January 2008 - Wine-devel - list.winehq.org

Re: shell32: Support getting the shell folder path for users other than the current user and the default user [resend]
by Alexandre Julliard 24 Jan '08

24 Jan '08

"Ge van Geldorp" <ge(a)thinstall.com> writes: > +static PSID _GetUserFromToken(HANDLE Token) > +{ > + char InfoBuffer[64]; > + PTOKEN_USER UserInfo; > + DWORD InfoSize; > + PSID Sid; > + DWORD SidSize; > + > + UserInfo = (PTOKEN_USER) InfoBuffer; > + if (! GetTokenInformation(Token, TokenUser, InfoBuffer, sizeof(InfoBuffer), > + &InfoSize)) > + { > + if (GetLastError() != ERROR_INSUFFICIENT_BUFFER) > + return NULL; > + UserInfo = HeapAlloc(GetProcessHeap(), 0, InfoSize); > + if (UserInfo == NULL) > + return NULL; > + if (! GetTokenInformation(Token, TokenUser, UserInfo, InfoSize, > + &InfoSize)) > + { > + HeapFree(GetProcessHeap(), 0, UserInfo); > + return NULL; > + } > + } > + > + SidSize = GetLengthSid(UserInfo->User.Sid); > + Sid = HeapAlloc(GetProcessHeap(), 0, SidSize); > + if (Sid == NULL) > + { > + if (UserInfo != (PTOKEN_USER) InfoBuffer) > + HeapFree(GetProcessHeap(), 0, UserInfo); > + return NULL; > + } > + > + if (! CopySid(SidSize, Sid, UserInfo->User.Sid)) > + { > + HeapFree(GetProcessHeap(), 0, Sid); > + Sid = NULL; > + } You should return a string directly, there's no reason to make a copy of the SID. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: PATCH - makedep, open in binary mode, move the end of file marker
by Alexandre Julliard 24 Jan '08

24 Jan '08

"Steven Edwards" <winehacker(a)gmail.com> writes: > As Julliard did not comment on this, I am assuming either it hit is > garbage filter because it was bad or because it was diffed from cvs. > Resending to hit the garbage filter for bad hacks...As discussed on > wine-devel, if its still not accepted then when I have some time I'll > look to seeing what needs to be done to teach MSYS about using windows > style line endings when doing checkouts with msys and cvs. Makefiles are text files, so they should be opened in text mode. If the mixed endings are confusing MSYS you could change makedep to rewrite the whole file instead of just appending. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: winex11.drv: try work around a crash
by Marcus Meissner 24 Jan '08

24 Jan '08

On Thu, Jan 24, 2008 at 09:00:32AM +0000, Louis. Lenders wrote: > > Hi, basically this is a resend of a patch i sent very long time ago ( http://www.winehq.org/pipermail/wine-patches/2007-April/037733.html ) > > It's a fix for bug http://bugs.winehq.org/show_bug.cgi?id=7380#c29 > > As explained in comment http://bugs.winehq.org/show_bug.cgi?id=7380#c29 > the cause of these crashes is some buggy mfc-function > , and as windows is able to work around this, wine should do the same. > > I've put the exception handler only around X11DRV_DIB_SetImageBits_24 , as that's where all the applications crash; if we want to mimic windows behaviour it should be put around all cases i guess, but i'm not sure we want to that and maybe hide other bugs. > > If this patch is rejected could you please comment why, to help fix the bug in a better way Remembering a different bug with the same issue I think you should not wrap DIB_SetImageBits_24 but just check descr->bits in X11DRV_DIB_SetImageBits() Something along the lines of: unsigned long off; unsigned char volatile x; __TRY { for (off=0;off<descr->sizeImage;off+=PAGE_SIZE) { x += ((unsigned char*)descr->bits)[off] } } ... Ciao, Marcus

1 1

Re: user32: Test destroying the cursor of a parent process.
by Dmitry Timoshkov 24 Jan '08

24 Jan '08

"Andrew Riedi" <andrewriedi(a)gmail.com> wrote: >I made some changes from the last time I talked to AJ - mainly exit code > stuff. > > changelog: user32: Test destroying the cursor of a parent process. Also there was a complaint that ok() call ersults in the child process are not included in the final statistics. There were no changes in that area in the patch. -- Dmitry.

