June 2007 - Wine-Devel - list.winehq.org

Wine fails to start browser
by Louis. Lenders June 3, 2007

June 3, 2007

Hi, i noticed some kind of regression in current git, for example when i try to start http://fe.trymedia.com/dm/infogrames/30m_default/t_20em/FreddiKelp.exe , a download manager for the game. With current git i get this: Usage: winebrowser URL That's it, whereas in wine-0.9.34, it opens my firefox fine, at the correct page , and starts to download the game. The only solution i found now is again add "%1" to the key HKEY_CLASSES_ROOT\http\shell\open\command\winebrowser, but that "fix' was not acceptable in wine, me was told a few weeks ago. I saw some patches from Misha Koshelev that might be related to this, maybe changing the key is acceptable now, or anyone know another fix? Thx --------------------------------- Yahoo! Answers - Got a question? Someone out there knows the answer. Tryit now.

1 0

Support for iFilter API
by Stefan Brüns June 3, 2007

June 3, 2007

Hi, recently i had a look at the iFilter API. iFilter is an interface to extract text, formatting and various attributes from text like documents (pdf, rtf, html, ...). The actual extraction is done by a format specific plugin which gets registered to the Registry. I started my tests by taking the headers from Visual Studio, and most tests went quite fine, so I have a winelib application which is able to extract text from some proprietary documents. Next, I would like to get rid of the requirement for the headers taken from VS, so my actual question is: How to implement these headers so they can be incorporated into the wine source? As it is just an interface, there are no dlls which have to be implemented. Bye, Stefan -- Stefan Brüns / Kastanienweg 6 - Zimmer 1206 / 52074 Aachen mailto:lurch(a)gmx.li http://www.kawo1.rwth-aachen.de/~lurchi/ phone: +49 241 169-4206 mobile: +49 160 3797725

1 0

Re: webpage: Update Fedora download information
by Saulius Krasuckas June 3, 2007

June 3, 2007

* On Sat, 2 Jun 2007, Andreas Bierfert wrote: > here is an upgrade for the fedora download information as I requested on > w-d-l. Hi, your patch isn't in unified fromat. Probably add "-u" option.

1 1

How to determine if something is "debuggable" ... ?
by Marc G. Fournier June 3, 2007

