[PATCH 1/2] gdiplus: Don't test font face of DEFAULT_GUI_FONT.

[Esme]

From: Vincent Povirk <vincent(a)codeweavers.com>

Wine-bug: https://bugs.winehq.org/show_bug.cgi?id=47872
Signed-off-by: Vincent Povirk <vincent(a)codeweavers.com>
---
 dlls/gdiplus/tests/font.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/dlls/gdiplus/tests/font.c b/dlls/gdiplus/tests/font.c
index 9abb80faaf5..7f4b1e876e5 100644
--- a/dlls/gdiplus/tests/font.c
+++ b/dlls/gdiplus/tests/font.c
@@ -808,29 +808,21 @@ static void test_font_metrics(void)
 
 static void test_font_substitution(void)
 {
-    WCHAR ms_shell_dlg[LF_FACESIZE];
+    const WCHAR ms_shell_dlg[] = {'M','S',' ','S','h','e','l','l',' ','D','l','g',0};
     char fallback_font[LF_FACESIZE];
     HDC hdc;
-    HFONT hfont;
     LOGFONTA lf;
     GpStatus status;
     GpGraphics *graphics;
     GpFont *font;
     GpFontFamily *family;
-    int ret;
 
     hdc = CreateCompatibleDC(0);
     status = GdipCreateFromHDC(hdc, &graphics);
     expect(Ok, status);
 
-    hfont = GetStockObject(DEFAULT_GUI_FONT);
-    ok(hfont != 0, "GetStockObject(DEFAULT_GUI_FONT) failed\n");
-
-    memset(&lf, 0xfe, sizeof(lf));
-    ret = GetObjectA(hfont, sizeof(lf), &lf);
-    ok(ret == sizeof(lf), "GetObject failed\n");
-    ok(!lstrcmpA(lf.lfFaceName, "MS Shell Dlg"), "wrong face name %s\n", lf.lfFaceName);
-    MultiByteToWideChar(CP_ACP, 0, lf.lfFaceName, -1, ms_shell_dlg, LF_FACESIZE);
+    memset(&lf, 0, sizeof(lf));
+    lstrcpyA(lf.lfFaceName, "MS Shell Dlg");
 
     status = GdipCreateFontFromLogfontA(hdc, &lf, &font);
     expect(Ok, status);
-- 
2.17.1

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=67784

Your paranoid android.


=== debiant (32 bit Japanese:Japan report) ===

gdiplus:
font.c:832: Test failed: wrong face name Arial
font.c:843: Test failed: wrong face name ™^‹?‹

=== debiant (32 bit Chinese:China report) ===

gdiplus:
font.c:832: Test failed: wrong face name Arial

[Esme (they/them)]

These are existing failures:
https://test.winehq.org/data/3ddf3a720f2a342141550c973f10854b573d80ed/linux_...

I'm not sure why this results in Arial specifically, but that test
machine lacks MS UI Gothic.

[Francois Gouget]

On Sun, 22 Mar 2020, Esme (they/them) wrote:
> These are existing failures:
> https://test.winehq.org/data/3ddf3a720f2a342141550c973f10854b573d80ed/linux_...
>
> I'm not sure why this results in Arial specifically, but that test
> machine lacks MS UI Gothic.

I can reproduce the failure in the ja_JP lcoale on my development 
machine but not on my Debian 10 TestBot machine. But I don't see which 
package could make a difference. They all have ttf-mscorefonts-installer 
but I have not found which package could be providing "MS UI Gothic". Do 
you know which package provides it on you Linux distribution? Or maybe 
the font's filename?


-- 
Francois Gouget <fgouget(a)codeweavers.com>

[Jeff Smith]

On Sun, Mar 22, 2020, 2:46 PM Esme <vincent(a)codeweavers.com> wrote:
> From: Vincent Povirk <vincent(a)codeweavers.com>
>
> Wine-bug: https://bugs.winehq.org/show_bug.cgi?id=47872
> Signed-off-by: Vincent Povirk <vincent(a)codeweavers.com>
> ---
>  dlls/gdiplus/tests/font.c | 14 +++-----------
>  1 file changed, 3 insertions(+), 11 deletions(-)
>
> diff --git a/dlls/gdiplus/tests/font.c b/dlls/gdiplus/tests/font.c
> index 9abb80faaf5..7f4b1e876e5 100644
> --- a/dlls/gdiplus/tests/font.c
> +++ b/dlls/gdiplus/tests/font.c
> @@ -808,29 +808,21 @@ static void test_font_metrics(void)
>
> static void test_font_substitution(void)
>  {
> -    WCHAR ms_shell_dlg[LF_FACESIZE];
> +    const WCHAR ms_shell_dlg[] = {'M','S',' ','S','h','e','l','l','
> ','D','l','g',0};

Especially for tests, is there any reason now to avoid using L””-style
literals for wchar strings?

If not, you could get rid of ms_shell_dlg completely and substitute L"MS
Shell Dlg" for the one remaining place ms_shell_dlg is used.

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=67791

Your paranoid android.


=== build (build log) ===

error: corrupt patch at line 43
Task: Patch failed to apply

=== debiant (build log) ===

Task: Patch failed to apply

=== debiant (build log) ===

Task: Patch failed to apply