1 0

Re: gitignore: added the missing include/devenum.h generated file.
by Reece Dunn 24 Jan '08

24 Jan '08

On 23/01/2008, Reece Dunn <msclrhd(a)googlemail.com> wrote: > The include/devenum.idl file does not have an entry in .gitignore to > prevent the generated file include/devenum.h from being included. This > patch fixes this issue. Please ignore this patch. - Reece

1 0

Re: credui: Full russian translation
by Kirill K. Smirnov 24 Jan '08

24 Jan '08

Hi, > + IDS_INCORRECTPASSWORDTITLE "Вход успешно произведён" The translation has opposite meaning than original. >+ IDS_INCORRECTPASSWORD "Убедитесь, что имя пользоватея\nи пароль верны." Typo in word "пользоватея" -- Kirill

1 0

Re: msxml3: Implement ownerDocument
by Alistair Leslie-Hughes 24 Jan '08

24 Jan '08

Dont apply this patch. "Alistair Leslie-Hughes" <leslie_alistair(a)hotmail.com> wrote in message news:fn6308$1gf$1(a)ger.gmane.org... > Hi, > > Changelog: > msxml3: Implement ownerDocument > > Best Regards > Alistair Leslie-Hughes > -------------------------------------------------------------------------------- > From 422440f49888d555c6031e3e40cafda0523c6ad8 Mon Sep 17 00:00:00 2001 > From: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> > Date: Wed, 23 Jan 2008 11:07:59 +1100 > Subject: [PATCH] Implement ownerDocument > To: wine-patches <wine-patches(a)winehq.org> > > --- > dlls/msxml3/node.c | 22 ++++++++++++++++++++-- > dlls/msxml3/tests/domdoc.c | 23 +++++++++++++++++++++-- > 2 files changed, 41 insertions(+), 4 deletions(-) > > diff --git a/dlls/msxml3/node.c b/dlls/msxml3/node.c > index 9d43c09..bc7a2a0 100644 > --- a/dlls/msxml3/node.c > +++ b/dlls/msxml3/node.c > @@ -602,8 +602,26 @@ static HRESULT WINAPI xmlnode_get_ownerDocument( > IXMLDOMNode *iface, > IXMLDOMDocument** DOMDocument) > { > - FIXME("\n"); > - return E_NOTIMPL; > + HRESULT hr = S_OK; > + xmlnode *This = impl_from_IXMLDOMNode( iface ); > + TRACE("\n"); > + > + if(!DOMDocument) > + return E_INVALIDARG; > + > + *DOMDocument = NULL; > + > + switch( This->node->type ) > + { > + case XML_DOCUMENT_NODE: > + hr = S_FALSE; > + break; > + default: > + *DOMDocument = (IXMLDOMDocument*)create_basic_node( > (xmlNodePtr)This->node->doc, NULL ); > > + break; > + } > + > + return hr; > } > > static HRESULT WINAPI xmlnode_cloneNode( > diff --git a/dlls/msxml3/tests/domdoc.c b/dlls/msxml3/tests/domdoc.c > index 4507d08..8802053 100644 > --- a/dlls/msxml3/tests/domdoc.c > +++ b/dlls/msxml3/tests/domdoc.c > @@ -687,10 +687,9 @@ static void test_domnode( void ) > { > owner = NULL; > r = IXMLDOMNode_get_ownerDocument( element, &owner ); > - todo_wine { > ok( r == S_OK, "get_ownerDocument return code\n"); > - } > ok( owner != doc, "get_ownerDocument return\n"); > + IXMLDOMDocument_Release(owner); > > type = NODE_INVALID; > r = IXMLDOMNode_get_nodeType( element, &type); > @@ -1869,6 +1868,7 @@ static void test_cloneNode(void ) > static void test_xmlTypes(void) > { > IXMLDOMDocument *doc = NULL; > + IXMLDOMDocument *docOwner = NULL; > IXMLDOMElement *pRoot; > HRESULT hr; > IXMLDOMComment *pComment; > @@ -1923,6 +1923,25 @@ static void test_xmlTypes(void) > ok(hr == S_OK, "ret %08x\n", hr ); > ok( !lstrcmpW( str, _bstr_("document") ), "incorrect nodeTypeString > string\n"); > SysFreeString(str); > + > + /* test ownerDocument */ > + hr = IXMLDOMDocument_get_ownerDocument(doc, NULL); > + ok(hr == E_INVALIDARG, "ret %08x\n", hr ); > + > + hr = IXMLDOMDocument_get_ownerDocument(doc, &docOwner); > + ok(hr == S_FALSE, "ret %08x\n", hr ); > + ok(docOwner == NULL, "ownerDocument return a value\n"); > + if(docOwner != NULL) > + { > + str = NULL; > + hr = IXMLDOMDocument_get_nodeName( docOwner, &str ); > + ok ( hr == S_OK, "get_nodeName wrong code\n"); > + ok ( str != NULL, "str is null\n"); > + ok( !lstrcmpW( str, szDocument ), "incorrect nodeName\n"); > + SysFreeString( str ); > + > + IXMLDOMDocument_Release(docOwner); > + } > > /* test implementation */ > hr = IXMLDOMDocument_get_implementation(doc, NULL); > -- > 1.5.3.7 > > -------------------------------------------------------------------------------- > >

