[PATCH] kernel32/tests: Georgian and Hindi now get a thread codepage of CP_UTF8.

[Francois Gouget]

As returned by a call to GetCPInfoExA(CP_THREAD_ACP...) in
Windows 10 >=1809.

Signed-off-by: Francois Gouget <fgouget(a)free.fr>
---

The new behavior seems ok to me so I marked the old one as broken and 
added a todo_wine. I can reverse this if the old behavior seems better.

 dlls/kernel32/tests/codepage.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/dlls/kernel32/tests/codepage.c b/dlls/kernel32/tests/codepage.c
index 8fcc575a29e..bb6c09a26d5 100644
--- a/dlls/kernel32/tests/codepage.c
+++ b/dlls/kernel32/tests/codepage.c
@@ -1151,9 +1151,9 @@ static void test_threadcp(void)
         if (lcids[i].threadcp)
             ok(cpi.CodePage == lcids[i].threadcp, "wrong codepage %u for lcid %04x, should be %u\n",
                cpi.CodePage, lcids[i].lcid, lcids[i].threadcp);
-        else
-            ok(cpi.CodePage == acp, "wrong codepage %u for lcid %04x, should be %u\n",
-               cpi.CodePage, lcids[i].lcid, acp);
+        else todo_wine
+             ok(broken(cpi.CodePage == acp) || cpi.CodePage == CP_UTF8, "wrong codepage %u for lcid %04x, should be CP_UTF8\n",
+               cpi.CodePage, lcids[i].lcid);
 
         /* WideCharToMultiByte - CP_THREAD_ACP */
         num = WideCharToMultiByte(CP_THREAD_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL);
-- 
2.20.1

[Alexandre Julliard]

Francois Gouget <fgouget(a)free.fr> writes:
> @@ -1151,9 +1151,9 @@ static void test_threadcp(void)
>          if (lcids[i].threadcp)
>              ok(cpi.CodePage == lcids[i].threadcp, "wrong codepage %u for lcid %04x, should be %u\n",
>                 cpi.CodePage, lcids[i].lcid, lcids[i].threadcp);
> -        else
> -            ok(cpi.CodePage == acp, "wrong codepage %u for lcid %04x, should be %u\n",
> -               cpi.CodePage, lcids[i].lcid, acp);
> +        else todo_wine
> +             ok(broken(cpi.CodePage == acp) || cpi.CodePage == CP_UTF8, "wrong codepage %u for lcid %04x, should be CP_UTF8\n",
> +               cpi.CodePage, lcids[i].lcid);
>
> /* WideCharToMultiByte - CP_THREAD_ACP */
>          num = WideCharToMultiByte(CP_THREAD_ACP, 0, foobarW, -1, NULL, 0, NULL, NULL);

It would be interesting to test if the WideCharToMultiByte conversion is
actually using utf-8.

-- 
Alexandre Julliard
julliard(a)winehq.org