wine-devel August 2005

wine-devel@winehq.org

132 participants
254 discussions

Re: Standardize the DllGetClassObject prototypes
by Robert Shearman 05 Aug '05

05 Aug '05

Francois Gouget wrote: > > According to MSDN the prototype of the DllGetClassObject() functions is: > > HRESULT WINAPI DllGetClassObject(REFCLSID pclsid, REFIID piid, LPVOID > *ppv); > > > But in Wine it has sometimes varied a bit, like returning a DWROD > instead of an HRESULT. Nothing too serious but I like order so I went > through them and made them all the same. > > There's one which is more seriously wrong though: > MSXML_DllGetClassObject() is missing the WINAPI. This cause > winapi_check to issue an error and is how I got into this. That wouldn't have happened if the function had been called DllGetClassObject instead of MSXML_DllGetClassObject. Is the uniqueness of the name really more important than the compiler being able to typecheck it? Can we please not introduce any more of the Dll* functions with DLLNAME_ prefixes? -- Rob Shearman

1 0

Re: msxml3: protect DOMElement implementation with HAVE_LIBXML2
by Mike McCormack 05 Aug '05

05 Aug '05

Michael Jung wrote: > This fixes msxml3 compilation on debian sarge, if libxml2-dev headers are not > installed. I've imitated Mike's scheme from domdoc.c. Thanks for fixing this. > +HRESULT DOMElement_create( IXMLDOMElement** DOMElement, void* xmldoc ) You shouldn't need this stub, because the function will only be used if HAVE_LIBXML2 is defined in domdoc.c Mike

1 0

Re: [DSHOW] Beginning of winestrmbase lib (resend)
by Dimi Paun 05 Aug '05

05 Aug '05

From: "Christian Costa" <titan.costa(a)wanadoo.fr> > Add winestrmbase library and make quartz use it. If this is wine specific, and it's supposed to be streambase library, I think winestreambase is maybe a clearer name, winestrmbase makes it sound like a string processing lib. :) -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

2 3

Re: Patch to fix GetCommState "Invalid Argument" errors with Linux usbserial devices
by Mike McCormack 05 Aug '05

05 Aug '05

James Liggett wrote: > This patch fixes an "Invalid Argument" error in GetCommState when a > program tries to get information about a usbserial device. With this > patch, Wine provides a "fake" serial device control block to a Windows > program requesting it, such that the program believes that it's > interfacing with an actual serial port, since usbserial devices don't > currently acknowledge traditional DCBs via tcgetattr/tcsetattr. Using > this patch I was able to synchronize a serial-based Palm PDA with Palm > Desktop on Wine with a pl2303 USB-serial converter. I'm also told that > this patch also works with FTDI devices. Hi James, I don't think this is quite right. The pl2303 is a standard USB-serial converter, and it should have a selectable baud rate (according to the Linux kernel's driver for it). It has a db-9 connector on one end, and a usb connector on the other, right? I'm also pretty sure that all serial ports are ttys in Linux, so your if(isatty()) check will always succeed. Have you tried writing a small test program that does GetCommState and SetCommState on that serial device in Windows? If it fails to set the baud rate perhaps there's an error code returned? Mike

2 1

Re: ntdll: Fix raw CD-ROM access on Linux
by Jesse Allen 04 Aug '05

04 Aug '05

On 8/3/05, Tobias Gruetzmacher <tobias(a)portfolio16.de> wrote: > Hi, > > after submitting my bugreport > http://bugs.winehq.com/show_bug.cgi?id=3175 I curefully checked the > transfer directions in dlls/ntdll/cdrom.c. I think, they are reversed at > two places. > > The attached patch fixes my reported bug. This may also fix bug 3143. > It does fix the bug. Actually, both games use SIntfNT.dll, so we got a duplicate bug report. Thanks for tracking down the issue. Jesse

