Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

NVDA does not read entire line with "report spelling errors" is not checked, winwordTextInproc branch #1964

Closed
nvaccessAuto opened this issue Nov 26, 2011 · 15 comments
Labels
Milestone

Comments

@nvaccessAuto
Copy link

Reported by surveyor on 2011-11-26 12:23
I've an E-book formatted as RTF, about 500 MS_Word
pages. I've observed that when reading it with arrow keys or with say all command, NVDA does not read all of the text in a line and using review
commands of NVDA is not possible. I've selected a section of the book and saved as a
separate doc. This time review commands worked but, the problem of not
reading all text in a line remained. I can read all with the snap in main
branch.
Attached a sample rtf file, the text in it is Turkish. But, I don't think it would be a problem.
To reproduce:
Open the document in MS_Word, mine is 2010.
Select all and copy the text to the clipboard.
Open an editor such as notepad.
Paste the text.
Focus the Word winndow and read for instance third line with NVDA.
Change to the Notepad and read the same line with NVDA.
You should see the difference.

@nvaccessAuto
Copy link
Author

Attachment test.rtf added by surveyor on 2011-11-26 12:27
Description:

@nvaccessAuto
Copy link
Author

Comment 1 by briang1 on 2011-11-26 17:21
I'm not sure about this, but I don't think I can get it to do as you say. Certainly lines get wrapped in Word, but that I assume is not what you mean, you mean its skipping whole chunks, I assume.
It might be easier if you could find an english file that does it, or is this a Turkish issue?

I am using Xp with an older version of word.

@nvaccessAuto
Copy link
Author

Comment 2 by kevinchao89 on 2011-11-26 20:00
Cannot confirm and please provide more detailed specifics, steps to reproduced, expected/actual results, and an English document.

Focus the Word winndow and read for instance

third line with NVDA.

third line is the letter "B"

Change to the Notepad and read the same line

with NVDA.

It also reads as the letter "B"

You should see the difference.

No, I don't, and please describe exactly what it is you're seeing in both cases and what you expect.

@nvaccessAuto
Copy link
Author

Comment 3 by pvagner on 2011-11-26 22:09
Originally this document displays in the protected view when open in word 2010 but once I'll get rid of protected view I can see some verry sstrange behaviour when it comes to retrieving line offsets.
In order to better understand the issue please enable reporting line numbers in the document formatting dialog.
When navigating by lines you can see it is not possible to navigate to the fourth line. If you navigate to the fourth line moving by character or by word you will notice the end of the fourth line is reported differently depending on the cursor position and / or direction of cursor movement.

@nvaccessAuto
Copy link
Author

Comment 4 by kevinchao89 on 2011-11-26 23:12
I don't think this is a NVDA issue or a screen reader issue, I tested it with NVDA 2011.3 and WinwordTextInproc 4816 and even SATogo.com. All are skipping line 4, it goes from line 3 to 5. Enabling report line numbers in NVDA did help see what the issue is.
Changes:
Changed title from "nvda/winwordTextInproc, texts partialy read" to "Turkish RTF Doc: Reading in Word will skip line 4, go from 3 to 5."

@nvaccessAuto
Copy link
Author

Comment 5 by surveyor (in reply to comment 1) on 2011-11-27 09:17
Replying to briang1:

I'm not sure about this, but I don't think I can get it to do as you say. Certainly lines get wrapped in Word, but that I assume is not what you mean, you mean its skipping whole chunks, I assume.

It might be easier if you could find an english file that does it, or is this a Turkish issue?


Ok, let me try to clerify with my inefficient English.
While reading a word document line by line or using sayAll command, NVDA does not read all the text in a line. This is valid only for nvda/winwordTextInproc branch. This time I've saved a piece from the UserGuide as a doc file for testing. For instance,
attention to the line below:
• Support for popular applications including web browsers, email clients, internet chat programs and office suites
NVDA reads it as:
• Support for popular applications including web browsers, email client

I hope this time is sufficiently clerified.

@nvaccessAuto
Copy link
Author

Attachment TestForNVDA.doc added by surveyor on 2011-11-27 09:18
Description:

@nvaccessAuto
Copy link
Author

Comment 6 by surveyor (in reply to comment 5) on 2011-11-28 11:11
I think the behavior is somehow related to "spelling error notification" function of NVDA. Because, I've discovered that, when I selected the check box about "spelling error notification" in "document formatting settings dialog of NVDA, the behavior of incomplete reading of the texts has disappeared. As I wrote before, NVDA main branch reads normally in iether cases.

Replying to surveyor:


Ok, let me try to clerify with my inefficient English.

While reading a word document line by line or using sayAll command, NVDA does not read all the text in a line. This is valid only for nvda/winwordTextInproc branch. This time I've saved a piece from the UserGuide as a doc file for testing. For instance,

attention to the line below:

• Support for popular applications including web browsers, email clients, internet chat programs and office suites

NVDA reads it as:

• Support for popular applications including web browsers, email client

I hope this time is sufficiently clerified.

@nvaccessAuto
Copy link
Author

Comment 7 by surveyor (in reply to comment 6) on 2011-12-06 11:05
It still persists even after 4824.
Furthermore, this time reads only one word per line, with "report spelling errors" check box not checked. Reviewing line by line is not still possible for large documents neither.

@nvaccessAuto
Copy link
Author

Comment 8 by mdcurran on 2011-12-06 11:10
I assume though that checking/unchecking the reportSpellingErrors checkbox does not change the behaviour?
So to be clear, was it true that previous winwordTextInproc could read the entire line if reportSpellingErrors was checked?

@nvaccessAuto
Copy link
Author

Comment 9 by surveyor (in reply to comment 8) on 2011-12-06 11:33
Replying to mdcurran:

I assume though that checking/unchecking the reportSpellingErrors checkbox does not change the behaviour?

Actually, it does.

So to be clear, was it true that previous winwordTextInproc could read the entire line if reportSpellingErrors was checked?

Yes, it was. And it's valid for 4824 too.

Changes:
Changed title from "Turkish RTF Doc: Reading in Word will skip line 4, go from 3 to 5." to "NVDA does not read entire line with "report spelling errors" is not checked, winwordTextInproc branch"

@nvaccessAuto
Copy link
Author

Comment 10 by surveyor on 2011-12-07 23:15
I really wonder if it's only me facing this problem.
I hope this piece of log helps.
NVDA version main 4833.

@nvaccessAuto
Copy link
Author

Attachment Nvda_log.txt added by surveyor on 2011-12-07 23:17
Description:
NVDA log after opening a document with MSWord 2010.

@nvaccessAuto
Copy link
Author

Comment 11 by surveyor on 2011-12-12 20:03
Appologies! it's not spesifically related to knotification of speling errors. Please uncheck all options in document formatting dialog to reproduce.

@nvaccessAuto
Copy link
Author

Comment 12 by mdcurran on 2011-12-12 23:50
Fixed in 053cd86. To reproduce the bug all formatting had to be disabled, including lists etc. I have removed the code that was causing this, which was no longer doing what it was written to do, since the addition of many more features in our Word support.
Changes:
Milestone changed from None to 2012.1
State: closed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant