September 2008 - Wine-Devel

Re: [1/3] mshtml: Implement HTMLElement2 get_scrollLeft (try 2)
by Alexandre Julliard Sept. 10, 2008

Sept. 10, 2008

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > + nsres = nsIDOMHTMLElement_QueryInterface(This->nselem, &IID_nsIDOMNSHTMLElement, (void**)&nselem); > + if(NS_SUCCEEDED(nsres)) > + { > + nsres = nsIDOMNSHTMLElement_GetScrollLeft(nselem, &left); > + nsIDOMNSHTMLElement_Release(nselem); > + if(NS_FAILED(nsres)) > + left = 0; > + } > + else > + { > + ERR("Could not get nsIDOMNSHTMLElement interface: %08x\n", nsres); > + } This ERR obviously needs the same treatment as the other one. Also please fix your tab size. -- Alexandre Julliard julliard(a)winehq.org

1 0

re: coverity ... new run finally
by Dan Kegel Sept. 10, 2008

Sept. 10, 2008

Marcus wrote: > A new Coverity run (274) was done finally, so > we can restart looking at issues ;) > > Still a lot of NULL ptr migration issues which suck > to fix, but well. Yay! Here are the new errors introduced since the end of May, sorted by source file. There are only 23 not in test code: 725 NULL_RETURNS Context_CopyProperties crypt32/context.c 715 FORWARD_NULL CryptMsgGetAndVerifySigner crypt32/msg.c 709 DEADCODE DEVENUM_ReadPinTypes devenum/createdevenum.c 717 FORWARD_NULL DEVENUM_ReadPinTypes devenum/createdevenum.c 726 OVERRUN_STATIC test_EnumGroupsInGroup dplayx/tests/dplayx.c 718 FORWARD_NULL MSI_GetUserInfo msi/msi.c 747 REVERSE_INULL ILockBytes_RemoteReadAt_Stub ole32/ole32_objidl_p.c 748 REVERSE_INULL ILockBytes_RemoteReadAt_Stub ole32/ole32_objidl_p.c 749 REVERSE_INULL ISequentialStream_RemoteRead_Stub ole32/ole32_objidl_p.c 750 REVERSE_INULL ISequentialStream_RemoteRead_Stub ole32/ole32_objidl_p.c 751 REVERSE_INULL IEnumOLEVERB_RemoteNext_Stub ole32/ole32_oleidl_p.c 752 REVERSE_INULL IEnumOLEVERB_RemoteNext_Stub ole32/ole32_oleidl_p.c 373 REVERSE_INULL IDispatch_GetIDsOfNames_Stub oleaut32/oleaut32_oaidl_p.c 753 REVERSE_INULL ITypeLib_RemoteFindName_Stub oleaut32/oleaut32_oaidl_p.c 754 REVERSE_INULL IEnumConnectionPoints_RemoteNext_Stub oleaut32/oleaut32_ocidl_p.c 755 REVERSE_INULL IEnumConnections_RemoteNext_Stub oleaut32/oleaut32_ocidl_p.c 714 FORWARD_NULL MediaSeekingImpl_SetPositions quartz/control.c 713 FORWARD_NULL TestFilter_Create quartz/tests/filtergraph.c 719 FORWARD_NULL ME_FindPixelPos riched20/caret.c 708 DEADCODE array_compute_and_size_conformance rpcrt4/ndr_marshall.c 720 FORWARD_NULL CreateContext wined3d/context.c 721 FORWARD_NULL CreateContext wined3d/context.c 722 FORWARD_NULL test_OpenRequest winhttp/tests/winhttp.c 723 FORWARD_NULL test_SendRequest winhttp/tests/winhttp.c 724 FORWARD_NULL test_WinHttpAddHeaders winhttp/tests/winhttp.c 716 FORWARD_NULL InternetCrackUrlW wininet/internet.c 712 FORWARD_NULL test_mmioOpen winmm/tests/mmio.c 710 FORWARD_NULL ldap_parse_sort_controlW wldap32/parse.c 711 FORWARD_NULL ldap_parse_vlv_controlW wldap32/parse.c Sure would be nice if Coverity could identify the cl which introduced each problem :-) - Dan

6 8

Re: [2/3] mshtml: Add tests for get_scrollLeft
by Alistair Leslie-Hughes Sept. 10, 2008

