[PATCH 1/2] kernel32/locale: Set whether we used the default char when using Unix codepage.

[Gabriel Ivăncescu]

Signed-off-by: Gabriel Ivăncescu 
---
 dlls/kernel32/locale.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/dlls/kernel32/locale.c b/dlls/kernel32/locale.c
index b546cb0..2698b3d 100644
--- a/dlls/kernel32/locale.c
+++ b/dlls/kernel32/locale.c
@@ -2726,6 +2726,7 @@ INT WINAPI WideCharToMultiByte( UINT page, DWORD flags, LPCWSTR src, INT srclen,
         {
             ret = wine_cp_wcstombs( unix_cptable, flags, src, srclen, dst, dstlen,
                                     defchar, used ? &used_tmp : NULL );
+            if (used) *used = used_tmp;
             break;
         }
         /* fall through */

[PATCH 2/2] kernel32/locale: Don't fail with non-NULL args when the Unix codepage falls back to UTF-8.

[Gabriel Ivăncescu]

Signed-off-by: Gabriel Ivăncescu 
---

In my opinion, when using CP_UNIXCP, the caller shouldn't care whether it
fell back to UTF-8 or not, and it should not fail if the last two parameters
are non-NULL.

This should simplify caller code that uses CP_UNIXCP with a non-NULL 'used'
parameter.

dlls/kernel32/locale.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/dlls/kernel32/locale.c b/dlls/kernel32/locale.c
index 2698b3d..1770319 100644
--- a/dlls/kernel32/locale.c
+++ b/dlls/kernel32/locale.c
@@ -2729,6 +2729,9 @@ INT WINAPI WideCharToMultiByte( UINT page, DWORD flags, LPCWSTR src, INT srclen,
             if (used) *used = used_tmp;
             break;
         }
+        if (used) *used = FALSE;
+        defchar = NULL;
+        used = NULL;
         /* fall through */
     case CP_UTF8:
         if (defchar || used)

[PATCH 2/2] kernel32/locale: Don't fail with non-NULL args when the Unix codepage falls back to UTF-8.

[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=56234

Your paranoid android.

=== debian10 (32 bit report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (32 bit Chinese:China report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (32 bit WoW report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (64 bit WoW report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

[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=56233

Your paranoid android.

=== debian10 (32 bit report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (32 bit Chinese:China report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (32 bit WoW report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)

=== debian10 (64 bit WoW report) ===

kernel32:
Unhandled exception: page fault on write access to 0x00000000 in 32-bit code (0x00439c82).

Report errors:
kernel32:virtual prints too much data (91050 bytes)