... graphic", which is the pdf.js placeholder for a page that is loading. It completely skips over the image on the first page.

I'm not sure how XFA encodes alt text for images, but Acrobat reports "Canadian Flag image. Citizenship and Immigration Canada." as the label for that image.

While the image isn't particularly important in this case, an image could well provide vital information to the user in another document.

Comment 1

[James Teh [:Jamie]]

The extracted alt text could be placed in the alt attribute on the image.

Comment 2

[GitHub Bugzilla PR Linker]

Attachment: Link to GitHub master push: https://github.com/mozilla/pdf.js/commit/2cef6d879671dc193f15e213b066ff8c2062993a

Commit merged into master by GitHub
Authored by Brendan Dahl (brendandahl)

Comment 3

[Daniel Bodea [:danibodea]]

I can confirm that the alternative text "Canadian Flag image. Citizenship and Immigration Canada." and "Image of text Canada." are being read in Windows 10 with NVDA, in Ubuntu 20 with ORCA, and in Mac OS 11 with VoiceOver.
This implementation has been verified on Nightly v93.0a1 from 2021-08-15 and Beta v92.0b4, while the release channel is affected, but disabled under pref.