Sept. 10, 2008

"Alistair Leslie-Hughes" <leslie_alistair(a)hotmail.com> wrote in message news:48C65494.2010204(a)hotmail.com... > Hi, > > Changelog: > mshtml: Add tests for get_scrollLeft > > Best Regards > Alistair Leslie-Hughes Hi, Swap the order of patch 2 and 3 and everything all tests will pass ok. Best Regards Alistair Leslie-Hughes

1 0

re: Debugging applications running on wine
by Dan Kegel Sept. 9, 2008

Sept. 9, 2008

Kevin K wrote: > Is winedbg the only method of debugging applications being > developed for Windows on Wine? > > For instance, assume a program developed with Visual Studio > in C or C++, and I needed to debug it on Linux? If winedbg doesn't work for you, we should fix it. Same goes for other debuggers. For instance, ollydbg seems to work http://appdb.winehq.org/objectManager.php?sClass=application&iId=2619 Visual C++ 6's debugger seems to work at least a little, too: http://appdb.winehq.org/objectManager.php?sClass=version&iId=31 Please file bugs for any problems you run into. - Dan

6 6

Re: msi: Uninitialized variable fix (Coverity)
by James Hawkins Sept. 9, 2008

Sept. 9, 2008

On Tue, Sep 9, 2008 at 3:54 PM, Andrew Talbot <andrew.talbot(a)talbotville.com> wrote: > Fix for Coverity error CID: 762. > > --- > Changelog: > msi: Uninitialized variable fix (Coverity). > > diff --git a/dlls/msi/action.c b/dlls/msi/action.c > index 8a8efe7..48557f0 100644 > --- a/dlls/msi/action.c > +++ b/dlls/msi/action.c > @@ -3838,7 +3838,7 @@ static UINT ACTION_PublishFeatures(MSIPACKAGE *package) > MSIFEATURE *feature; > UINT rc; > HKEY hkey; > - HKEY userdata; > + HKEY userdata = NULL; > > if (!msi_check_publish(package)) > return ERROR_SUCCESS; > @@ -3952,7 +3952,7 @@ static UINT ACTION_PublishFeatures(MSIPACKAGE *package) > > end: > RegCloseKey(hkey); > - RegCloseKey(userdata); > + if (userdata) RegCloseKey(userdata); > return rc; Please don't add another NULL-before-free check. -- James Hawkins

3 4

Adding Flawfinder to Patchwatcher
by Austin English Sept. 9, 2008

Sept. 9, 2008

I had a discussion with Dan about adding Flawfinder to the patchwatcher. Currently, it's got some pretty generic errors, but it seems able to test only patches, so we wouldn't be flooded with old nonbugs (or we could set up a blacklist of safe errors). For reference, I've run it on today's git. I'm attaching the full log, as well as a condensed version of the most common errors (1 per error type). Looks like a lot of chances for buffer overflows.. Thoughts? -Austin

15 19

Patchwatcher online
by Dan Kegel Sept. 9, 2008

Sept. 9, 2008

OK, I've checked in the updated Patchwatcher source http://code.google.com/p/winezeug/source/detail?r=177 and have turned email notifications back on. I also told it to rerun the most recent 11 patches, just to test that it's sending the good patches to http://groups.google.com/group/wine-patches-filtered again. The results page http://kegel.com/wine/patchwatcher/results/ looks nice and green; the blacklist at http://winezeug.googlecode.com/svn/trunk/patchwatcher/blacklist.txt filters out all the tests that tend to fail on my test machine infrequently enough that they show up in the five baseline test runs, but frequently enough that they caused bogus failure reports in the few hundered patches I've processed. It would be *very good* to whittle that blacklist down to zero. I should update http://bugs.winehq.org/show_bug.cgi?id=9916 with a script that runs just the problematic tests over and over again. (Or maybe I should start a new bug, that one is pretty long.)