1 0

help with typelib support
by Alistair Leslie-Hughes 24 Jan '08

24 Jan '08

Hi, This msxml bug requires the TYPELIB resource to be added. http://bugs.winehq.org/show_bug.cgi?id=11257 I modified the followng file as follows. Makefile.in IDL_TLB_SRCS = msxml3_v1.idl msxml3_v1.idl (new File, with a single line) #include "msxml2.idl" version.rc 1 TYPELIB LOADONCALL DISCARDABLE msxml3_v1.tlb The issue I am seing is the msxml3_v1.tlb is not being created. Output. ../../tools/widl/widl -I. -I. -I../../include -I../../include -I/usr/include/libxml2 -I/usr/include/libxml2 -D__WINESRC__ -DCOM_NO_WINDOWS_H -t -T msxml3_v1.tlb msxml3_v1.idl ../../tools/wrc/wrc --nostdinc -I. -I. -I../../include -I../../include -I/usr/include/libxml2 -I/usr/include/libxml2 -D__WINESRC__ -DCOM_NO_WINDOWS_H -foversion.res version.rc version.rc:26:47: Error: Cannot open file msxml3_v1.tlb Any ideas on what is wrong? Best Regards Alistair Leslie-Hughes

3 4

user32: listbox failures
by Reece Dunn 24 Jan '08

24 Jan '08

Running the listbox tests from dlls/user32/tests, I don't get any test failures on Vista. Can anyone else confirm this? Also, the LB_DIR failures on http://test.winehq.org/ are due to it looking for files in dlls/user32/tests as part of its test run. Since these are not available when running it as part of the test suite that generates those test results - or if you are running it as as standalone in another directory (e.g. building it on one machine, then running it on another) - the tests are failing. My thoughts on this is to create some dummy data used by the tests, then cleaning this up after the tests are run. This will all be self-contained in listbox.c. Does anyone have a better idea? If not, I'll look into this and provide a patch. - Reece

1 0

[msi/tests] Install tests on win9x and NT4
by Paul Vriens 23 Jan '08

23 Jan '08

Hi, Currently the install test fail/crash on at least win95 and NT4. The main reason I found was that pMsiQueryComponentStateA and pMsiSourceListGetInfoA are not guarded. Does it makes sense to guard every instance of these calls and just continue: if (pMsiSourceListGetInfoA) { call test outcome } or should we skip the whole subtest when the function is not available. Guarding each single item requires 28 of these if-statements or is there an other alternative? (like creating a function/macro to deal with this) Just guarding the functions makes the tests not crash but they still fail (that is another issue of course). -- Cheers, Paul.

3 3