Re: [Audacity-translation] A problem about language name in Preferences

[trebmuh]

Le 2017-11-29 10:10, James Crook a écrit :

On 11/29/2017 1:09 AM, Paul Licameli wrote:
I thought it over and now it think it would be easier to maintain a 
small
text file of all the language self-names in legible Unicode text, then 
just
have the program read it at run-time, rather than maintain less 
readable

source code.

Then it would not be too hard to research all the correct names and 
cut and

paste them.

Not a large project, but not small enough for this limitied release.

PRL


A detail...

Slightly better might be if Audacity.pot contained the English names
of the languages.

German, French, Dutch   etc.
Translators would then translate these.


This sounds like a good plan.


We would also add the few miscellaneous strings that come from the
'desktop' file and the Audacity setup program that we currently put in
the right place by hand.


Sounds good.


Then we would run a script  to automatically
create a suitable source code file, just as we already run a script to
create the 'git identity' source code file (that is a file the lets
users know exactly what version of the source code was used).

There are two advantages of this:

1) Translators are each maintaining a single .po file.
2) Audacity does not have to load an extra file at run time.

Anyway I agree with Paul, NOT for 2.2.1.


+1

Olivier

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Audacity-translation mailing list
Audacity-translation@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-translation

Re: [Audacity-translation] A problem about language name in Preferences

[James Crook]

On 11/29/2017 1:09 AM, Paul Licameli wrote:

I thought it over and now it think it would be easier to maintain a small
text file of all the language self-names in legible Unicode text, then just
have the program read it at run-time, rather than maintain less readable
source code.

Then it would not be too hard to research all the correct names and cut and
paste them.

Not a large project, but not small enough for this limitied release.

PRL


A detail...

Slightly better might be if Audacity.pot contained the English names of 
the languages.


German, French, Dutch   etc.
Translators would then translate these.

We would also add the few miscellaneous strings that come from the 
'desktop' file and the Audacity setup program that we currently put in 
the right place by hand.  Then we would run a script  to automatically 
create a suitable source code file, just as we already run a script to 
create the 'git identity' source code file (that is a file the lets 
users know exactly what version of the source code was used).


There are two advantages of this:

1) Translators are each maintaining a single .po file.
2) Audacity does not have to load an extra file at run time.

Anyway I agree with Paul, NOT for 2.2.1.

--James.




--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Audacity-translation mailing list
Audacity-translation@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-translation

Re: [Audacity-translation] A problem about language name in Preferences

[Paul Licameli]

I thought it over and now it think it would be easier to maintain a small
text file of all the language self-names in legible Unicode text, then just
have the program read it at run-time, rather than maintain less readable
source code.

Then it would not be too hard to research all the correct names and cut and
paste them.

Not a large project, but not small enough for this limitied release.

PRL


On Tue, Nov 28, 2017 at 5:41 PM, Paul Licameli 
wrote:

> Look at the file src/Languages.cpp and you will see the table where these
> language names are defined as strings in C++ code.  The names are not
> defined in any .po files, nor can I find where they are named in wxWidgets
> sources.  Is there something in `man locale` that tells you how to retrieve
> the language's self-name in Unicode with some C library call?  I don't know
> of it.
>
> So, correcting this table of names may be difficult.  For instance:
>
>localLanguageName[wxT("fr")] = wxT("Francais");
>
> To make the proper c with cedilla, as a C++ wide character string literal,
> looks like this:
>
>localLanguageName[wxT("fr")] = wxT("Fran\u00e7ais");
>
> I just tried this and it makes the dialog display better, but the code is
> harder to read, and probably not easy to write either.
>
> I don't have the time now to figure out the correct Unicode escapes for
> all of them.
>
> But maybe one of you has the dedication to do that, and we can get it into
> version 2.2.2.
>
> I would rather fix it all at once, then only partly.  So someone should
> also research all the languages in the list without any active workers in
> this group now.
>
>
> On Tue, Nov 28, 2017 at 5:30 PM, Paul Licameli 
> wrote:
>
>> Look at the file src/Languages.cpp and you will see the table where these
>> language names are defined as strings in C++ code.  The names are not
>> defined in any .po files, nor can I find where they are named in wxWidgets
>> sources.  Is there something in `man locale
>>
>> On Tue, Nov 28, 2017 at 8:34 AM,  wrote:
>>
>>> Le 2017-11-26 13:12, mkpoli a écrit :
>>>
 I recently noticed that language JA is using the romanized
 autoglossonym "Nihongo" in the combo box, rather than the English name
 for it "Japanese". I also found "Suomi", "Svenska" ,"Russky",
 "Francais" and "Ellinika".

