- Wine-Devel - list.winehq.org

wingcc link problem
by Vitaly Lipatov Sept. 24, 2005

Sept. 24, 2005

Last days I can't compile any program with winegcc: $ winegcc buildtest.c a.out.exe-u5twQI.spec.o(.data+0x28): In function `__wine_spec_nt_header': : undefined reference to `__wine_spec_exe_entry' a.out.exe-u5twQI.spec.o(.init+0x1): undefined reference to `__wine_spec_init_ctor' collect2: ld returned 1 exit status winegcc: gcc failed. $ cat buildtest.c int main(int argc, char **argv) { exit(0); } Any suggestions? -- Vitaly Lipatov, ALT Linux Team Russia, Saint-Petersburg, www.etersoft.ru

1 0

Need help copying and painting with BitBlt
by James Hawkins Sept. 24, 2005

Sept. 24, 2005

Hey guys, I'm working on implementing the SizeBar in HTML Help, but I'm running into troubles with BitBlt. The expected behavior is as follows: * User left-clicks on the SizeBar window. WM_LBUTTONDOWN - Copy the contents of the SizeBar window to a bitmap. * User "drags" the SizeBar left or right. WM_MOUSEMOVE - Draw the bitmap containing the SizeBar contents with XOR so the old position is erased and the new position is drawn. * User releases the SizeBar. WM_LBUTTONUP - Release bitmap resources. - Resize internal windows. The only thing not working is the color of the moving SizeBar. The color in the bar is black with very few specks of white here and there. Also, when the bar passes over text, the parts of the bar with text under it turns white. Another problem is that the bar is drawn 23 pixels too high, so the bar starts at y=26 even though I specify y=49 in BitBlt. Right now I'm using a couple global variables temporarily until I can get the feature to work, but the following is the function that copies the bits from the SizeBar window into the bitmap: static void SB_CopySizeBar(HHInfo *pHHInfo) { HDC hdc = GetDC(pHHInfo->hwndSizeBar); RECT rc; hdcMem = CreateCompatibleDC(hdc); SB_GetSizeBarRect(pHHInfo, &rc); hbm = CreateCompatibleBitmap(hdc, rc.right, rc.bottom); SelectObject(hdcMem, hbm); BitBlt(hdcMem, rc.left, rc.top, rc.right, rc.bottom, hdc, 0, 0, SRCCOPY); } When I paint the bar I use: BitBlt(hdc, rc.left, rc.top, rc.right, rc.bottom, hdcMem, 0, 0, SRCINVERT); I've attached a screen shot of what it should look like, and what it actually looks like. If anyone has any ideas about getting this to work, I would greatly appreciate it. Thanks, James Hawkins

1 0

Re: XEmbed Systray Support
by Alexandre Julliard Sept. 23, 2005

Sept. 23, 2005

Robert Shearman <rob(a)codeweavers.com> writes: > + ev.xclient.type = ClientMessage; > + ev.xclient.window = systray_window; > + ev.xclient.message_type = x11drv_atom(_NET_SYSTEM_TRAY_OPCODE); > + ev.xclient.format = 32; > + ev.xclient.data.l[0] = CurrentTime; > + ev.xclient.data.l[1] = SYSTEM_TRAY_REQUEST_DOCK; > + ev.xclient.data.l[2] = data->whole_window; > + XSendEvent( display, systray_window, False, NoEventMask, &ev ); > + XSync( display, False ); You need to get the tsx11 lock when calling X functions. Also what's the reason for having an XSync here? -- Alexandre Julliard julliard(a)winehq.org

3 4

(no subject)
by dank＠kegel.com Sept. 23, 2005

Sept. 23, 2005

Dimi wrote: >> http://kegel.com/wine/qa > >Good page. Thanks! >I'm afraid that it will not get enough >exposure up on your site though, Don't underestimate the power of search engines to find obscure content... but I agree. >it seems to me >it would fit perfectly into our Wiki. Ditto for the >other (excellent) Wine pages you have there. Sorry, I'm allergic to wikis. I much prefer editing in vi. I'd be happy to add it to the winehq.org web site once I've gotten it up to snuff. [ If the wiki stored its files in the same cvs archive as winehq.org does, we'd have the best of both worlds... wonder why no wikis are set up like that.] By the way, do you happen to know how one gets permission to modify bugzilla reports? That's kind of needed so the more experienced bug triage volunteers can add keywords, etc., and I'd like to add that info to my page. - Dan

1 0

memory leak detection patch
by Mike McCormack Sept. 23, 2005

Sept. 23, 2005

Hi, This patch is written for MSI, but could be adapted to other code. It keeps a linked list of allocated memory and removes free'd memory from the list. It requires that you change all the HeapAlloc's or mallocs in a dll to msi_alloc (for example), but can detect unfree'd memory and invalid free's. I have discovered quite a few problems in my own code with this patch, so hopefully somebody else will find it useful too. Mike

4 4

oh, and a question about old config options
by Robbert Xerox Sept. 23, 2005

Sept. 23, 2005

Hi, i also would like to ask the following question: I've sent in a patch to add an option to run applications in Managed mode (or not). My question is : what is the status of this option (and others from the old config file) in current wine: is there a policy to get rid of these options, or will they remain in wine. ATM this option still helps some apps (like winamp) to run in a better way. Furthermore: will there be an option added to winecfg where you can set the debug options (like RelayExclude) or do we have to do this for eternity in the registry now (which i find rather annoying) At least: what i as i user would find very handy is some kind of environment variable where you can set the searchpath for native dlls. Again, atm this has to be done in the registry, but a more userfriendly way would be better i think. Any thoughts? ___________________________________________________________ To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com

1 0

ntsepkg.h and security.h missed in wine
by Peter Lemenkov Sept. 23, 2005

Sept. 23, 2005

Hello, All! I'm trying to compile Miranda-IM with Winelib (main module already compiled) and find oud that these two necessary headers are missed in Wine-includes. In order to compile miranda-im, I took 'em from MinGW win32api distribution: http://cvs.sourceforge.net/viewcvs.py/*checkout*/mingw/w32api/include/ntsec… http://cvs.sourceforge.net/viewcvs.py/*checkout*/mingw/w32api/include/secur… Are there any ways to officially include these files into Wine? -- With best regards, Peter Lemenkov.

