On 21/07/09 03:45, Takao Fujiwara wrote:
> (07/20/09 20:49), Tony Mechelynck-san wrote:
>> Won't Vim +iconv accept any charset known by your installed version of
>> iconv, even charsets not compiled into Vim? You can list them (at least
>
> I'm not sure if I understand 'Vim +iconv' exactly however 'vim
+iconv is a shortcut for "compiled with the +iconv feature".
> --version' shows '+iconv' in my env.
>
> % env vim --version
> VIM - Vi IMproved 7.2 (2008 Aug 9, compiled Jul 8 2009 16:34:32)
> Included patches: 1-148
> Huge version without GUI. Features included (+) or not (-):
> ... +iconv ...
Then you've got it; however, on Windows "+iconv/dyn" is more frequent,
and then whether you have it is determined at run-time.
:echo has('iconv')
gives a nonzero answer (usually 1) if it is present _this time_ then.
>
> The problem is:
> 1. Invoke vim windows-31j-text.txt
> 2. Invoke ':set fileencoding=windows-31j' on vim screen.
>
> Then vim cannot show the WINDOWS-31J text.
> After I applied my patch, vim can show the contents correctly.
>
>> if you have a posix-compliant iconv executable in addition to the iconv
>> shared library) by means of "iconv -l" (pipe the output into more or
>> less since there are a lot of them) or maybe, in Vim, ":new | 0r !iconv
>> -l".
>
> Yes, 'iconv -l' shows the correct encodings.
> However vim has the hard-coded charset table in itself as I attached the
> patch.
>
>>
>> Most of the charsets you want are already accepted by my iconv version;
>> the exceptions are two Sun encodings (with no equivalents that I can
>> see) and cswindows31j for cswindows-31j, ms_kanji for ms-kanji.
>>
>> I'm appending a small file containing a paragraph from jp.wikipedia.org
>> which I saved in "cseucpkdfmtjapanese" encoding in gvim 7.2.234 after
>> removing HTML formatting. I suppose you could do
>>
>> :view ++enc=cseucpkdfmtjapanese pkdfmt.txt
>
> When I applied my patch, this line can shows the charset correctly in my
> env.
> I cannot show the content of pkdfmt.txt correctly without my patch.
Mine could, obviously, and without any patching.
>
>>
>> in gvim to see if it is proper Japanese. In any case it is not UTF-8,
>> and it looks like gibberish in Latin1, but the above ex-command gives me
>> what "looks" to me (the illiterate Westerner) as Japanese writing.
>>
>>
>> Personally, I would rather trust iconv than your statement that, let's
>> say, "shift-jisx0213 is not cp932 exactly but making it an alias would
>> help".
>
> My iconv is 2.10.1
> % iconv --versoin
> iconv (GNU libc) 2.10.1
>
> My understanding is vim has the table enc_canon_table[] in src/mbyte.c .
> If we add a new codepage DBCS_JPN_SHIFT_JIS0213 in the table, we will
> need the furthermore investigations.
> I mean the suggestion of my patch can work with a lot of code points at
> the moment without changing more source codes.
>
> Thanks,
> fujiwara
>
>>
>> FWIW, this is what I get in answer to "iconv --version":
>>> iconv (GNU libc) 2.9
>>> Copyright (C) 2008 Free Software Foundation, Inc.
>>> This is free software; see the source for copying conditions. There
>>> is NO
>>> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
>>> PURPOSE.
>>> Written by Ulrich Drepper.
>>
>>
>> Best regards,
>> Tony.
>
>
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
