On 20/07/09 21:09, Bram Moolenaar wrote:
>
>
> Takao Fujiwara wrote:
>
>> (07/19/09 05:25), Bram Moolenaar-san wrote:
>>> Can you please explain for each entry why it should be added?
>>
>> + {"ms932", IDX_CP932},
>>
>> MS932 is added in glibc/iconvdata/gconv-modules
>>
>> + {"windows-31j", IDX_CP932},
>>
>> WINDOWS-31J is added in glibc/iconvdata/gconv-modules
>> WINDOWS-31J is also registered in IANA.
>>
>> + {"cswindows-31j", IDX_CP932},
>>
>> CWINDOWS-31J is added in glibc/iconvdata/gconv-modules
>>
>> + {"shift-jisx0213", IDX_CP932},
>>
>> SHIFT_JISX0213 is added in glibc/iconvdata/shift_jisx0213.c
>>
>> + {"sjis-win", IDX_CP932},
>>
>> SJIS_WIN is added in glibc/iconvdata/gconv-modules
>>
>> + {"sjis-open", IDX_CP932},
>>
>> SJIS_OPEN is added in glibc/iconvdata/gconv-modules
>>
>> + {"pck", IDX_CP932},
>>
>> PCK is a Solaris encoding.
>>
>> + {"euc-jisx0213", IDX_EUC_JP},
>>
>> EUC_JISX0213 is added in glibc/iconvdata/euc-jisx0213.c
>>
>> + {"cseucpkdfmtjapanese", IDX_EUC_JP},
>>
>> CSEUCPKDFMTJAPANESE is added in glibc/iconvdata/gconv-modules
>>
>> + {"osf00030010", IDX_EUC_JP},
>>
>> OSF00030010 is added in glibc/iconvdata/gconv-modules
>>
>> + {"euc-jp-ms", IDX_EUC_JP},
>>
>> EUC_JP_MS is added in glibc/iconvdata/euc-jp-ms.c
>>
>> + {"eucjp-ms", IDX_EUC_JP},
>>
>> EUCJP-MS is added in glibc/iconvdata/gconv-modules
>>
>> + {"eucjp-open", IDX_EUC_JP},
>>
>> EUCJP-OPEN is added in glibc/iconvdata/gconv-modules
>>
>> + {"eucjp-win", IDX_EUC_JP},
>>
>> EUCJP-WIN is added in glibc/iconvdata/gconv-modules
>>
>> + {"csshiftjis", IDX_SJIS},
>>
>> CSSHIFTJIS is added in glibc/iconvdata/gconv-modules
>>
>> + {"ms-kanji", IDX_SJIS},
>>
>> MS-KANJI is added in glibc/iconvdata/gconv-modules
>>
>> + {"ansi-1251", IDX_CP1251},
>>
>> ANSI-1251 is a Solaris encoding.
>>
>> Actually vim doesn't have some charsets, e.g. "shift-jisx0213" is not
>> IDX_CP932 exactly. However adding the similar alias names would help
>> users instead of falling back to a ISO.
I'm wary of this kind of approximations
>> I would think the additional charsets might not effect the performance
>> of vim so much. So if the suggested charsets are upstreamed, it would
>> be nice.
>> I'll provide the patched binaries to users as they request the feature.
>
> I'm not worried about performance. I do want to avoid that we add
> things in here that might cause a problem for someone.
>
> I invite everybody to make comments. Anything that looks like it might
> be causing problems?
>
> What is glibc/iconvdata/gconv-modules? Isn't this an implementation
> instead of a specification of encodings?
>
I'd rather we rely on iconv, which knows most of these encodings
already. AFAICT, Vim (with +iconv, of course) can use any 'fileencoding'
known to iconv (and convert it to and from UTF-8, or via UTF-8 to and
from each other, the file's data permitting).
Best regards,
Tony.
--
Support the right of unborn males to bear arms!
-- A public service announcement from Phyllis Schlafly,
the Catholic Church, and the National Rifle
Association
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
