Re: [resend] gdi32/tests: Add tests for GdiGetCodePage.

[Dmitry Timoshkov]

"Huang, Zhangrong"  wrote:
> As discussed in Bug 4065 (http://bugs.winehq.org/show_bug.cgi?id=4065)

In the bug 4065 I suggested to comment out the failing tests before
sending it to wine-patches.

[resend] gdi32/tests: Add tests for GdiGetCodePage.

[Huang, Zhangrong]

Sorry, I didn't get that. These tests will pass on Windows
2000/XP/Vista in different languages except Wine, there are no failing
tests. Should I mark these as todo_wine?

And could we hardcode CJK codepages to fix GdiGetCodePage?
Thanks.

2008/6/24 Dmitry Timoshkov :
> In the bug 4065 I suggested to comment out the failing tests before
> sending it to wine-patches.

[resend] gdi32/tests: Add tests for GdiGetCodePage.

[Dmitry Timoshkov]

"Huang, Zhangrong"  wrote:
> Sorry, I didn't get that. These tests will pass on Windows
> 2000/XP/Vista in different languages except Wine, there are no failing
> tests. Should I mark these as todo_wine?

If todo_wine doesn't mangle the test too much that's of cause more
preferrable. My impression was that with commenting out the failing
test you would avoid a lot of trouble.
> And could we hardcode CJK codepages to fix GdiGetCodePage?

What do you mean?

[resend] gdi32/tests: Add tests for GdiGetCodePage.

[Huang, Zhangrong]

2008/6/24 Dmitry Timoshkov :
> > And could we hardcode CJK codepages to fix GdiGetCodePage?
>
> What do you mean?

I mean fix GdiGetCodePage like:

if (charset == ANSI_CHARSET || charset == DEFAULT_CHARSET)
{
     cp = GetACP();
     if (cp == 936 || cp == 950 || cp == .....)
          return cp;

else (TranslateCharsetInfo(....))
         ....
}

I didn't see any better way.
If we could do so, I'll try to send a patch for GdiGetCodePage with tests.
Is't OK? Thanks.

[resend] gdi32/tests: Add tests for GdiGetCodePage.

[Dmitry Timoshkov]

"Huang, Zhangrong"  wrote:
> I mean fix GdiGetCodePage like:
>
> if (charset == ANSI_CHARSET || charset == DEFAULT_CHARSET)
> {
>     cp = GetACP();
>     if (cp == 936 || cp == 950 || cp == .....)
>          return cp;
>
> else (TranslateCharsetInfo(....))
>         ....
> }
>
> I didn't see any better way.

That definitely will break any charset specified in the font (like
symbol, russian, etc.) in CJK locale. Most likely what happens is
a font linking/substituting for ANSI_CHARSET.

[resend] gdi32/tests: Add tests for GdiGetCodePage.

[Huang, Zhangrong]

2008/6/24 Dmitry Timoshkov :
> That definitely will break any charset specified in the font (like
> symbol, russian, etc.) in CJK locale. Most likely what happens is
> a font linking/substituting for ANSI_CHARSET.

I think it doesn't. If users create font with ANSI_CHARSET or
DEFAULT_CHARSET or any specified charset on CJK locale, they would get
correct font with correct charset, and if it doesn't, it's another
issue.

Now I'm talking about is GdiGetCodePage returns incorrect codepage on
CJK locale, it's not the same with the created font's charset when
using ANSI_CHARSET or DEFAULT_CHARSET. What GdiGetCodePage returns
doesn't break any font charset.

Anyway, I'll remove the failing tests and mark tests as todo_wine.