3 2

Fwd: [RESEND] WININET: use ICU_BROWSER_MODE in InternetCanonicalizeUrl
by Denis Lukianov Sept. 23, 2005

Sept. 23, 2005

Anything wrong with my patch? It hasn't been accepted Begin forwarded message: > From: Denis Lukianov <denis(a)voxelsoft.com> > Date: 22 September 2005 03:06:12 BDT > To: wine-patches(a)winehq.org > Subject: [RESEND] WININET: use ICU_BROWSER_MODE in > InternetCanonicalizeUrl > Reply-To: wine-devel(a)winehq.org > > > Fixes bug 3319. > > =================================================================== > RCS file: dlls/wininet/internet.c,v > retrieving revision 1.1 > diff -u -r1.1 dlls/wininet/internet.c > --- dlls/wininet/internet.c 2005/09/18 11:17:09 1.1 > +++ dlls/wininet/internet.c 2005/09/22 01:55:14 > @@ -1514,6 +1514,12 @@ > dwURLFlags |= URL_UNESCAPE; > dwFlags &= ~ICU_ESCAPE; > } > + > + if(dwFlags & ICU_BROWSER_MODE) > + { > + dwURLFlags |= URL_BROWSER_MODE; > + dwFlags &= ~ICU_BROWSER_MODE; > + } > if(dwFlags) > FIXME("Unhandled flags 0x%08lx\n", dwFlags); > TRACE("%s %p %p %08lx\n", debugstr_a(lpszUrl), lpszBuffer, > @@ -1553,6 +1559,12 @@ > dwURLFlags |= URL_UNESCAPE; > dwFlags &= ~ICU_ESCAPE; > } > + > + if(dwFlags & ICU_BROWSER_MODE) > + { > + dwURLFlags |= URL_BROWSER_MODE; > + dwFlags &= ~ICU_BROWSER_MODE; > + } > if(dwFlags) > FIXME("Unhandled flags 0x%08lx\n", dwFlags); > TRACE("%s %p %p %08lx\n", debugstr_w(lpszUrl), lpszBuffer, > >

1 0

Re: OLEAUT32: add full multiplication/division support for DECIMAL (in small patches)
by Dimi Paun Sept. 23, 2005

Sept. 23, 2005

From: "Alex Villacís Lasso" <a_villacis(a)palosanto.com> > By popular request, I am resending the implementation for DECIMAL > multiplication/division in small patches. Good, but please send one patch per email. -- Dimi Paun <dimi(a)lattica.com> Lattica, Inc.

2 1

[DOCS, POLICY] Documenting stuff that doesn't work right?
by Holly Bostick Sept. 23, 2005

Sept. 23, 2005

Hey, So I just blew away my previous install of Wine and reinstalled from yesterday's CVS, and went to install a program, so I could take notes on the most current user experience for the docs. I immediately realized that I hadn't run winecfg to set my drives, so I went to do that-- and that's where I ran into this problem. I was installing Icewind Dale, and like all games and Wine-installed programs, I meant to install to the partition I have set aside for that purpose, which is mounted to /usr/local/games, with a symlink in my home directory. So naturally I usually set this partition to be a Wine drive, and I prefer to use d: as the browse dialog naturally defaults to the c: drive and it's faster to change if I only have to change to the next drive. When I went to the 'Drives' tab in winecfg, I had c: (of course) e: (tmp) f: (my home directory) z: (/) You will notice that no removeable media device appears. So I go to add a drive, which is automatically called D: , as you would expect, and I set it to point to /usr/local/games, and I also add G: and set it to point to /media/cdrecorder. Hit apply and all appears well (no errors or anything of that sort). Run the IWD install again and get to the installation location section, and G appears in the browser, but D does not. Cancel and back to winecfg to try again. After several attempts, I finally opened up a file manager and went to ~/.wine/dosdevices and looked at what was there. To my surprise, there was already a link to d:, which was set to /cdrom (which is an invalid location for me anyway since I use hal, which puts stuff in /media). Since I had already set the correct mount point of my DVD drive to the g: link, I just changed d: to my preferred location of /usr/local/games, and was able to complete the install to the d:\iwd directory as I had planned. So the problem is that an automatically created (but wrongly targeted, in these days of hal/dbus) link is created to the cdrom device does not appear in the Winecfg drive list (unless you do autodetect, which may well add additional, unwanted drives to the list that the user would have to remove), making a drive letter that 'seems' to be available in fact unavailable for use (the reason that I couldn't set /usr/local/games to d was because d already existed and couldn't apparently be changed in its invisibility), and also causing the user to think that no drive letter is set for the CD/DVD device, so that they create an unnecessary second link. Obviously I know how to get around this, having just done so, but is this appropriate to document? Or is it likely to be fixed before the updated docs go 'live'? I only ask because I can't think of a way to write this up atm without explicitly exposing a flaw in Wine, and Wine doesn't really deserve that-- even winecfg doesn't deserve that, being much improved from even last month's release (which improvement is obvious, even though I've only done this single operation). I mean, I'll take notes on everything, but this looks like a policy question to me. Holly

3 4