3 3

Re: ws2_32: ioctlsocket should try to check if argp is valid (try 2)
by Juan Lang Sept. 9, 2008

Sept. 9, 2008

Hi Jeff, + if(sock == INVALID_SOCKET) + skip("Creating the socket failed: %d, skipping test.\n", WSAGetLastError()); skip needs a return too. + ret = ioctlsocket(sock, cmds[i], (u_long *)1); + if(ok(ret == SOCKET_ERROR, "ioctlsocket succeeded unexpectedly\n")) + { I think you mean: ok(ret == SOCKET_ERROR...) if (ret == SOCKET_ERROR) { } although really the if is unnecessary, as you won't cause a crash if the first test fails. --Juan

2 1

Re: ws2_32: ioctlsocket should try to check if argp is valid
by Juan Lang Sept. 9, 2008

Sept. 9, 2008

Hi Jeff, I have a stylistic complaint about your tests in this patch: + if(sock == INVALID_SOCKET) + { + ok(0, "Creating the socket failed: %d, skipping test\n", WSAGetLastError()); + return; + } If you intend to skip a test, shouldn't you use skip instead? + ret = ioctlsocket(sock, cmds[i], (u_long *)1); + if(ret != SOCKET_ERROR) + ok(0, "ioctlsocket succeeded unexpectedly\n"); + else + { I'm not a fan of the ok(0) approach here. Why not use ok(ret == SOCKET_ERROR) instead? --Juan

2 1

Launching a Linux Executable from a Windows app Running under Wine
by Mike Engelhardt Sept. 9, 2008

Sept. 9, 2008

I write a Windows application that I want to be useful under Linux & WINE. I want to use xchm to view the help documentation .chm file since WINE doesn't yet properly display this .chm file which was generated with third party tools and includes advanced features. xchm does render the file acceptably and I'll e-mail the .chm file to anyone interested in that. But for the time being, I want to simply launch xchm from my app whenever my app finds itself being run under WINE. What I have is done so far is system() a command with path names of xchm and my .chm file that seem to basically be trying to outsmart WINE outsmarting Windows. Clearly this is an ugly hack and is not a long term solution. Is there a supported/recommended manner of launching an X Window System executable with Linux absolute path arguments from a MS Windows application running under Linux & WINE? --Mike

3 3

[PATCH] Open file in binary mode to work around a windows bug
by Steven Edwards Sept. 9, 2008

Sept. 9, 2008

-- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

3 5

Re: James Hawkins : netapi32: Fix two failing tests in NT4.
by Kai Blin Sept. 9, 2008

Sept. 9, 2008

On Wednesday 03 September 2008 14:43:55 Alexandre Julliard wrote: > - ok(ret == NERR_BadUsername, "Adding user with too long username > returned 0x%08x\n", ret); + ok(ret == NERR_BadUsername || > + broken(ret == NERR_PasswordTooShort), /* NT4 */ > + "Adding user with too long username returned 0x%08x\n", ret); > I'd argue that this is not broken, but the only error NT4 returns for bad username/password combinations. At least the comments in the surrounding code should be updated. Sorry for not catching this one earlier, I didn't have that much time for Wine recently. Cheers, Kai -- Kai Blin WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin Samba team member http://www.samba.org/samba/team/ -- Will code for cotton.

1 0

Re: user32.dll: Implementation for LockWorkStation() based on xdg-screensaver
by Steven Edwards Sept. 9, 2008

Sept. 9, 2008

On Tue, Sep 9, 2008 at 11:25 AM, Paul Chitescu <paulc(a)voip.null.ro> wrote: > Changelog: > user32.dll: Implementation for LockWorkStation() based on > xdg-screensaver > > Thanks to Steven Edwards (winehacker(a)gmail.com) for the idea of > using "xdg-screensaver lock". When I submitted the patch before, Alexandre and I discussed this both on and off list, I think there is another thread somewhere regarding it and I think he wanted it in explorer because of applications running in Desktop mode. I think the right behavior is to send a message to Explorer and if a wine application is not running in desktop mode then to invoke the xdg-screensaver, otherwise we should stub out or load a windows style screensaver in the emulated desktop. -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

1 0

Re: user32.dll: Implementation for LockWorkStation() based on xdg-screensaver
by Michael Karcher Sept. 9, 2008