June 3, 2007

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I realize that that may be an odd question, but I also don't want spend an huge amount of time trying to get something to work that just won't work "at this time" ... I've been asked to get a Windows application running under FreeBSD ... optimally, I'd prefer to do it under Wine, since the last thing I want is a) to buy a Windows License and b) to run Windows when all I need is this one application ... First and foremost, I'm most impressed with how far Wine has come ... I remember back when you needed the Windows OS to do anything ... with current Wine, I was able to run the Setup / Installer for this application, it built my .wine directory, installed everything beautifully ... .. the problem arises when I try and run the app after install ... and the odd thing is that it doesn't appear to actually do anything, it just hangs there, no errors to the screen. I pop'd into IRC and asked about debugging there, and was pointed to the WINEDEBUG=+all setting, which I've since and re-tried to run, without output redirected to a log file ... it very quickly gets to 65M in size and 900k lines ... most of which looks to be looping around the same dozen lines: 0009:trace:seh:call_stack_handlers calling handler at 0x55779e code=c0000005 flags=0 0009:Call msvcrt._except_handler3(0035f8e4,0035fab0,0035f618,0035f54c) ret=9c1a70fd 0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 0009:trace:seh:_except_handler3 filter = 0x557756 0009:Call msvcrt._XcptFilter(c0000005,0035f400) ret=00557767 0009:trace:seh:_XcptFilter (c0000005,0x35f400) 0009:Ret msvcrt._XcptFilter() retval=00000000 ret=00557767 0009:trace:seh:_except_handler3 filter returned CONTINUE_SEARCH 0009:trace:seh:_except_handler3 reached TRYLEVEL_END, returning ExceptionContinueSearch 0009:Ret msvcrt._except_handler3() retval=00000001 ret=9c1a70fd 0009:trace:seh:call_stack_handlers handler at 0x55779e returned 1 Using grep -n, it looks like starting around like 43k in the log is where this happens, and continues ad finitum: 43048:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 43059:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 43070:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 43081:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 43092:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub 43103:0009:trace:seh:_except_handler3 exception c0000005 flags=0 at 0x0 handler=0x55779e 0x35f618 0x35f54c semi-stub <alot of lines deleted> 917026:0009:trace:seh:_except_handler3 exception c000013a flags=0 at 0x9c0605d7 handler=0x55779e 0x35f128 0x35f06c semi-stub 917037:0009:trace:seh:_except_handler3 exception c000013a flags=0 at 0x9c0605d7 handler=0x55779e 0x35f128 0x35f06c semi-stub 917048:0009:trace:seh:_except_handler3 exception c000013a flags=0 at 0x9c0605d7 handler=0x55779e 0x35f128 0x35f06c semi-stub 917059:0009:trace:seh:_except_handler3 exception c000013a flags=0 at 0x9c0605d7 handler=0x55779e 0x35f128 0x35f06c semi-stub 917070:0009:trace:seh:_except_handler3 exception c000013a flags=0 at 0x9c0605d7 handler=0x55779e 0x35f128 0x35f06c semi-stub Looking at the log just before that loop started (not sure how many lines before is significant), there is: 0009:Call msvcrt._initterm(00575000,0057508c) ret=005576e2 0009:trace:msvcrt:_initterm (0x575000,0x57508c) 0009:trace:msvcrt:_initterm Call init function 0x5578f4 0009:trace:seh:raise_exception code=c0000005 flags=0 addr=0x0 0009:trace:seh:raise_exception info[0]=00000000 0009:trace:seh:raise_exception info[1]=00000000 0009:trace:seh:raise_exception eax=005578f4 ebx=9ca6b678 ecx=00000000 edx=00000038 esi=00575004 edi=0057508c 0009:trace:seh:raise_exception ebp=0035f968 esp=0035f93c cs=0033 ds=003b es=003b fs=1007 gs=001b flags=00010206 The fun part is that I don't know what is important, and what isn't ... if someone would like to see the complete log before the semi-stub errors start to loop, please feel free to ask ... the software I'm trying to run is available as a free download, if someone wishes to actually try the software ... Compared to the graphical games that ppl show listed on AppDB, I would have thought this one would have been simple ... :) How best to proceed? Thank you ... - ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email . scrappy(a)hub.org MSN . scrappy(a)hub.org Yahoo . yscrappy Skype: hub.org ICQ . 7615664 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (FreeBSD) iD8DBQFGYMaY4QvfyHIvDvMRAjYAAKDm/YJ8ExAv/NphfJomp9igJ1CH/gCg0zTK pyudGfNTQ9o8bgorDhcF9MA= =DEaH -----END PGP SIGNATURE-----

5 12

Re: crypt32: remove 'recursive registry key delete' function
by Paul Vriens June 3, 2007

June 3, 2007

On 6/1/07, Stefan Leichter <Stefan.Leichter(a)camline.com> wrote: > ChangeLog > --------- > replace CRYPT_RecurseDeleteKey with RegDeleteTreeW > > > > > Hi Stefan, I don't think that's enough. There's a difference between SHDeleteKeyW and RegDeleteTreeW (see also http://wiki.winehq.org/RecursiveRegistryKeyDelete, but I guess that's a known page to you). Basically you have to remove the key itself as well. Cheers, Paul.