1 0

compile error in current cvs
by Stefan Leichter 04 Aug '05

04 Aug '05

Hello, current cvs does not compile for me. Im on SuSE 9.0 Bye Stefan make[2]: Entering directory `/usr/src/wine/wine-build/dlls/msxml3' gcc -c -I../../../wine/dlls/msxml3 -I. -I../../../wine/include -I../../include -D__WINESRC__ -DCOM_NO_WINDOWS_H -D_REENTRANT -fPIC -Wall -pipe -mpreferred-stack-boundary=2 -fno-strict-aliasing -gstabs+ -Wpointer-arith -g -O2 -o domdoc.o ../../../wine/dlls/msxml3/domdoc.c In file included from ../../../wine/dlls/msxml3/domdoc.c:36: ../../../wine/dlls/msxml3/msxml_private.h:26:27: libxml/parser.h: Datei oder Verzeichnis nicht gefunden In file included from ../../../wine/dlls/msxml3/domdoc.c:36: ../../../wine/dlls/msxml3/msxml_private.h:28: error: parse error before "xmlDocPtr" make[2]: *** [domdoc.o] Fehler 1 make[2]: Leaving directory `/usr/src/wine/wine-build/dlls/msxml3' make[1]: *** [msxml3] Fehler 2 make[1]: Leaving directory `/usr/src/wine/wine-build/dlls' make: *** [dlls] Fehler 2

4 4

Re: secur32: add first SSPI test to actually see what work needs to be done
by Kai Blin 04 Aug '05

04 Aug '05

* Kai Blin <blin(a)gmx.net> [03/08/05, 02:03:27]: > + sec_status = QuerySecurityPackageInfo( sec_pkg_name, &pkg_info); > + > + ok((long)sec_status >= 0L, "Return value of QuerySecurityPackageInfo should be >= 0, but is %ld\n", (long)sec_status); > + Never mind that. After running the tests on Win32, I realized I needed to check for SEC_E_OK, not 0. I'll resubmit the patch. Kai -- Kai Blin, (blin at gmx dot net) "It might help if we ran the MBA's out of Washington." -- Admiral Grace Hopper

2 3

Re: Fix Using DDraw to Access the Desktop
by Christian Costa 04 Aug '05

04 Aug '05

Hi Rob (and Lionel), While debugging a game where there is the same problem, I did exactly the same fix. However, although this works, it is horribly slow. Do you have the same issue? Maybe this is the reason this was commented out. I wonder if SYNC_UPDATE and OWN_WINDOW stuff has been written to workaround this. There is not comment about that and the code is from transgaming and rather old so I don't know. Lionel do you remember something about that? Robert Shearman wrote: > Hi, > > Previously, using ddraw to write to a small area of the desktop (or > even none at all) would cause the screen to go black because the > contents of the screen wasn't copied to the surface before the surface > was copied back to the screen. This patch fixes that. > > Changelog: > Copy the current contents of the DC to the surface on locking so that > when it is used to write to the desktop then the previous contents > plus the updates are correctly written out. > >------------------------------------------------------------------------ > >Index: dlls/ddraw/surface_user.c >=================================================================== >RCS file: /home/wine/wine/dlls/ddraw/surface_user.c,v >retrieving revision 1.1 >diff -u -p -r1.1 surface_user.c >--- dlls/ddraw/surface_user.c 5 Jun 2005 17:50:49 -0000 1.1 >+++ dlls/ddraw/surface_user.c 25 Jul 2005 14:45:32 -0000 >@@ -47,6 +47,7 @@ static void User_destroy_own_window(IDir > static DWORD CALLBACK User_update_thread(LPVOID); > #endif > static void User_copy_to_screen(IDirectDrawSurfaceImpl* This, LPCRECT rc); >+static void User_copy_from_screen(IDirectDrawSurfaceImpl* This, LPCRECT rc); > > static HWND get_display_window(IDirectDrawSurfaceImpl* This, LPPOINT pt); > >@@ -224,10 +225,8 @@ static void User_DirectDrawSurface_wait_ > void User_DirectDrawSurface_lock_update(IDirectDrawSurfaceImpl* This, > LPCRECT pRect, DWORD dwFlags) > { >-#if 0 > if (!(dwFlags & DDLOCK_WRITEONLY)) > User_copy_from_screen(This, pRect); >-#endif > if (dwFlags & DDLOCK_WAIT) User_DirectDrawSurface_wait_update(This); > > if (pRect) { >@@ -594,7 +593,6 @@ static void User_copy_to_screen(IDirectD > } > } > >-#if 0 > static void User_copy_from_screen(IDirectDrawSurfaceImpl* This, LPCRECT rc) > { > if (This->surface_desc.ddsCaps.dwCaps & DDSCAPS_PRIMARYSURFACE) >@@ -603,6 +601,13 @@ static void User_copy_from_screen(IDirec > HWND hDisplayWnd = get_display_window(This, &offset); > HDC hDisplayDC = GetDC(hDisplayWnd); > RECT drawrect; >+ HDC hSurfaceDC; >+ >+ if (FAILED(This->get_dc(This, &hSurfaceDC))) >+ { >+ ReleaseDC(hDisplayWnd, hDisplayDC); >+ return; >+ } > > drawrect.left = 0; > drawrect.right = This->surface_desc.dwWidth; >@@ -611,7 +616,7 @@ static void User_copy_from_screen(IDirec > if (rc) > IntersectRect(&drawrect,&drawrect,rc); > >- BitBlt(This->hDC, >+ BitBlt(hSurfaceDC, > drawrect.left, drawrect.top, > drawrect.right-drawrect.left, drawrect.bottom-drawrect.top, > hDisplayDC, >@@ -621,7 +626,6 @@ static void User_copy_from_screen(IDirec > ReleaseDC(hDisplayWnd, hDisplayDC); > } > } >-#endif > > static const IDirectDrawSurface7Vtbl User_IDirectDrawSurface7_VTable = > { > >