Sept. 9, 2008

Am Dienstag, den 09.09.2008, 18:25 +0300 schrieb Paul Chitescu: > Known problems: > 1. Although trying to check with access() if /usr/bin/xdg-screensaver is > executable there is no error returned from spawnvp() if exec() eventually > fails after fork() succeeds. Unlikely but possible. I see that as a bug in spawnvp(). You can not do anything about that. But I dislike checking using access, because a) it is a well-known race condition (in this case, it does not seem to matter, as /usr/bin should be writable by root only). b) you sacrifice the possibility to find /usr/local/bin/xdg-screensaver if the utility has been locally added. (more important point). Would removing the access call and just trying to call xdg-screensaver have any negative implication except for not letting the windows application know that locking failed? If not, I would remove the call, as MSDN says that locking can also fail on Windows for various reasons even if LockWorkStation returned TRUE. > 2. No way to disable it - maybe users don't want their screen locked. So what? If a windows process calls LockWorkStation when the user does not want the workstation to be locked, it's a bug in the windows process, except... > 3. Doesn't take into account virtual desktops, and I have no idea what it > should do in that case, just deny access? ... if running in virtual desktop mode. (I suppose you mean "explorer /desktop" and not virtual desktop provided by X Window managers or virtual desktops provided by the linux kernel) As the real right way to implement this call is to SendMessage to explorer.exe, and explorer knows whether it runs in virtual desktop mode, I would suggest that the patch be extended to handle the screensaver start in Wine's explorer process, and just do nothing in the virtual desktop case. Regards, Michael Karcher

1 0

I just found out that when the input content is already GB2312, “wine” automatically mistakes it for unicode
by hawaii.wine wong Sept. 9, 2008

Sept. 9, 2008

*Hi there, I am a Chinese programmer. I have a problem here with "wine" As we know "wine" is able to convert unicode to GB2312-- but weirdly, I just found out that when the input content is already GB2312, "wine" automatically mistakes it for unicode and therefore attempts to convert it again—and of course it gets a wrong output. **I thought "wine" could recognize GB2312 input and knows this input needs no conversion. Thanks in advance!!*

2 6

Patchwatcher status: added urlmon:protocol.c and urlmon:url.c to blacklist
by Dan Kegel Sept. 9, 2008

Sept. 9, 2008

These tests failed randomly for a few people: urlmon:url.c:2125: Test succeeded inside todo block: unexpected OnProgress_CONNECTING urlmon:url.c:2129: Test failed: expected OnProgress_SENDINGREQUEST urlmon:url.c:2131: Test failed: expected OnResponse urlmon:protocol.c:1597: Test failed: Read failed: 00000001 (0 bytes) urlmon:protocol.c:365: Test failed: expected ReportData urlmon:protocol.c:1598: Test failed: Read failed: 8000000a (0 bytes) so I've added urlmon:url.c and urlmon:protocol.c to the patchwatcher blacklist, All of the tests that currently touch the real internet probably need to be changed to touch a mock internet if we want to have reliable tests.

1 0

Re: user32.dll: Add stub for LockWorkStation
by Steven Edwards Sept. 9, 2008

Sept. 9, 2008

On Mon, Sep 8, 2008 at 2:32 PM, Paul Chitescu <paulc(a)voip.null.ro> wrote: > Changelog: > user32.dll: Stub for LockWorkStation() > > Not sure if user_main.c is the right place but I put it together with > ExitWindowsEx. Would you mind taking a shot at implementing this properly? It was on my TODO but I most likely will never get back to it. Here is my patch from before which was wrong. http://www.winehq.org/pipermail/wine-patches/2007-May/039240.html What we need to do is use SendMessage to talk to the existing explorer.exe process and have it do the locking. Feel free to use what i have to make a patch if your interested. -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

2 1

re: Patchwatcher security improvements
by Dan Kegel Sept. 9, 2008

Sept. 9, 2008

Interesting. One of my goals is to support Solaris and BSD; have you tried your stuff there? I'm currently refactoring patchwatcher.sh; I've pulled the generic stuff out into libpatchwatcher.sh and the wine-specific stuff into wine-slave.sh. Your changes will fit nicely into wine-slave.sh, I hope. I'm surprised you had to give up on the chroot... I was planning on trying to run just wine-slave.sh in a chroot jail, since it's the only part that would actually try to run any part of the wine build system. - Dan

2 2

Re: user32/test Added test about window sizes
by Michael Karcher Sept. 9, 2008

Sept. 9, 2008

Am Dienstag, den 09.09.2008, 14:13 +0200 schrieb Florian Köberle: > + ok( width == 500 && height == 6, "rect is not 6x6 but %dx%d: %d,%d-%d,%d\n", > + width, height, rect.left, rect.top, rect.right, rect.bottom); [...] > + ok( width == 6 && height == 500, "rect is not 6x6 but %dx%d: %d,%d-%d,%d\n", > + width, height, rect.left, rect.top, rect.right, rect.bottom); These messages contain a copy & paste mistake. Please fix the 6x6. Regards, Michael Karcher