>>>
>>> I can confirm for French here.
>>>
>>> The are all anglicized romanization of
 autoglossonym of that language. I think it is not clear at all
 especially for languages not using Latin script (even worse if using
 non-phonemic writing system, e.g. at least Ellinika can (nearly)
 one-to-one correspond to Ελληνικά), and it is also not
 consistent since Chinese, Korean and some other languages are in
 English name.

 It would probably cause confusion in finding the desired language.

 I disagree using English name for languages, it is also not that easy.
 I think at least language combo box should contain autoglossonym.

 English name + autoglossonym (Preferable):
 Chinese (Simplified)  简体中文
 Chinese (Traditional) 繁體中文
 Japanese  日本語
 Finnish   Suomi

>>>
>>> + 1 for this suggestion since it allows language readers + English
>>> readers to do understand it.
>>
>>
>
> I think just the locale abbreviation and the self-name of the language,
> not the English name.
>
> After all, if I show the English name for French, when I am in English
> locale... should I show the French name for English if I'm in French locale
> but switching out?
>
> No, I don't want to bother getting n^2 strings for n languages!
>
> Just listing each language's self name is what I am used to seeing in
> Wikipedia page sidebars, so it should be all right here too.
>
>
>>
>>>
>>> autoglossonym only (Preferable):
  简体中文
 繁体中文
 日本语
 Suomi

>>>
>>> +0.5 : since it doesn't allow me (as an example) to understand that
>>> Chinese is supported since I don't read Chinese.
>>>
>>> English name only (Not good):
 Chinese (Simplified)
 Chinese (Traditional)
 Japanese
 Finnish

>>>
>>> -1 since it *only* allow English readers to understand it (my
>>> grand-father is out of the run here)
>>>
>>> current (Worse):

 Chinese (Simplified)
 Chinese (Traditional)
 Nihongo
 Suomi

>>>
>>> -2 : really bad
>>>
>>> Another possibility which would requires some work code-wise and
>>> translation-wise would be to have them all translated to every language and
>>> then display with  (example with French locales) :
>>>
>>
>>> Localised name + autoglossonym only + English name:
 Chinois (simplifié)   简体中文  Chinese (Simplified)
 Chinois (traditionnel)繁體中文  Chinese (Traditional)
 Japonais  日本語Japanese
 Finois   

Re: [Audacity-translation] A problem about language name in Preferences

[Paul Licameli]

Look at the file src/Languages.cpp and you will see the table where these
language names are defined as strings in C++ code.  The names are not
defined in any .po files, nor can I find where they are named in wxWidgets
sources.  Is there something in `man locale` that tells you how to retrieve
the language's self-name in Unicode with some C library call?  I don't know
of it.

So, correcting this table of names may be difficult.  For instance:

   localLanguageName[wxT("fr")] = wxT("Francais");

To make the proper c with cedilla, as a C++ wide character string literal,
looks like this:

   localLanguageName[wxT("fr")] = wxT("Fran\u00e7ais");

I just tried this and it makes the dialog display better, but the code is
harder to read, and probably not easy to write either.

I don't have the time now to figure out the correct Unicode escapes for all
of them.

But maybe one of you has the dedication to do that, and we can get it into
version 2.2.2.

I would rather fix it all at once, then only partly.  So someone should
also research all the languages in the list without any active workers in
this group now.


On Tue, Nov 28, 2017 at 5:30 PM, Paul Licameli 
wrote:

> Look at the file src/Languages.cpp and you will see the table where these
> language names are defined as strings in C++ code.  The names are not
> defined in any .po files, nor can I find where they are named in wxWidgets
> sources.  Is there something in `man locale
>
> On Tue, Nov 28, 2017 at 8:34 AM,  wrote:
>
>> Le 2017-11-26 13:12, mkpoli a écrit :
>>
>>> I recently noticed that language JA is using the romanized
>>> autoglossonym "Nihongo" in the combo box, rather than the English name
>>> for it "Japanese". I also found "Suomi", "Svenska" ,"Russky",
>>> "Francais" and "Ellinika".
>>>
>>
>> I can confirm for French here.
>>
>> The are all anglicized romanization of
>>> autoglossonym of that language. I think it is not clear at all
>>> especially for languages not using Latin script (even worse if using
>>> non-phonemic writing system, e.g. at least Ellinika can (nearly)
>>> one-to-one correspond to Ελληνικά), and it is also not
>>> consistent since Chinese, Korean and some other languages are in
>>> English name.
>>>
>>> It would probably cause confusion in finding the desired language.
>>>
>>> I disagree using English name for languages, it is also not that easy.
>>> I think at least language combo box should contain autoglossonym.
>>>
>>> English name + autoglossonym (Preferable):
>>> Chinese (Simplified)  简体中文
>>> Chinese (Traditional) 繁體中文
>>> Japanese  日本語
>>> Finnish   Suomi
>>>
>>
>> + 1 for this suggestion since it allows language readers + English
>> readers to do understand it.
>
>

I think just the locale abbreviation and the self-name of the language, not
the English name.

After all, if I show the English name for French, when I am in English
locale... should I show the French name for English if I'm in French locale
but switching out?

No, I don't want to bother getting n^2 strings for n languages!

Just listing each language's self name is what I am used to seeing in
Wikipedia page sidebars, so it should be all right here too.


>
>>
>> autoglossonym only (Preferable):
>>>  简体中文
>>> 繁体中文
>>> 日本语
>>> Suomi
>>>
>>
>> +0.5 : since it doesn't allow me (as an example) to understand that
>> Chinese is supported since I don't read Chinese.
>>
>> English name only (Not good):
>>> Chinese (Simplified)
>>> Chinese (Traditional)
>>> Japanese
>>> Finnish
>>>
>>
>> -1 since it *only* allow English readers to understand it (my
>> grand-father is out of the run here)
>>
>> current (Worse):
>>>
>>> Chinese (Simplified)
>>> Chinese (Traditional)
>>> Nihongo
>>> Suomi
>>>
>>
>> -2 : really bad
>>
>> Another possibility which would requires some work code-wise and
>> translation-wise would be to have them all translated to every language and
>> then display with  (example with French locales) :
>>
>
>> Localised name + autoglossonym only + English name:
>>> Chinois (simplifié)   简体中文  Chinese (Simplified)
>>> Chinois (traditionnel)繁體中文  Chinese (Traditional)
>>> Japonais  日本語Japanese
>>> FinoisSuomiFinnish
>>>
>>
>>
More trouble that it is worth, I think.

PRL



>

> Hope that helps
>> Olivier
>>
>> 
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> ___
>> Audacity-translation mailing list
>> Audacity-translation@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/audacity-translation
>>
>
>
--
Check out the vibrant 

Re: [Audacity-translation] A problem about language name in Preferences

[Paul Licameli]

Look at the file src/Languages.cpp and you will see the table where these
language names are defined as strings in C++ code.  The names are not
defined in any .po files, nor can I find where they are named in wxWidgets
sources.  Is there something in `man locale

On Tue, Nov 28, 2017 at 8:34 AM,  wrote:

> Le 2017-11-26 13:12, mkpoli a écrit :
>
>> I recently noticed that language JA is using the romanized
>> autoglossonym "Nihongo" in the combo box, rather than the English name
>> for it "Japanese". I also found "Suomi", "Svenska" ,"Russky",
>> "Francais" and "Ellinika".
>>
>
> I can confirm for French here.
>
> The are all anglicized romanization of
>> autoglossonym of that language. I think it is not clear at all
>> especially for languages not using Latin script (even worse if using
>> non-phonemic writing system, e.g. at least Ellinika can (nearly)
>> one-to-one correspond to Ελληνικά), and it is also not
>> consistent since Chinese, Korean and some other languages are in
>> English name.
>>
>> It would probably cause confusion in finding the desired language.
>>
>> I disagree using English name for languages, it is also not that easy.
>> I think at least language combo box should contain autoglossonym.
>>
>> English name + autoglossonym (Preferable):
>> Chinese (Simplified)  简体中文
>> Chinese (Traditional) 繁體中文
>> Japanese  日本語
>> Finnish   Suomi
>>
>
> + 1 for this suggestion since it allows language readers + English readers
> to do understand it.
>
> autoglossonym only (Preferable):
>>  简体中文
>> 繁体中文
>> 日本语
>> Suomi
>>
>
> +0.5 : since it doesn't allow me (as an example) to understand that
> Chinese is supported since I don't read Chinese.
>
> English name only (Not good):
>> Chinese (Simplified)
>> Chinese (Traditional)
>> Japanese
>> Finnish
>>
>
> -1 since it *only* allow English readers to understand it (my grand-father
> is out of the run here)
>
> current (Worse):
>>
>> Chinese (Simplified)
>> Chinese (Traditional)
>> Nihongo
>> Suomi
>>
>
> -2 : really bad
>
> Another possibility which would requires some work code-wise and
> translation-wise would be to have them all translated to every language and
> then display with  (example with French locales) :
>
> Localised name + autoglossonym only + English name:
>> Chinois (simplifié)   简体中文  Chinese (Simplified)
>> Chinois (traditionnel)繁體中文  Chinese (Traditional)
>> Japonais  日本語Japanese
>> FinoisSuomiFinnish
>>
>
> Hope that helps
> Olivier
>
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> Audacity-translation mailing list
> Audacity-translation@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/audacity-translation
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Audacity-translation mailing list
Audacity-translation@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-translation

