March 2007 - Wine-devel - list.winehq.org

Problem with program
by unknown＠example.com 22 Mar '07

22 Mar '07

Using wine "memoriesontv" it does not show the images and the generation of video interrupts

2 1

Re: SoC idea: add BITS support
by Juan Lang 22 Mar '07

22 Mar '07

> Anyway, I don't know what shape our iphlpapi is in, but I think it > should be possible to at least get the basic features working in the > context of a SoC project. Okay, I'll bite. I looked at the BITS docs too, and it uses GetBestInterface to pick an interface, and some interface stats to determine how busy an interface is. Our iphlpapi should be up to the task (but maybe that's just a parent's pride ;) ) --Juan ____________________________________________________________________________________ No need to miss a message. Get email on-the-go with Yahoo! Mail for Mobile. Get started. http://mobile.yahoo.com/mail

2 1

Sugestion: A few changes to the
by Giles Cameron 22 Mar '07

22 Mar '07

After learning about the wine debuging output's level system, and seeing it be used for a while, I think its time we make a couple of new channels. They would be like the 'trace' channel, only for more spefic purposes, these could by default be compiled out, but when needed put in. Their primary purpos would be for tracking down bugs in the API, ones that may not be as easy to find in a test case. Any thoughts? Also, if more channels is not the way to go, maybe per-function filtering? Thanks for the time, Giles Cameron

4 5

Re: dplayx: corrected the way nameserver initializes the sessiondesc structure on broadcast request reply
by Kai Blin 22 Mar '07

22 Mar '07

On Thursday 22 March 2007 17:20, Alessandro Pignotti wrote: > Whitout this patch the directplay nameserver does not fully copy the > Session Description structure when replying to session request broadcast Good catch. Kai -- Kai Blin, <kai Dot blin At gmail Dot com> WorldForge developer http://www.worldforge.org/ Wine developer http://wiki.winehq.org/KaiBlin/ -- Will code for cotton.

1 0

Re: [netapi32/tests] Use GetModuleHandle and skip
by Paul Vriens 22 Mar '07

22 Mar '07

Paul Vriens wrote: > Hi, > > netapi32 is already linked to. > > Changelog > Use GetModuleHandle and skip > > Cheers, > > Paul. > > > ------------------------------------------------------------------------ > >>From 6595fe64b93f2a81b4f6cbf77df58b2ef67c29ae Mon Sep 17 00:00:00 2001 > From: Paul Vriens <Paul.Vriens.Wine(a)gmail.com> > Date: Fri, 16 Mar 2007 12:00:08 +0100 > Subject: [PATCH] Use GetModuleHandle and skip > > --- > dlls/netapi32/tests/ds.c | 7 ++----- > 1 files changed, 2 insertions(+), 5 deletions(-) > > diff --git a/dlls/netapi32/tests/ds.c b/dlls/netapi32/tests/ds.c > index 7d0413d..7ee8a73 100644 > --- a/dlls/netapi32/tests/ds.c > +++ b/dlls/netapi32/tests/ds.c > @@ -78,15 +78,12 @@ START_TEST(ds) > { > HMODULE hnetapi32; > > - hnetapi32 = LoadLibraryA("netapi32.dll"); > - if (!hnetapi32) > - return; > + hnetapi32 = GetModuleHandleA("netapi32.dll"); > > - SetLastError(0xdeadbeef); > pDsRoleGetPrimaryDomainInformation=(void*)GetProcAddress(hnetapi32,"DsRoleGetPrimaryDomainInformation"); > if (!pDsRoleGetPrimaryDomainInformation) > { > - trace("DsRoleGetPrimaryDomainInformation not implemented : (%d), stop testing\n", GetLastError()); > + skip("DsRoleGetPrimaryDomainInformation is not available\n"); > return; > } > pDsRoleFreeMemory=(void*)GetProcAddress(hnetapi32,"DsRoleFreeMemory"); Hi, this patch is already committed but something is wrong. For some reason GetModuleHandleA doesn't return a handle. Does anyone know why? The Makefile.in contains: IMPORTS = netapi32 advapi32 kernel32 Cheers, Paul.

