[Scilab-Dev] A few changes in localization and configure options.

[Sylvestre Ledru]

I commited this fix.

S
Le jeudi 26 juin 2008 à 21:33 +0800, Yung-Jang Lee a écrit :
> Slvestre,
>  
> From the email I just send to you, I found a bug  (in the  file
> localetoutf.c )
>  
> #ifndef _MSC_VER   <---------------------bug , 
>       /*  Under Windows only Code page 1252 (iso-8859-1) is subset of
> UNICODE
> ....
>  
> should be 
>  
> #ifdef _MSC_VER <------- change to ifdef 
>       /*  Under Windows only Code page 1252 (iso-8859-1) is subset of
> UNICODE
>  
>  
> This bug always trigger the converter. No harm but not what we want.
>  
>  
> YungLee
>  
> 2008/6/26, Yung-Jang Lee <yjlee123 at gmail.com>: 
>          
>         In previous version I use the converter  only for two-byte
>         character  set (  Chinese,Japanese and Korean). 
>          
>         -----------------------------previous version------------------------------------------    
>                              stricmp("zh_TW", lang)==0 ||
>         stricmp("zh_CN", lang)==0 ||
>         stricmp("ko_KR", lang)==0 || stricmp("ja_JP", lang)==0
>          
>         ----------------------------------------------------------------------------------------------  
>           
>         But I find that under Windows, only code page 1252 (Latin 1
>         languages: Afrikaans, Basque, Catalan, Danish, Dutch, English,
>         Faroese, Finnish, French, Galician, German, Icelandic,
>         Indonesian, Italian, Malay, Norwegian, Portuguese, Spanish,
>         Swahili, Swedish.  ) does not need converter.  So I change it
>         to
>          
>         ----------------------------------current
>         version-----------------------------------------------    
>         #ifndef _MSC_VER
>                // under Windows, it need a converter  if Code Page is
>         not CP1252 (Latin 1 languages)
>                if ( stricmp("CP1252", encoding) !=0 )
>         unicodeSubset=FALSE;
>         #else
>               /*Under linux,  it need a converter for
>         non-unicode encoding */
>               if ( stricmp("utf-8", encoding) !=0 && stricmp("utf8",
>         encoding)!=0 && stricmp("", encoding) !=0 )
>                   unicodeSubset=FALSE;
>         #endif
>         ----------------------------------------------------------------------------------------------
>          
>          
>         I think Russian (CP1251)  for example, also need   a charset
>         converter under Windows and current implementation should work
>         for this case. 
>          
>         Before final release of Scilab 5, it may be better to
>         create necessary localization files
>         ( scilab.mo, localization.jar ...) and let each language
>         translator to test their translations.
>          
>         YungLee
>          
>          
>         
>          
>         2008/6/25, Sylvestre Ledru <sylvestre.ledru at inria.fr>: 
>                 Hi,
>                 
>                 A quick emails to explain the few commits I just did.
>                 
>                 * Localization
>                 Thanks to Yung-Jang Lee who took the time to explain
>                 me how asian
>                 countries are dealing with their charset before
>                 Unicode and to develop
>                 the piece of code for this.
>                 All this stuff has been commited in the revision
>                 25545.
>                 
>                 Just a short question, Yung-Jang, we had this in a
>                 previous version:
>                                          stricmp("zh_TW", lang)==0 ||
>                 stricmp("zh_CN", lang)==0 ||
>                 stricmp("ko_KR", lang)==0 || stricmp("ja_JP", lang)==0
>                 
>                 How are you handling them now ?
>                 
>                 * options in the configure
>                 In order to have some more consistent options (and
>                 because I was unhappy
>                 about the old names), I renamed a few of them.
>                 --with-swig-generation => --enable-build-swig
>                 --with-giws-generation => --enable-build-giws
>                 --with-localization-generation =>
>                 --enable-build-localization
>                 
>                 * help management
>                 I started to work on the build process of the
>                 documentation. There are
>                 too many hardcoded paths & duplicate codes for now.
>                 
>                 Sylvestre
>                 
>                 
>         
>