Re: [Audacity-translation] A problem about language name in Preferences

[trebmuh]

Le 2017-11-26 13:12, mkpoli a écrit :

I recently noticed that language JA is using the romanized
autoglossonym "Nihongo" in the combo box, rather than the English name
for it "Japanese". I also found "Suomi", "Svenska" ,"Russky",
"Francais" and "Ellinika".


I can confirm for French here.


The are all anglicized romanization of
autoglossonym of that language. I think it is not clear at all
especially for languages not using Latin script (even worse if using
non-phonemic writing system, e.g. at least Ellinika can (nearly)
one-to-one correspond to Ελληνικά), and it is also not
consistent since Chinese, Korean and some other languages are in
English name.

It would probably cause confusion in finding the desired language.

I disagree using English name for languages, it is also not that easy.
I think at least language combo box should contain autoglossonym.

English name + autoglossonym (Preferable):
Chinese (Simplified)  简体中文
Chinese (Traditional) 繁體中文
Japanese  日本語
Finnish   Suomi


+ 1 for this suggestion since it allows language readers + English 
readers to do understand it.



autoglossonym only (Preferable):
 简体中文
繁体中文
日本语
Suomi


+0.5 : since it doesn't allow me (as an example) to understand that 
Chinese is supported since I don't read Chinese.



English name only (Not good):
Chinese (Simplified)
Chinese (Traditional)
Japanese
Finnish


-1 since it *only* allow English readers to understand it (my 
grand-father is out of the run here)



current (Worse):

Chinese (Simplified)
Chinese (Traditional)
Nihongo
Suomi


-2 : really bad

Another possibility which would requires some work code-wise and 
translation-wise would be to have them all translated to every language 
and then display with  (example with French locales) :



Localised name + autoglossonym only + English name:
Chinois (simplifié)   简体中文  Chinese (Simplified)
Chinois (traditionnel)繁體中文  Chinese (Traditional)
Japonais  日本語Japanese
FinoisSuomiFinnish


Hope that helps
Olivier

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Audacity-translation mailing list
Audacity-translation@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-translation

[Audacity-translation] A problem about language name in Preferences

[mkpoli]

I recently noticed that language *ja *is using the romanized autoglossonym
"Nihongo" in the combo box, rather than the English name for it "Japanese".
I also found "Suomi", "Svenska" ,"Russky", "Francais" and "Ellinika". The
are all anglicized romanization of autoglossonym of that language. I think
it is not clear at all especially for languages not using Latin script
(even worse if using non-phonemic writing system, e.g. at least Ellinika
can (nearly) one-to-one correspond to Ελληνικά), and it is also not
consistent since Chinese, Korean and some other languages are in English
name.

It would probably cause confusion in finding the desired language.

I disagree using English name for languages, it is also not that easy. I
think at least language combo box should contain autoglossonym.

English name + autoglossonym (Preferable):
Chinese (Simplified)  简体中文
Chinese (Traditional) 繁體中文
Japanese  日本語
Finnish   Suomi
autoglossonym only (Preferable):
 简体中文
繁体中文
日本语
Suomi
English name only (Not good):
Chinese (Simplified)
Chinese (Traditional)
Japanese
Finnish
current (Worse):
Chinese (Simplified)
Chinese (Traditional)
Nihongo
Suomi
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Audacity-translation mailing list
Audacity-translation@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-translation