2 2

Re: user32:tests:msg Add test for switch mdi childs
by Dmitry Timoshkov June 3, 2007

June 3, 2007

"Anatoly Lyutin" <vostok(a)etersoft.ru> wrote: > +/* Switch visible mdi childs */ Please use the word 'children' to avoid confusion. > + trace("Switch child window.\n"); > + > + ret = SendMessageA(mdi_client, WM_MDIACTIVATE, (WPARAM)mdi_child2, 0); > + ok(!ret, "Child not been deactivate(Switch test)\n"); active_child == mdi_child2, so it's not clear what you try to do in the test above. Besides, you don't check any subsequences of the call. Since below starts another test 'flush_sequence();' should be added before. > + ret = SendMessageA(mdi_client, WM_MDIACTIVATE, (WPARAM)mdi_child, 0); > + > + ok(!ret, "Child not activate(Switch test)\n"); > + > + ok_sequence(WmSwitchChild,"Child not switch correctly\n",TRUE); Again, you don't check any subsequences of the call. Also if you could add comments in the added message sequence what is going on, it will help a lot later figuring out how to make the test pass under Wine. Something along the following lines present in the existing sequences: /* maximize the 1st MDI child */ ... /* restore the 2nd MDI child */ ... { WM_CHILDACTIVATE, sent|defwinproc|wparam|lparam, 0, 0 }, /* in the 1st MDI child */ { WM_NCACTIVATE, sent|wparam|defwinproc, 0 }, /* in the 1st MDI child */ { WM_MDIACTIVATE, sent|defwinproc }, /* in the 1st MDI child */ ... { WM_KILLFOCUS, sent }, /* in MDI client */ { WM_IME_SETCONTEXT, sent|wparam|optional, 0 }, /* in MDI client */ etc. I understand that this is a tedious work, but it should be done. -- Dmitry.

1 0

Re: crypt32: implement CryptSIPLoad (try 5)
by Juan Lang June 2, 2007

June 2, 2007

> I think it's worth figuring out why native does it like this. Believe me, I've been trying to. I'm only inferring that it should be loaded because using native wintrust calls CryptSIPLoad, and my implementation makes it continue on to call other functions that seem reasonable in that context. I don't know what tests I can write to show that I should not pass the DLL's function pointers directly back to the caller. That might make sense if crypt32 were to try to unload unused function pointers periodically, since the API doesn't provide an unload function - but that seems overly complicated, and I unload them at DLL unload time instead. If you have suggestions on tests I can write, I'm all ears. > Also the dll loading part of the patch is quite confusing, hopefully > doing it closer to the native way would help make it clearer. The complication comes from the fact that the registry allows one to have a separate DLL for each function in a SIP, but CryptAddSIPProvider does not. I therefore fail if the DLL names are different. This allows me to use an HMODULE as the hSIP member of a SIP_DISPATCH_INFO, rather than invent some more complicated scheme. I can add more comments to try to explain that if you like. --Juan ____________________________________________________________________________________ Bored stiff? Loosen up... Download and play hundreds of games for free on Yahoo! Games. http://games.yahoo.com/games/front

3 2

Re: wordpad: (1/6) Split code into set_caption function
by Andrey Turkin June 2, 2007

June 2, 2007

Alexander Nicolaysen Sørnes wrote: > This will be useful for the file saving code. > > > Regards, > > Alexander N. Sørnes > + WCHAR wszCaption[MAX_PATH]; + + if(wszNewFileName) + { + lstrcpyW(wszCaption, wszNewFileName); + lstrcatW(wszCaption, wszSeparator); + lstrcatW(wszCaption, wszAppTitle); Looks like there are possible buffer overflow

1 0

Re: How to determine if something is "debuggable" ... ?
by Tom Spear June 2, 2007

June 2, 2007

On 6/1/07, Marc G. Fournier <scrappy(a)hub.org> wrote: Another thing you can _try_, since it looks like it messing up in msvcrt or possibly even dbghelp, is to get msvcrt.dll and dbghelp.dll from a windows install, place them in your wine's c:\windows\system32 folder (usually ~/.wine/drive_c/windows/system32), and then use winecfg to set both dll's to run (native, builtin) for that application.. See if that helps, no guarantees tho, especially since it is FreeBSD. -- Thanks Tom