1 0

x11drv: XDND: Fix file drop to properly support file:/// URIs (debugging problem)
by Maciej Katafiasz 22 Mar '07

22 Mar '07

Hi, (please ignore the other copy I sent from the wrong address by accident) I'm working on a bug in winex11drv/xdnd.c, where file:/// URIs are not correctly interpreted, resulting in DnD broken most of the time. There are two parts of that bug: 1) X11DRV_XDND_BuildDropFiles() fails to properly handle absolute unix paths, which is exactly what you get from file:///, so it won't find the file unless $PWD and the constructed Windows drive-absolute path happen to have the same root. For example, file:///home/mathrick/foo.txt will yield "\\home\\mathrick\\foo.txt" and work only if the current directory is on a drive letter on which that path is valid. 2) X11DRV_XDND_DeconstructTextPlain() fails to accommodate for the fact file:/// URIs are, in fact, URIs and not paths, and doesn't check for validity nor does it unescape % sequences. The result is that paths with spaces, etc. in them won't work. The attached patch is my first take at 1), it works, but there is a problem. My test application is μTorrent, and it'll randomly lock up with the following errors printed: err:ntdll:RtlpWaitForCriticalSection section 0x110020 "heap.c: main process heap section" wait timed out in thread 001f, blocked by 0019, retrying (60 sec) err:ntdll:RtlpWaitForCriticalSection section 0x110020 "heap.c: main process heap section" wait timed out in thread 001e, blocked by 0019, retrying (60 sec) I can reproduce them by attempting to drop a file on µT's window right after startup. If I wait a while, it works. It also happens if I switch to another virtual desktop, then back to the one µT is on and then try to drop. I tried to check the code as well as I could (win32 API is not exactly my everyday environment), and I believe it's correct. At least all the allocations appear to be correct, nothing is overwritten. I've tried debugging in several ways using winedbg, but it fails pretty hard with various exceptions that aren't there when running outside the debugger, or fails to attach to the process, etc. From the circumstances, I believe there's some kind of race condition in allocation from process heap. However, I'm unable to find it as I have no idea about wine internals and how exactly DnD happens, and what threads do what and when. I'd be therefore very grateful if someone knowledgeable would take a look at it and see what's wrong, or at least help me with debugging the issue, as my ideas how to attack it have run out. Cheers, -- Being really good at C++ is like being really good at using rocks to sharpen sticks. (Thant Tessman) -------- Maciej Katafiasz <ml(a)mathrick.org> http://mathrick.org

5 5

Fwd: kernel32: Implement ReplaceFileA/ReplaceFileW (revised)
by Erich Hoover 22 Mar '07

22 Mar '07

Request for comments, as per http://www.winehq.com/site/sending_patches. Thanks! Erich Hoover ehoover(a)mines.edu ---------- Forwarded message ---------- From: Erich Hoover <ehoover(a)mines.edu> Date: Mar 3, 2007 2:37 PM Subject: kernel32: Implement ReplaceFileA/ReplaceFileW (revised) To: wine-patches(a)winehq.org Real Name: Erich Hoover Description: Implements the functions ReplaceFileA and ReplaceFileW in kernel32 (Bug #7544). This revision includes a lot of changes as a result of feedback from Felix, see wine-devel for details. Also provides conformance test code to ensure proper functionality. Changelog: kernel32: Implement ReplaceFileA/ReplaceFileW

2 2

Regedit output to STDOUT / force registry files sync
by Vit Hrachovy 22 Mar '07

22 Mar '07

Hi, I'm trying to find some elegant method to access registry keys added during one WINE session, i.e. without restarting WINE. Test case description: 1. Open WINE session wineconsole --backend=user cmd 2. During session, insert some new registry key regedit new-entry.reg 3. Try to access the new registry entry from bash cat ~/.wine/system.reg |grep AutoHotkey (fails) 4. Close WINE session I've found there was at times an option to save registry periodically, but this can't help me here. This problem can be solved in two ways, both of them can be helpful: a) Create small program to force sync of registry in memory with registry files - 'wineflushregistry'. b) Enhance regedit to be able to output to STDOUT. By default registry search output is done to a specified file. It can be redirected to STDERR, though. (tested on 0.9.29, 0.9.33) Enhancing regedit ----------------- regedit /es "HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\App Paths\AutoHotkey.exe" which will be similar to (non-functional) regedit /e /dev/stdout "HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\App Paths\AutoHotkey.exe" Regards Vit Hrachovy

4 11

Deadline for Summer of Code is March 26th!
by Dan Kegel 22 Mar '07

22 Mar '07

Watch out, those deadlines can creep up on you! If you're a student interested in working on Wine as part of Google's Summer of Code, you have only five or so more days to apply. See http://code.google.com/soc or http://groups.google.com/group/google-summer-of-code-announce/web/guide-to-… for more details. And see http://wiki.winehq.org/SummerOfCode for a few ideas. Feel free to post your ideas to wine-devel first for quick feedback. So far, there are only two applications, both of which look pretty good. We will probably have more than two slots, so it's wide open right now! Sharpen your pencils, and good luck! - Dan

2 2

SoC idea: pressure sensitive tablet support
by Dan Kegel 22 Mar '07

22 Mar '07

While looking around for a project idea, I noticed that people are starting to ask for pressure sensitive tablet support in Photoshop on Wine. Seems like it could be a fun project for some Summer of Code student. - Dan

1 0