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

Punctuation / symbol level: most: Some symbols on the right is not voiced (Turkish) #12556

Closed
SeanTolstoyevski opened this issue Jun 17, 2021 · 8 comments

Comments

@SeanTolstoyevski
Copy link

If you are reading a sentence in quotes ("SENTENCES WORDS"), NVDA will ignore the quotation mark on the right.
The problem is that if there is any symbol next to the quotation mark on the right, NVDA will consider the quotation mark on the right. For example "hello."
Now if we remove the fullstop from the example, we will only hear quote once.

A note:

I don't know if this problem is caused by Espeak. NVDA may be sending the wrong data for Turkish to Espeak.
Repeat the same steps for English and everything seems correct.
Since I don't know how to analyze this process more deeply, I wanted to create an issue. I will close this issue with your referrals and recreate it in Espeak-NG repo.
I don't know how to repeat these steps for Espeak. I can ask these questions, but I can also listen to your thoughts and suggestions.

And if this problem is in NVDA codebase I want to submit a pr. Could you please tell me which files I can analyze?
I want to make a small improvement. I love the open source world and NVDA is a great project.

Thanks for reading the note.

Steps to reproduce:

  1. Change Punctuation/symbol level: most

  2. Change Synthesizer: EspeakNG

  3. Change synthesizer language: Turkish

  4. Typing
    "Merhaba dostum" dedi.
    Or coding example:
    functionCallProduce("a string")

  5. Scroll through the text with the up and down arrows.

Actual behavior:

In both examples, you will hear the quotation mark only 1.

So:
tırnak Merhaba dostum dedi.

Or:
functionCallProduce açparantez tırnak a string kapaparantez

Expected behavior:

The following output should be voiced:
tırnak Merhaba dostum tırnak dedi.

Code example:
functionCallProduce açparantez tırnak a string tırnak kapaparantez

System configuration

NVDA installed/portable/running from source:

Tested installed, portable

NVDA version:

2020.4

Windows version:

Windows 10, 64 bit

Name and version of other software in use when reproducing the issue:

Other information about your system:

Other questions

Does the issue still occur after restarting your computer?

Yes

Have you tried any other versions of NVDA? If so, please report their behaviors.

If add-ons are disabled, is your problem still occurring?

yes

Does the issue still occur after you run the COM Registration Fixing Tool in NVDA's tools menu?

@seanbudd
Copy link
Member

seanbudd commented Jun 17, 2021

Thanks @SeanTolstoyevski ,

Thanks for the thorough investigation. I believe this is a duplicate of #11779, which should be fixed by #11856. Your enthusiasm for fixing this is highly appreciated, feel free to have a review of that PR and issue.

Cheers

@SeanTolstoyevski
Copy link
Author

Hi @seanbudd ,

I downloaded the build as you said in pr and tested it with the same configuration.
It doesn't fix what i'm talking about.

@seanbudd seanbudd reopened this Jun 18, 2021
@CyrilleB79
Copy link
Collaborator

Hello
Seems a problem in Turkish translation.
In the Turkish symbol file, we have the following line:
in-word ' (?<=[^\W_])['’'"'”'-]
In the English symbol file we have:
in-word ' (?<=[^\W_])['’]
The English line should be used.
The following characters should no tappear in the regexp: " ” -

You may want to contact Turkish translators for this. I think it is @Surveyor123

@Surveyor123
Copy link
Contributor

I've just Updated symbols.dic and Committed. Will be fixed in NVDA2021.1. Thanks

@CyrilleB79
Copy link
Collaborator

@SeanTolstoyevski and @Surveyor123, have you been able to check with the new symbol file that the issue is fixed?

If it is confirmed that this issue aws due to symbol file, this issue should be closed. Thanks.

@Surveyor123
Copy link
Contributor

Surveyor123 commented Jun 18, 2021 via email

@SeanTolstoyevski
Copy link
Author

Can you share a build with the latest commits active?
As the owner of the issue, I want to confirm this.

@CyrilleB79
Copy link
Collaborator

As explained, translation/localization are not made in this Git repository but in another SVN repository. That's why you were asked to close this issue and to contact translators to make the modifications.

The modifications are not yet in this repo and there is no automatic build containing them. You will be able to see them when a new beta (if any) is released or in NVDA 2021.1rc1 build.

If you want to test manually without expecting these builds:

  1. Create a portable version of NVDA 2021.1beta3 (if you do not have already)
  2. Download symbols.dic.txt and rename it without the txt extension. That's the last version of the symbol file modified by @Surveyor123
  3. Go to YourPortableNVDAFolder\locale\tr
  4. Paste there the file you have downloaded and renamed. You will need to accept to overwrite the existing file.
  5. Start your portable copy of NVDA and test

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

No branches or pull requests

4 participants