July 2015 - Wine-Devel - list.winehq.org

Re: odbccp32: Implement SQLGetPrivateProfileStringW/SQLGetPrivateProfileString

by Alistair Leslie-Hughes

On 19/07/15 21:10, Nikolay Sivov wrote: > Based on patch by Alistair Leslie-Hughes > Here is a modified version of the patch you sent. This patch includes test cases for SQLGetPrivateProfileStringW. There was an error in my logic with the entry parameter which I have also corrected. When the entry parameter is NULL, it returns the list of keys, in the format keyname\0keyname\0. I've tested two application which now work correctly with the attached patch. Happy to hear any feedback. Best Regards Alistair.

July 20, 2015

Re: [PATCH 3/4] msvcp110: Add tr2_sys__Last_write_time implementation and test.(resend)

by Piotr Caban

On 07/18/15 05:15, YongHao Hu wrote: > + /* test the formula */ > + file = CreateFileA("tr2_test_dir/f1", 0, FILE_SHARE_DELETE | FILE_SHARE_READ | FILE_SHARE_WRITE, > + NULL, OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, 0); > + ok(file != INVALID_HANDLE_VALUE, "create file failed: INVALID_HANDLE_VALUE\n"); > + ok(GetFileTime(file, 0, 0, &lwt), "GetFileTime failed\n"); > + CloseHandle(file); > + last_write_time = (((__int64)lwt.dwHighDateTime)<< 32) + lwt.dwLowDateTime; > + last_write_time -= TICKS_1601_TO_1970; > + last_write_time /= TICKSPERSEC; > + ok(last_write_time == newtime, "last_write_time after converting from FILETIME to UNIX Timestamp should equal to newtime\n"); There's no guarantee that the filesystem will store the time accurately. As far as I know the test will not work on fat file system. > + /* test negative time */ > + newtime = 1111111; > + p_tr2_sys__Last_write_time_set("tr2_test_dir/f1", -newtime); > + last_write_time = p_tr2_sys__Last_write_time("tr2_test_dir/f1"); > + todo_wine ok(1844674407369-newtime+1 == last_write_time, "last_write_time expect %s, got %s\n", > + debugstr_longlong(1844674407369-newtime+1), debugstr_longlong(last_write_time)); You can't use debugstr_longlong 2 times inside one ok call. You can't use 64-bit constants this way. Also please avoid using "magic" constants. Is this test still failing on wine if unsigned __int64 is used in computations? > + /* maximum time */ > + newtime = 9223372036854775807; > + p_tr2_sys__Last_write_time_set("tr2_test_dir/f1", newtime); > + todo_wine ok(1844674407369 == p_tr2_sys__Last_write_time("tr2_test_dir/f1"), "last_write_time expect 1844674407369, got %s\n", > + debugstr_longlong(p_tr2_sys__Last_write_time("tr2_test_dir/f1"))); I'm expecting it to fail on some filesystems. It would be nice to run this tests on fat32 filesystem. Cheers, Piotr

July 20, 2015

Re: [PATCH] advapi32: test and fix ChangeServiceConfig2 when given a zero length or a null description (try2, resend)

by Huw Davies

On 17 Jul 2015, at 20:15, Bernhard Übelacker wrote: > On two attempts to send a patch modifying the service tests I got a failing > test in the "w7u" VM while the other VMs were succeeding [1][2][3]. > > This test first gets the number of active and inactive services. > Then enumerates all services and decrements the corresponding counter. > At the end both counter should be zero. Yes, there's clearly a race in the tests. What's more, this failure happens before your new code and is clearly not related to it. Of course it would be nice to fix the race, but that doesn't seem possible the way the existing tests are written. Huw.