1 0

coverity ... new run finally
by Marcus Meissner Sept. 9, 2008

Sept. 9, 2008

Hi folks, A new Coverity run (274) was done finally, so we can restart looking at issues ;) Still a lot of NULL ptr migration issues which suck to fix, but well. Ciao, Marcus

1 0

Re: notepad: Only append .txt to missing files without a dot in them (try2)
by Rob Shearman Sept. 9, 2008

Sept. 9, 2008

2008/9/8 Alexander Nicolaysen Sørnes <alex(a)thehandofagony.com>: > - if (!lstrcmp(txtW, cmdline + lstrlen(cmdline) - lstrlen(txtW))) > + if (strchrW(PathFindFileNameW(cmdline), '.')) You're using one shell helper function already so why not use PathFindExtensionW instead? -- Rob Shearman

1 0

Re: [1/5] secur32: Require gnutls for schannel
by Alexandre Julliard Sept. 9, 2008

Sept. 9, 2008

"Henri Verbeet" <hverbeet(a)gmail.com> writes: > diff --git a/dlls/secur32/Makefile.in b/dlls/secur32/Makefile.in > index df9695a..9805d92 100644 > --- a/dlls/secur32/Makefile.in > +++ b/dlls/secur32/Makefile.in > @@ -6,6 +6,8 @@ MODULE = secur32.dll > IMPORTLIB = secur32 > IMPORTS = netapi32 advapi32 kernel32 ntdll > DELAYIMPORTS = crypt32 > +EXTRALIBS = @GNUTLSLIBS@ > +EXTRAINCL = @GNUTLSINCL@ You need to load the library dynamically, we don't want the whole dll to fail to load if gnutls is missing at run-time. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: mshtml: Implement HTMLElement2 get_scrollLeft
by Alexandre Julliard Sept. 9, 2008

Sept. 9, 2008

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > @@ -831,8 +831,31 @@ static HRESULT WINAPI HTMLElement2_put_scrollLeft(IHTMLElement2 *iface, long v) > static HRESULT WINAPI HTMLElement2_get_scrollLeft(IHTMLElement2 *iface, long *p) > { > HTMLElement *This = HTMLELEM2_THIS(iface); > - FIXME("(%p)->(%p)\n", This, p); > - return E_NOTIMPL; > + nsIDOMNSHTMLElement *nselem; > + PRInt32 left = 0; > + nsresult nsres; > + > + TRACE("(%p)->(%p)\n", This, p); > + > + if(!This->nselem) { > + FIXME("NULL nselem\n"); > + return E_NOTIMPL; > + } > + > + nsres = nsIDOMHTMLElement_QueryInterface(This->nselem, &IID_nsIDOMNSHTMLElement, (void**)&nselem); > + if(NS_SUCCEEDED(nsres)) { > + nsres = nsIDOMNSHTMLElement_GetScrollLeft(nselem, &left); > + nsIDOMNSHTMLElement_Release(nselem); > + if(NS_FAILED(nsres)) > + ERR("GetScrollLeft failed: %08x\n", nsres); > + }else { > + ERR("Could not get nsIDOMNSHTMLElement interface: %08x\n", nsres); > + } Printing an ERR is not a substitute for handling errors, you can't just continue as if nothing happened. Or if the error really can be legitimately ignored then you shouldn't have an ERR. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: Simplify tools/wrc a bit (RESEND)
by Alexandre Julliard Sept. 9, 2008

Sept. 9, 2008

Gerald Pfeifer <gerald(a)pfeifer.com> writes: > Any thoughts on this patch? It breaks the WANT_NEAR_INDICATION case. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: add minimal ctxtcall.h
by Rob Shearman Sept. 9, 2008

Sept. 9, 2008

2008/9/8 Louis. Lenders <xerox_xerox2000(a)yahoo.co.uk>: > +typedef HRESULT (*PFNCONTEXTCALL)(ComCallData *pparam); You don't explicitly tell widl that this function is a __stdcall function. Please check if that is automatically added in the generated header file. If not, you'll have to specify it here. > + > +/***************************************************************************** > + * IContextcallback interface > + */ Please change "IContextcallback" to "IContextCallback". -- Rob Shearman

1 0