On 2014/12/31, Wed 03:55, Stefan Küng wrote;
> On 30.12.2014 17:21, kosuebi22_at_yahoo.co.jp wrote:
>> Date: 2014/12/29, Mon 23:51, Stefan Küng wrote:
>>> On 28.12.2014 14:14, kosuebi22_at_yahoo.co.jp wrote:
>>>> I find a mismatch of encoding of the font name.
>>> Could you maybe send the patch for Scintilla to the Scintilla dev group
>> Thanks for the advice.
>> But I think it is no good to change official Scintilla API.
>> Has a problem in another program's compatibility.
>> Or, It it difficult to add new Scintilla API by me...
>> Alternative plan.
>> I create new patch.
>> I don't change Scintilla.
>> DirectWrite mode set UTF-8.GDI mode set ANSI.
>> I compile/test on Windows 7(jp) and Visual Studio Community 2013.
>> There is limitation to the GDI mode. Font name is only in CP_ACP charctor.
>> (same as until now)
> Since DirectWrite mode is the default now, I think this is the best
me too. best solution.
>> This patch is difficult back port than the previous patch.
>> There is need to change "..\UDiffFontName" and Direct2D
> default to FALSE.
> Don't understand: why do you want to set Direct2D by default to FALSE?
> And what change is required for UDiffFontName?
It is that if you want to back-port to 1.8.x.
r26087 D2D default
r25977 new setting page for the TortoiseUDiff...
> But before we go any further here:
> I think the real question is: why is the localized font name not
> recognized by CreateFontIndirectA ? I mean, the font name is converted
> from utf16 to ansi (local code page). And the local code page should be
> able to correctly handle that font name?
My PC case,
TortoiseSVN fonts setting are list of 53 items.
(1) 30 fonts: English(ASCII) name by Japanese/English system locale.
(2) 21 fonts: Japanese name by Japanese system locale.
and alternative English name by English system locale.
(3) 2 fonts: Japanese name by Japanese/English system locale.
(1) or (2) are OK by CreateFontIndirectA.
(3) converted font name to "????"... at utf16 to ansi(Latin1) NG.
If utf16 to ansi(Japanese/Shift_JIS) OK.
CreateFontIndirectA is not a perfect.
but (3) are miner fonts.
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_tortoisesvn.tigris.org].
Received on 2015-01-04 15:56:25 CET