1 0

GDI+: headers and one test
by Evan Stade June 1, 2007

June 1, 2007

Changelog: *Added public GDI+ headers in include/ *Added private GDI+ headers in dlls/gdiplus *Added small implementation of GDI+ pens in dlls/gdiplus/pen.c *Added (first) GDI+ test dlls/gdiplus/Makefile.in | 3 + dlls/gdiplus/gdip_defines.h | 43 +++++++++++++++++++ dlls/gdiplus/gdip_enumerations.h | 57 ++++++++++++++++++++++++++ dlls/gdiplus/gdiplus.c | 27 ++++++++++++ dlls/gdiplus/gdiplus.spec | 8 ++-- dlls/gdiplus/gdiplus_private.h | 43 +++++++++++++++++++ dlls/gdiplus/pen.c | 84 ++++++++++++++++++++++++++++++++++++++ dlls/gdiplus/pen.h | 30 ++++++++++++++ dlls/gdiplus/tests/Makefile.in | 13 ++++++ dlls/gdiplus/tests/pen.c | 44 ++++++++++++++++++++ include/Gdiplus.h | 10 +++++ include/GdiplusEnums.h | 39 ++++++++++++++++++ include/GdiplusFlat.h | 23 ++++++++++ include/GdiplusGpStubs.h | 10 +++++ include/GdiplusInit.h | 26 ++++++++++++ include/GdiplusTypes.h | 7 +++ 16 files changed, 462 insertions(+), 5 deletions(-) -Evan Stade

6 5

Re: Added GDI+ public headers, small implementation of pen, memory and initialization, and two tests (2nd try)
by Detlef Riekenberg June 1, 2007

June 1, 2007

On Do, 2007-05-31 at 18:02 -0700, Evan Stade wrote: > Changelog: > *Added GDI+ public headers > *Added some implementation (pen.c, gdiplus.c, memory.c) > *Added two tests (pen.c, init.c) I good indicator, that the Patch should be split in smaller Parts, is a Changelog (Subject) that is larger than a simple Line or includes an enumeration. You already use 3 Lines in the Changelog which describe the code for a minimum of 3 Patches. The usual way is one Patch for every new header to reduce the complexity of the Patch, but your headers are really small, so 2 Patches or maybe 1 single header-Patch might be ok (this depend also on the size). Thanks -- By by ... Detlef

3 2

Re: [1/2] uninstaller: add ability to scan HKCU for uninstall entries (Try 2)
by James Hawkins June 1, 2007

June 1, 2007

On 6/1/07, Tom Spear <speeddymon(a)gmail.com> wrote: > Here is a 2nd attempt at the last one I sent. This time the root > field is initialized, and this part of the patch is written against > current git, not against changes to my own tree. Oh and its onl;y a > 2-parter > > This will make wine's programs/uninstaller look thru not only > HKEY_LOCAL_MACHINE for uninstall entries, but also HKEY_CURRENT_USER. > + /* Loop thru HKCU first, then thru HKLM */ + for (iRootKey=0; iRootKey<sizeof(rootKeys) / sizeof(rootKeys[0]); ++iRootKey) + { + /* If there is no uninstall info in a specific root key, + * finish this run and go to the next */ + if (RegOpenKeyExW(rootKeys[iRootKey], PathUninstallW, 0, + KEY_READ, &hkeyUninst) != ERROR_SUCCESS) + continue; This whole patch would be a lot simpler if you just used one root variable: if (RegOpenKeyExW(HKCU....) == ERROR_SUCCESS) root = HKCU; else if (RegOpenKeyExW(HKLM....) == ERROR_SUCCESS) root = HKLM; else bail out Now just use root everywhere instead of having an array of just two constants, and you get rid of the index. -- James Hawkins

