Issue 90991 - [a11y] OOo Impress and Calc should name paragraphs starting at 1
Summary: [a11y] OOo Impress and Calc should name paragraphs starting at 1
Status: CLOSED FIXED
Alias: None
Product: General
Classification: Code
Component: code (show other issues)
Version: OOo 3.0 Beta
Hardware: All All
: P3 Trivial (vote)
Target Milestone: OOo 3.2
Assignee: williewalker
QA Contact: issues@framework
URL:
Keywords: accessibility
Depends on:
Blocks:
 
Reported: 2008-06-23 17:18 UTC by richburridge
Modified: 2010-01-08 09:11 UTC (History)
3 users (show)

See Also:
Issue Type: ENHANCEMENT
Latest Confirmation in: ---
Developer Difficulty: ---


Attachments

Note You need to log in before you can comment on or make changes to this issue.
Description richburridge 2008-06-23 17:18:38 UTC
See also Orca bug #539773 which is tracking this issue.
http://bugzilla.gnome.org/show_bug.cgi?id=539773

(This enhancement request affects Impress and Calc. I wasn't
sure how to file it, so I've placed it under "Presentation.
Please adjust accordingly).

We've noticed that when you are editing a spreadsheet cell 
with Calc or editing a slide placeholder in Impress, that the
accessible name for the first paragraph in "Paragraph 0".
Whilst most programmers have no problem counting from zero, we
feel it would be easier for the average blind OOo user if they
started from 1 (i.e. "Paragraph 1). Then when they hear
"Paragraph 3" spoken, they will know that they are on the third
paragraph and not the fourth.

Thanks.
Comment 1 eric.savary 2008-06-23 17:43:33 UTC
@OBR: as described. And apparently, Writer doesn't number its paragraphs at all.
I think it's an edit engine problematic... @TL?

@richburridge: you may file such issues under "Framework".
Comment 2 nospam4obr 2008-06-24 06:07:26 UTC
IIRC the gnopernicus team once requested to not give paragraphs any meaningless
names (as paragraph numbers don't stay sorted in writer).
Comment 3 thomas.lange 2009-03-23 16:25:07 UTC
Crosschecking with OBR I was told that this one should be handled by returning
empty strings as the accessible names for writer paragraphs.
Comment 4 malte_timmermann 2009-03-25 14:17:56 UTC
MT: I agree that there shoudnt be some automatic naming.

We work on EditEngine Accessibility for OOo 3.2 anyway, so I put this on 3.2 also.
Comment 5 thomas.lange 2009-04-01 13:48:47 UTC
As discussed with MT, getAccessibleName and getAccessibleDescription should
return an empty string for paragraphs.

When looking into this it was found that getAccessibleDescription for
ImageBullets also included the paragraph numbering. Here it should be removed
from the description. Thus the resulting description will be "Image bullet in
paragraph" only.
Comment 6 williewalker 2009-04-01 14:08:59 UTC
@tl: I'd like better understand what is mean by 'the resulting description will
be "Image bullet in paragraph" only.'

If an assistive technology can determine the object is an image bullet via other
means (e.g., looking at the text of the object), then this kind of description
seems redundant.  I'm not sure I understand the exact scenario where this
happens, though.  Can you attach an example document and/or describe this in
more detail?  I'd like to take a look at it with Orca.
Comment 7 thomas.lange 2009-04-01 14:15:52 UTC
Fixed in CWS swa11y32.

Files changed:
svx:
M      source\accessibility\AccessibleEditableTextPara.cx
M      source\accessibility\AccessibleImageBullet.cxx
M      source\accessibility\accessibility.src
M      inc\accessibility.hrc
Comment 8 thomas.lange 2009-04-01 14:23:43 UTC
tl->williewalker: It means that the the call to getAccessibleDescription for
image bullets (should be available in Impress only here) previously returned
thext like "Image bullet in paragraph: $(ARG)" where '$(ARG)' was replaced by
the index of the paragraph. And that those indices should not be part of the
description anymore. 
Comment 9 williewalker 2009-04-01 15:25:59 UTC
williewalker->tl: I'll need to test with Orca, but it seems like a description
of "image bullet for paragraph" might not be the optimal approach.  That is
separate from this bug, however, and I can file a separate request if needed.
Comment 10 williewalker 2009-04-01 23:26:42 UTC
williewalker->tl: just want to follow up that I'll also need a pointer to a
build with the fix in it and I'll give it a shot.  Thanks!
Comment 11 thomas.lange 2009-04-02 07:38:03 UTC
tl->williewalker: If you think the bullet image thing should be a seperate issue
than please submit a new one and assign it to me.

About the test build: Since I'm still fixing other issues and another colleague
starts fixing his a11y issues next week I think it will still take us some days
before we provide the test build.
Comment 12 thomas.lange 2009-07-31 10:13:01 UTC
tl->williewalker: Please verify. Thanks!
Drop me a note if you need a specific build.
Comment 13 thomas.lange 2009-07-31 10:13:39 UTC
.
Comment 14 williewalker 2009-08-04 16:05:13 UTC
This fix has been confirmed by the Orca team using the
swa11y32_2nd_en-US_SolarisIntel.tar file Thomas Lange made for me
(300m51(Build:9408)[CWS:swa11y32_2nd]).

Many thanks!
Comment 15 eric.savary 2009-08-04 22:59:41 UTC
Thanx for verifying!
Comment 16 malte_timmermann 2010-01-08 09:11:42 UTC
Fixed and integrated => closing now..