Hi,
I was looking for a way to kill some time when I ended up on (WineHQ's version of) Dimi's janitorial page. It seems the page could use an update since as of this moment some of the items seem to have been finished, or nearly so:
tools/winapi_check/winapi_check --none --progress --cross-call-win32-win16
[...] loader/module.c: kernel32: WIN32_GetProcAddress16: illegal call to GetProcAddress16 (Win32 -> Win16) scheduler/syslevel.c: kernel32: _KERNEL32_86: illegal call to _LeaveSysLevel (Win32 -> Win16) [...]
There must have been some problem converting these? Or were they introduced after the cleanup maybe? I decided to look further:
tools/winapi_check/winapi_check --none --progress --cross-call-unicode-ascii
[...] dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to HEAP_strdupWtoA (Unicode -> ASCII) dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to OpenDriverA (Unicode -> ASCII) [...]
What's the story here?
I decided to take on the 'Include statements should use <> instead of ""' item, as it seems nobody else is working on that (except Patrick Stridvall who created that useful patch to winapi_cleanup of course). My plan is to supply one patch for each directory in $WINE_SRC/* and $WINE_SRC/dlls/*, in 'ls' sorting order, to keep the patch size down. Is that OK with you?
-Hans
----------------------------------------------- Hans Leidekker Vrije Universiteit Dienst IT Systeembeheer Campusnetwerk De Boelelaan 1105 1081 HV Amsterdam +31 (0) 20 444 5385 The Netherlands -----------------------------------------------
On April 8, 2003 08:54 am, Hans Leidekker wrote:
Hi,
I was looking for a way to kill some time when I ended up on (WineHQ's version of) Dimi's janitorial page. It seems the page could use an update since as of this moment some of the items seem to have been finished, or nearly so:
The one on my site is a bit more up to date: http://www.dssd.ca/wine/Wine-Janitorial.html
I will try to update the WineHQ one on a regular basis, but the latest and greatest will always be on my site. For this reason, please send updates for this page to me against my version, and I'll merge them upstream.
As for the status, the page should be fairly up to date. I have tried to maintain it daily, if you notice than things are no longer current, please let me know.
tools/winapi_check/winapi_check --none --progress --cross-call-win32-win16
[...] loader/module.c: kernel32: WIN32_GetProcAddress16: illegal call to GetProcAddress16 (Win32 -> Win16) scheduler/syslevel.c: kernel32: _KERNEL32_86: illegal call to _LeaveSysLevel (Win32 -> Win16) [...]
There must have been some problem converting these? Or were they introduced after the cleanup maybe? I decided to look further:
These ones seem fishy -- remember, wineapi_check makes mistakes sometimes.
tools/winapi_check/winapi_check --none --progress --cross-call-unicode-ascii
[...] dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to HEAP_strdupWtoA (Unicode -> ASCII) dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to OpenDriverA (Unicode -> ASCII) [...]
What's the story here?
What do you mean? Why did you pick these 2 examples out of the entire list?
I decided to take on the 'Include statements should use <> instead of ""' item, as it seems nobody else is working on that (except Patrick Stridvall who created that useful patch to winapi_cleanup of course). My plan is to supply one patch for each directory in $WINE_SRC/* and $WINE_SRC/dlls/*, in 'ls' sorting order, to keep the patch size down. Is that OK with you?
I don't think you need to do that. It's really up to Alexandre to run the script, we've discussed this before and he does not want a patch.
BTW, if you feel like working on something, programs/winecfg is a good start. All you need is C and Win32 programming an you're all set. Besides, it blocks a big chuck of the 0.9 TODO: http://www.dssd.ca/wine/Wine-0.9-TODO.html#winecfg
On Tuesday 08 April 2003 15:21, Dimitrie O. Paun wrote:
The one on my site is a bit more up to date: http://www.dssd.ca/wine/Wine-Janitorial.html
Well, that one's not up to date either. Running the checks on current CVS shows only two calls left, both for --cross-call-win32-win16 and --cross-call-unicode-ascii. The [...] marks in my mail were there to skip some error messages winapi_check spits out.
tools/winapi_check/winapi_check --none --progress --cross-call-unicode-ascii
[...] dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to HEAP_strdupWtoA (Unicode -> ASCII) dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to OpenDriverA (Unicode -> ASCII) [...]
What's the story here?
What do you mean? Why did you pick these 2 examples out of the entire list?
As I say, those are the only ones left at present.
I decided to take on the 'Include statements should use <> instead of ""' item, as it seems nobody else is working on that (except Patrick Stridvall who created that useful patch to winapi_cleanup of course). My plan is to supply one patch for each directory in $WINE_SRC/* and $WINE_SRC/dlls/*, in 'ls' sorting order, to keep the patch size down. Is that OK with you?