3 3

Re: [2/2] uninstaller: indentation cleanups (try 2)
by Tom Spear June 1, 2007

June 1, 2007

On 6/1/07, James Hawkins <truiken(a)gmail.com> wrote: > On 6/1/07, Tom Spear <speeddymon(a)gmail.com> wrote: > > Here is part 2 of try 2. Just cleans up the indentation of my changes > > in the previous patch in this series. > > > > I think you misunderstood what Dan meant about indentation changes. > You shouldn't have a patch that doesn't properly indent if you add a > new block, and then come in with another patch to fix the indentation. That is exactly what he told me to do, and I even submitted it to him like this and he said to send it in. -- Thanks Tom

1 0

Re: [1/2] uninstaller: add ability to scan HKCU for uninstall entries (Try 2)
by James Hawkins June 1, 2007

June 1, 2007

On 6/1/07, Tom Spear <speeddymon(a)gmail.com> wrote: > Here is a 2nd attempt at the last one I sent. This time the root > field is initialized, and this part of the patch is written against > current git, not against changes to my own tree. Oh and its onl;y a > 2-parter > > This will make wine's programs/uninstaller look thru not only > HKEY_LOCAL_MACHINE for uninstall entries, but also HKEY_CURRENT_USER. > + /* Loop thru HKCU first, then thru HKLM */ + for (iRootKey=0; iRootKey<sizeof(rootKeys) / sizeof(rootKeys[0]); ++iRootKey) + { + /* If there is no uninstall info in a specific root key, + * finish this run and go to the next */ + if (RegOpenKeyExW(rootKeys[iRootKey], PathUninstallW, 0, + KEY_READ, &hkeyUninst) != ERROR_SUCCESS) + continue; Indentation? -- James Hawkins

2 1

Re: [2/2] uninstaller: indentation cleanups (try 2)
by James Hawkins June 1, 2007

June 1, 2007

On 6/1/07, Tom Spear <speeddymon(a)gmail.com> wrote: > Here is part 2 of try 2. Just cleans up the indentation of my changes > in the previous patch in this series. > I think you misunderstood what Dan meant about indentation changes. You shouldn't have a patch that doesn't properly indent if you add a new block, and then come in with another patch to fix the indentation. -- James Hawkins

1 0

Re: [PATCH 3/4] msi: msiserver.idl: Maintain spacing as spaces, not tabs.
by James Hawkins June 1, 2007

June 1, 2007

On 6/1/07, Misha Koshelev <mk144210(a)bcm.edu> wrote: > --- > dlls/msi/msiserver.idl | 6 +++--- > 1 files changed, 3 insertions(+), 3 deletions(-) > Stupid eclipse...only applies my tab preferences to c files. -- James Hawkins

1 0

Re: mshtml #2: Added warning if a wrong Wine Gecko package version was found.
by Alexandre Julliard June 1, 2007

June 1, 2007

Jacek Caban <jacek(a)codeweavers.com> writes: > @@ -212,8 +213,10 @@ static BOOL load_wine_gecko(PRUnichar *gre_path) > if(res != ERROR_SUCCESS || type != REG_SZ) > return FALSE; > > - if(TRACE_ON(mshtml)) > - check_version(gre_path); > + /* We may consider message box here */ > + if(ERR_ON(mshtml) && !check_version(gre_path)) > + MESSAGE("Found a version of Wine Gecko package that is not dedicated " > + "to your Wine version. Reinstalation is recommended.\n"); If you want to check versions, there needs to be a way to specify in the URL which version you want to download, and old versions need to be kept available. The target CPU should also be part of the URL so we can support multiple platforms. -- Alexandre Julliard julliard(a)winehq.org

2 5

Bug triage, or spam?
by Tom Spear June 1, 2007

June 1, 2007

Hi all, I just have a quick question.. I got a message from someone last night asking me to stop closing bugs because I'm spamming the list, however I also received a message from someone a couple of nights ago thanking me for doing bug triage.. Which is it, and if it is both, then what draws the line between triage and spam, and how do I close stale bugs without it being considered spam? -- Thanks Tom