1 0

threats from MS over Wine's CD/DVD recognition??
by johan kdks 04 Aug '05

04 Aug '05

My friend has a Linux PC and when he inserts a CD or DVD, it popups up the cover art automatically and displays all sorts of extended data, like the actors in the film, the studio that released the cd, etc. And we tried 100 or so discs and it recognized every one perfectly, just as good as Windows XP Media Center does, and it even seems to cache the entries since the 2nd time you insert the disc the recognition is instant (first time takes a few seconds). None of the other Linux media players I've used do this--they all use freedb which doesn't work for DVD's, doesn't have extended attributes, and doesn't display any cover art. I've been poking around all day to figure out how it's done. Someone said it seems this Linux project is pulling the data from Microsoft's Windows media server, like XP MC. If that's true, I'm sure Bill isn't too happy. But anyway I figured that code must have been lifted from Wine--part of your "emulation". But I've been digging into Wine and it seems Wine doesn't do this either, at least not anymore. In fact it's conspicuously missing since Wine is a Windows emulator, and under Windows you put a movie in the drive and see the cover art and all the info on the disc, and with Wine you don't. Was this in Wine at some point, and then got yanked because of threats from Microsoft? I actually see no reference to it anywhere. If it's another one of those modules like libdvdcss that every needs to have, but nobody wants to host, it's buried so far underground I can't find it. If Plutohome.org is going to have to pull it soon, somebody should mirror that stuff soon so it doesn't get lost. -- _______________________________________________ NEW! Lycos Dating Search. The only place to search multiple dating sites at once. http://datingsearch.lycos.com

3 2

ownerdraw paint recursion in listbox
by Alexander Yaworsky 04 Aug '05

04 Aug '05

Hello, recently i tried to install some application and it hung when i tried to select options. It uses listbox with ownerdraw items with checkboxes. When listbox is initially painted everything is ok. But when i try to select other item, an extra WM_PAINT is sent to listbox when application draws item. This causes infinite recursion. I've been implemented a simple workaround adding boolean to listbox structure but i have doubts. I briefly looked at some other controls and had not noticed similar countermeasures. Have somebody ever seen such behaviour? It would be interesting to test other controls sending WM_PAINT from ownerdraw handler but i'm short of time to do it soon. Should this workaround implemented in this way or something is wrong deeper with WM_PAINT handling? Thanks.

2 7

← Newer
1
...
19
20
21
22
23
24
25
26
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

wine-devel August 2005