Heidisql: Incomplete Chinese translation with auto-detected system language

Created on 25 Jan 2019  ·  30Comments  ·  Source: HeidiSQL/HeidiSQL

Current behavior

The default state, does not specify the application language for simplified Chinese, that is: automatic detection

Expected behavior

When the system default language is detected to be simplified Chinese, the entire application USES simplified Chinese and does not need to be specified as: simplified Chinese

Environment

Application language: Auto detect

  • HeidiSQL version:
    10.0 Revision 5461
  • Database system and version:
  • Operating system:
    Win 10 Chinese
bug

All 30 comments

You will need to explain again. I don't get if this is a bug. If your system default language is simplified Chinese, then why shouldn't HeidiSQL use exactly that language as its default?

tim 20190125153445
tim 20190125153534
tim 20190125153659

When set to auto detection, only a few are displayed in Chinese.

When set to Chinese, all are displayed as Chinese, indicating that the language package is complete.

We have simplified and traditional Chinese in the language selector. Could it be that one of them is incomplete and you manually selected the other one which is translated completely? Please please test by selecting the other Chinese option.

1
2

Traditional Chinese is also complete.

The problem is that the previous version 9.5.0 does not exist, and it will be automatically upgraded. It started from one version, but I can't remember which version it is. All the way to 10.0. The question remains.

I suppose a good idea for debugging is to find out what the auto-detection does in the back. Will try to add the detected language in the dropdown item.

So the language drop down in the preferences dialog now shows the current locale abbreviation. For me it is "de_DE". Please update to the next build and post what you see there.

grafik

1

Auto detect (zh_CN)

I just updated the translation unit from https://sourceforge.net/p/dxgettext/ , which should fix some potentially important missing compiler directives. Please test again with the next build.

The defect is still in 5499

image

Oh you tested exactly one build too early :) Sorry for the confusion. Please update again now, the relevant build is now downloadable.

image

In 5500

I just found that, besides zh_CN and zh_TW, we have a third language: "zh". I'm not sure why it's not listed in the language dropdown, but the translation file is there, and it is more complete than zh_CN:

grafik

On Transifex, zh_CN is missing yet 1000 translations:
grafik

To me both zh and zh_CN look different, so zh_CN may be relevant, although it's just incomplete. In that case the translation should just be completed by anyone (feel free :). The other issue is to add zh_CN to the language selector in HeidiSQL's preferences dialog - this is just missing.

Same here. Is there any possibility to redirect "zh-CN" to "zh"? Because it is meaningless as the different between "en" and "en-US".

That's what I asked for, as I'm not aware of Chinese writing. If zh_CN is highly incomplete, and if it's the same language as zh which is translated completely, then I can remove zh_CN from Transifex. Or maybe there's a way to disable it.

I just removed zh_CN, after making a local backup. So the next build should auto-detect zh instead of zh_CN which more exact but no longer there.

@ansgarbecker I download the latest version which is 10.2.0.5999 but the system default language is still detected to be zh_CN instead of zh
2019-07-08_14-30-18

Thanks for your feedback. I had the impression this is fixed as there was no more feedback until now.

@ansgarbecker Hi, Ansgar Becker, zh: Chinese in general. It includes simplified Chinese and traditional Chinese in various regions. Perhaps because the language pack comes from contributors in multiple parts of the world, it is more complete. Before consulting my proposal, please refer to the following information:

1.
Locale | Description
zh-CN | Chinese (Simplified, PRC)
zh-SG | Chinese (Simplified, Singapore)
zh-TW | Chinese (Traditional, Taiwan)
zh-HK | Chinese (Traditional, Hong Kong S.A.R.)
zh-MO | Chinese (Traditional, Macao S.A.R.)

2.
Traditional Chinese, zh-Hant, it includes: zh-TW、zh-HK、zh-MO
Simplified Chinese, zh-Hans, it includes: zh-CN、zh-SG

3.
Reference: http://www.iana.org/assignments/language-subtag-registry/language-subtag-registry

%%
Type: language
Subtag: zh
Description: Chinese
Added: 2005-10-16
Scope: macrolanguage
%%
%%
Type: redundant
Tag: zh-Hant
Description: traditional Chinese
Added: 2003-05-30
%%
%%
Type: redundant
Tag: zh-Hans
Description: simplified Chinese
Added: 2003-05-30
%%
%%
Type: region
Subtag: CN
Description: China
Added: 2005-10-16
%%

zh-CN, which includes all Chinese in mainland China, such as big dialects, small dialects, traditional Chinese, simplified Chinese, not just simplified Chinese.

=================================
Based on the above information, I suggest making the following changes to the language options without emphasizing specific regions:

  1. zh: Chinese(Simplified) modified to zh: Chinese(China)
  2. zh_CN: Chinese(China) modified to zh-Hans: Chinese(Simplified)
  3. zh_TW: Chinese(Traditional) modified to zh-Hant: Chinese(Traditional)

For the problem of automatic detection of location language package, since zh is more complete, all Chinese is located to it by default. Zh_CN does not need to be deleted, and it may be complete in the near future.

I think we should remove those unmaintained languages (like zh_CN) first.

zh_CN was previously recreated at some point in the recent weeks. And also, my build process did not delete translation files if they were left out by the download process. They just were not updated, and the old content got baken into heidisql.exe. Now I', firing a rmdir /Q /S out\locale to get rid of all translation files, and then pull fresh copies with tx.py35.exe pull -a -s.

Current build does not have zh_CN any longer.

@ansgarbecker Good ,However, there are still some improvements to be made.
image

You mean, the auto-detect still says "zh_CN", or ...?

Is that screen taken from a new build?

Yes, it's the latest version, just a display problem.

in 5621

If zh_CN is still there, then this is not just a display problem. Or am I misunderstanding you?

The auto-detected language package is now zh and zh_CN is no longer in the language option. It's just that there's still a zh_CN after the automatic detection.

image

Oh I see. The gettext unit detects this language code, while I can't fix it to the one which is offered as a language selection.

Maybe it doesn't show it, after "Auto detect", like before I asked this issue, it just display "Auto detect".

Zh_CN is shown, but it is not in the option, which is not very friendly.

Was this page helpful?
0 / 5 - 0 ratings