6 17

The ddraw tests don't compile with Visual C++ 2005.
by Francois Gouget June 1, 2007

June 1, 2007

Visual C++ 2005 does not like the way we compute the infinity and NaN values. I've tried replacing the current divisions by zero with arithmetic on FLT_MAX (at least for the infinity calculations), but it did not like that either. Does anyone know how to make these calculations portable? How does one get at these values on Windows? diff --git a/dlls/ddraw/tests/d3d.c b/dlls/ddraw/tests/d3d.c [...] light.dvAttenuation0 = -1.0 / 0.0; /* -INFINITY */ [...] light.dvAttenuation0 = 1.0 / 0.0; /* +INFINITY */ [...] light.dvAttenuation0 = 0.0 / 0.0; /* NaN */ -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ 1 + e ^ ( i * pi ) = 0

4 4

Re: msvfw32: Fix enumeration of VFW video codecs that are registered in the registry.
by Detlef Riekenberg June 1, 2007

June 1, 2007

On Fr, 2007-06-01 at 00:25 +0200, Julien Muchembled wrote: > There are 2 ways to register a VFW video codec: > (1) the [drivers32] section of system.ini : "vidc.<FOURCC>=<dll_name>" > (2) the "HKLM\Software\Microsoft\Windows NT\CurrentVersion\Drivers32" key: > - value name: "vidc.<FOURCC>" > - value data: "<dll_name>" There ia a magic with the Name "IniFileMapping": On NT, (1) is redirected to (2). Wine is still affected by bug 4096 -- By by ... Detlef

1 0

FPS tool for wine
by Tomas.Zijdemans＠student.uib.no June 1, 2007

June 1, 2007

The largest gaming site in Norway recently did an extensive review of gaming on Linux, but Wine was left out of the benchmark because "no FPS tool exists for Wine". Surely this can't be true?

10 11

Re: crypt32: implement CryptSIPLoad (try 5)
by Paul Vriens June 1, 2007

June 1, 2007

On 6/1/07, Juan Lang <juan_lang(a)yahoo.com> wrote: > Answering my own question, I see that the previous try introduced compiler > warnings. Fixed. > > ChangeLog: implement CryptSIPLoad > > --Juan > Hi Juan, This patch still uses the non-native behavior. Is there a consensus on this? I guess we are the only ones talking about this but I hope there are more opinions. I said before that I also don't know why native does it like that and I don't see any benefit (yet). If this patch does get accepted we should have a proper comment in the source (or are the comments in the tests enough?). Cheers, Paul.

2 1

Re: Added GDI+ public headers, small implementation of pen, memory and initialization, and two tests (2nd try)
by Alexandre Julliard June 1, 2007

June 1, 2007

"Evan Stade" <estade(a)gmail.com> writes: > I made the fixes that Dmitry suggested. Additionally I changed the > public header files to all lower-case (to match the wine headers > rather than platform sdk headers). I also consolidated a couple of > source files so there wouldn't be files with only 2 functions in them. That's good, please do the same with the init test, there's no reason to have a separate file for that. > I believe there was a question earlier on the list about why there is > "#ifndef __cplusplus ... #endif" around portions of the headers. This > is because this implementation of gdi+ only supports c and not c++. That's not OK, the implementation clearly has to support C++ too, even if it is written in C. You should also send that stuff in smaller chunks. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re:wine.inf: Create fake dll for iexplore.exe.
by Louis. Lenders June 1, 2007

June 1, 2007

>wine.inf: Create fake dll for >iexplore.exe. >Vitaliy Margolen wine-patches at kievinfo.com >Thu May 24 08:50:10 CDT 2007 >Some older programs check if IE is installed looking for c:\Program >Files\Internet Explorer\iexplore.exe --- tools/wine.inf | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)

2 1