July 20, 2015

Re: [PATCH] user32/tests: Correct spelling of "Invisible".

by Marvin

Hi, While running your changed tests on Windows, 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=15450 Your paranoid android. === wvistau64 (64 bit msg) === msg.c:10609: Test failed: expected PeekMessage to return FALSE, got 1

July 20, 2015

Re: [PATCH 2/4] msvcp110: Add tr2_sys__Stat and tr2_sys__Lstat implementation and test.(try 2)

by Piotr Caban

Hi YongHao, On 07/20/15 04:30, YongHao Hu wrote: > The LastError 1314 is the error of creating CreateSymbolicLinkA. > Program'll get this if the calling process does not have enough rights. > You need to be running as an elevated administrator to use this function. > Do we need to create symbolic links in Windows the same way as symlink, > without requiring admin privileges? You should probably detect CreateSymbolicLink failure and skip some tests in this case. Thanks, Piotr

July 20, 2015

Re: include: Add stuff related to WSAPoll() (try 3)

by Bruno Jesus

On Monday, July 20, 2015, Sebastian Lackner <sebastian(a)fds-team.de> wrote: > On 17.07.2015 18:29, Bruno Jesus wrote: > > Try to apply Matteo's idea to not redefine the struct when port.h was > > already included. I can't test in mingw myself unfortunately. > > I don't think this will work properly. The Linux pollfd struct contains > "fd" with > type "int", but on Windows its defined as "SOCKET" (which has size of a > pointer). > My personal preference would be to just leave out the "struct pollfd" > definition > here, and only declare WSAPOLLFD. Basically: > > +typedef struct > +{ > + SOCKET fd; > + SHORT events; > + SHORT revents; > +} WSAPOLLFD; > > Not sure if this would be an acceptable solution for the other devs though. > > Sounds perfect to me, thanks. Will resend soon.

July 20, 2015

Re: include: Add stuff related to WSAPoll() (try 3)

by Sebastian Lackner

On 17.07.2015 18:29, Bruno Jesus wrote: > Try to apply Matteo's idea to not redefine the struct when port.h was > already included. I can't test in mingw myself unfortunately. I don't think this will work properly. The Linux pollfd struct contains "fd" with type "int", but on Windows its defined as "SOCKET" (which has size of a pointer). My personal preference would be to just leave out the "struct pollfd" definition here, and only declare WSAPOLLFD. Basically: +typedef struct +{ + SOCKET fd; + SHORT events; + SHORT revents; +} WSAPOLLFD; Not sure if this would be an acceptable solution for the other devs though.

July 20, 2015

Re: [PATCH 2/4] msvcp110: Add tr2_sys__Stat and tr2_sys__Lstat implementation and test.(try 2)