I don't think you need to do that. It's really up to Alexandre to run the script, we've discussed this before and he does not want a patch.
I see. I missed that discussion. <Insert some self punishment here, and some mumbling about properly searching the archives>
BTW, if you feel like working on something, programs/winecfg is a good start. All you need is C and Win32 programming an you're all set. Besides, it blocks a big chuck of the 0.9 TODO: http://www.dssd.ca/wine/Wine-0.9-TODO.html#winecfg
Well, thanks for the hint, but I think I'll go back to that regression test I started for file change notifications then.
-Hans
----------------------------------------------- Hans Leidekker Vrije Universiteit Dienst IT Systeembeheer Campusnetwerk De Boelelaan 1105 1081 HV Amsterdam +31 (0) 20 444 5385 The Netherlands -----------------------------------------------
On April 8, 2003 09:53 am, Hans Leidekker wrote:
Well, that one's not up to date either. Running the checks on current CVS shows only two calls left, both for --cross-call-win32-win16 and --cross-call-unicode-ascii. The [...] marks in my mail were there to skip some error messages winapi_check spits out.
Well, the pages says that the win32-win16 task is done -- all entries are grayed out, so no problem there. As for the unicode-ascii case, I don't understand where all those problems went. Running winapi_check on my tree aborts now with a problem in winapi_check itself:
dlls/lzexpand/lzexpand.specdlls/advapi32/advapi32.spec: 2: syntax error: '@ stdcall AbortSystemShutdownA(ptr)' BEGIN failed--compilation aborted at tools/winapi_check/winapi_check line 55, <IN> line 2.
On Wednesday 09 April 2003 08:14, Dimitrie O. Paun wrote:
I don't understand where all those problems went. Running winapi_check on my tree aborts now with a problem in winapi_check itself:
dlls/lzexpand/lzexpand.specdlls/advapi32/advapi32.spec: 2: syntax error: '@ stdcall AbortSystemShutdownA(ptr)' BEGIN failed--compilation aborted at tools/winapi_check/winapi_check line 55, <IN> line 2.
Yes, it's because the spec files don't have explicit handlers anymore. I fixed that:
http://cvs.winehq.com/patch.py?id=7801
winapi_check still has some problems parsing Wine code though:
tools/winapi_check/winapi_check --none --progress --cross-call-unicode-ascii
dlls/d3d8/shader.c: 243: can't parse: 'typedef FLOAT D3DMATRIX44[4][4];' dlls/d3d8/shader.c: 244: can't parse: 'typedef FLOAT D3DMATRIX43[4][3];' dlls/d3d8/shader.c: 245: can't parse: 'typedef FLOAT D3DMATRIX34[4][4];' dlls/d3d8/shader.c: 246: can't parse: 'typedef FLOAT D3DMATRIX33[4][3];' dlls/d3d8/shader.c: 247: can't parse: 'typedef FLOAT D3DMATRIX32[4][2];' dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to HEAP_strdupWtoA (Unicode -> ASCII) dlls/winmm/driver.c: winmm & winmm: OpenDriverW: illegal call to OpenDriverA (Unicode -> ASCII) libs/port/getopt.c (file 754 of 945)*** Can't parse '#if !defined __STDC__ || !__STDC__' *** *** Can't parse '#if !defined _LIBC && defined __GLIBC__ && __GLIBC__ >= 2' *** *** Can't parse '#if HAVE_STRING_H - 0' *** *** Can't parse '#if || defined _LIBC' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** libs/port/getopt.c (file 754 of 945): getenv*** Can't parse '#if && !defined strlen' *** libs/port/getopt.c (file 754 of 945): strlen*** Can't parse '#if defined __STDC__ && __STDC__' *** libs/port/getopt.c (file 754 of 945): exchange*** Can't parse '#if defined _LIBC && defined USE_NONOPTION_FLAGS' *** *** Can't parse '#if defined __STDC__ && __STDC__' *** libs/port/getopt.c (file 754 of 945): _getopt_initialize*** Can't parse '#if defined _LIBC && defined USE_NONOPTION_FLAGS' *** *** Can't parse '#if defined _LIBC && defined USE_NONOPTION_FLAGS' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** *** Can't parse '#if defined _LIBC && defined USE_IN_LIBIO' *** libs/port/getopt1.c (file 755 of 945)*** Can't parse '#if !defined __STDC__ || !__STDC__' *** *** Can't parse '#if !defined _LIBC && defined __GLIBC__ && __GLIBC__ >= 2' ***
-
Dimitrie O. Paun -
Hans Leidekker