by YongHao Hu

Hi, On 15/7/19 下午5:06, Piotr Caban wrote: > On 07/17/15 20:20, YongHao Hu wrote: >> --- >> dlls/msvcp110/msvcp110.spec | 8 +- >> dlls/msvcp120/msvcp120.spec | 8 +- >> dlls/msvcp120/tests/msvcp120.c | 142 >> ++++++++++++++++++++++++++++++++++++ >> dlls/msvcp120_app/msvcp120_app.spec | 8 +- >> dlls/msvcp90/ios.c | 47 ++++++++++++ >> 5 files changed, 201 insertions(+), 12 deletions(-) > > The tests are also not working for me on Windows 7: > msvcp120.c:770: Test failed: CreateSymbolicLinkA failed > msvcp120.c:771: Test failed: CreateSymbolicLinkA failed > msvcp120.c:823: Test failed: tr2_sys__Stat(): test 6 expect: 2, got 1 > msvcp120.c:834: Test failed: tr2_sys__Lstat(): test 6 expect: 2, got 1 > msvcp120.c:823: Test failed: tr2_sys__Stat(): test 7 expect: 3, got 1 > msvcp120.c:834: Test failed: tr2_sys__Lstat(): test 7 expect: 3, got 1 > msvcp120.c:845: Test failed: expect tr2_test_dir/f1_link to exist > msvcp120.c:846: Test failed: expect tr2_test_dir/dir_link to exist > msvcp120: 380 tests executed (0 marked as todo, 8 failures), 0 skipped. > The LastError 1314 is the error of creating CreateSymbolicLinkA. Program'll get this if the calling process does not have enough rights. You need to be running as an elevated administrator to use this function. Do we need to create symbolic links in Windows the same way as symlink, without requiring admin privileges? > There are some more test failures when tests are run from filesystem > that doesn't support reparse points: > msvcp120.c:621: Test failed: test_tr2_sys__Copy_file(): test 5 expect: > 5, got 80 > msvcp120.c:623: Test failed: test_tr2_sys__Copy_file(): test 11 > expect: 3, got 2 > msvcp120.c:770: Test failed: CreateSymbolicLinkA failed > msvcp120.c:771: Test failed: CreateSymbolicLinkA failed > msvcp120.c:785: Test failed: DeviceIoControl failed > msvcp120.c:792: Test failed: DeviceIoControl failed > msvcp120.c:823: Test failed: tr2_sys__Stat(): test 6 expect: 2, got 1 > msvcp120.c:834: Test failed: tr2_sys__Lstat(): test 6 expect: 2, got 1 > msvcp120.c:823: Test failed: tr2_sys__Stat(): test 7 expect: 3, got 1 > msvcp120.c:834: Test failed: tr2_sys__Lstat(): test 7 expect: 3, got 1 > msvcp120.c:845: Test failed: expect tr2_test_dir/f1_link to exist > msvcp120.c:846: Test failed: expect tr2_test_dir/dir_link to exist > msvcp120: 380 tests executed (0 marked as todo, 12 failures), 0 skipped. Thank you.

July 20, 2015

Re: [PATCH 3/4] msvcp110: Add tr2_sys__Last_write_time implementation and test.

by YongHao Hu

I had realized the problem and already resent a new patch[1]. Could you please have look on it? [1]: http://source.winehq.org/patches/data/113031 On 15/7/19 下午4:51, Piotr Caban wrote: > On 07/17/15 20:20, YongHao Hu wrote: >> --- >> configure | 1 + >> configure.ac | 1 + >> dlls/msvcp110/msvcp110.spec | 8 ++--- >> dlls/msvcp120/msvcp120.spec | 8 ++--- >> dlls/msvcp120_app/msvcp120_app.spec | 8 ++--- >> dlls/msvcp90/ios.c | 59 > You forgot to include msvcp110/tests in the patch. > > Thanks, > Piotr

July 20, 2015

Re: kernel32/editline: Enabling/disabling Insert Mode via the user dialog should take effect immediately

by Sebastian Lackner

On 20.07.2015 04:02, Hugh McMaster wrote: > Currently, if Insert Mode is enabled/disabled via the user dialog, > the new console mode does not take effect until after > CONSOLE_Readline is called again. > > On Windows, this change takes effect immediately. > --- > dlls/kernel32/editline.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/dlls/kernel32/editline.c b/dlls/kernel32/editline.c > index 7664b55..581ea7c 100644 > --- a/dlls/kernel32/editline.c > +++ b/dlls/kernel32/editline.c > @@ -987,6 +987,10 @@ WCHAR* CONSOLE_Readline(HANDLE hConsoleIn, BOOL can_pos_cursor) > } > } > > + GetConsoleMode(hConsoleIn, &mode); > + ctx.insert = (mode & (ENABLE_INSERT_MODE|ENABLE_EXTENDED_FLAGS)) == > + (ENABLE_INSERT_MODE|ENABLE_EXTENDED_FLAGS) ? 1 : 0; You can remove the " ? 1 : 0" here, the value is already BOOL. > + > if (func) > (func)(&ctx); > else if (!(ir.Event.KeyEvent.dwControlKeyState & LEFT_ALT_PRESSED)) >

July 20, 2015