wine-devel

wine-devel@winehq.org

84463 discussions

ASIO and wine
by Tom Hibbert 22 Sep '02

22 Sep '02

Hello, I am very interested in creating a home studio system with Linux. I get 1/4 the latency i do using Windows with the same hardware.. there is more control.. etc etc.. there are many reasons why I want to do this. I would especially like to use Cubase SX. The main barrier hampering this (besides copy protection ;) is the lack of ASIO support under Wine. Secondly, timing issues, but that is easy to tweak by the user. I am wondering if there are any progects or plans to implement ASIO in Wine/WineX. If so, I would like to contribute. If not, I will have to go to university and study computer science as well as electronic music. ;) Oh, yes - i know there are a lot of quality open source audio products (Audacity, Psycle, Csound, MAX/MSP come to mind...) I am mainly interested in seeing some VST implementation on Linux because VST is a very widely used and mature base for software synthesisers and effects. I'm told the only barrier to a NATIVE VST-capable app (Yes, the code is all there in the Audacity sources AFAIK) is the porting of a library known as libvstgui... why not run this through Winelib is the next thing i am going to ask the developers of that particular product :) Tom Hibbert Work: someday Home: +649 815 1838 Obtain my GPG Key from: http://cryptocracy.hn.org/pubkey.txt

2 2

Re: fix listview trace crash
by Dimitrie O. Paun 22 Sep '02

22 Sep '02

On September 21, 2002 02:46 am, Rein Klazes wrote: > Hi, > > With these fixes the original reason to debug disappeared as well ;-) > > dlls/comctl32/ : listview.c > Fix some crashes running with --debugmsg +listview > + ZeroMemory(&item, sizeof(item)); We shouldn't need these. In fact, they just hide bugs, and most of them are superfluous anyways. The first hunk is a good find though, I've integrated it in my patch. Alexandre, please use this one instead. ChangeLog -- Fix outputing of a text pointer in debug mode (by Rein Klazes) -- Remove a large bunch of superfluous ZeroMemory calls -- Fix couple of bugs Index: dlls/comctl32/listview.c =================================================================== RCS file: /var/cvs/wine/dlls/comctl32/listview.c,v retrieving revision 1.148 diff -u -r1.148 listview.c --- dlls/comctl32/listview.c 18 Sep 2002 18:29:28 -0000 1.148 +++ dlls/comctl32/listview.c 21 Sep 2002 15:08:20 -0000 @@ -337,7 +337,7 @@ wstr = HeapAlloc(GetProcessHeap(), 0, len * sizeof(WCHAR)); if (wstr) MultiByteToWideChar(CP_ACP, 0, (LPCSTR)text, -1, wstr, len); } - TRACE(" wstr=%s\n", debugstr_w(wstr)); + TRACE(" wstr=%s\n", text == LPSTR_TEXTCALLBACKW ? "(callback)" : debugstr_w(wstr)); return wstr; } @@ -423,7 +423,7 @@ " pszText=%s, cchTextMax=%d, iImage=%d, lParam=%lx, iIndent=%d}", lpLVItem->mask, lpLVItem->iItem, lpLVItem->iSubItem, lpLVItem->state, lpLVItem->stateMask, - debugtext_tn(lpLVItem->pszText, isW, 80), + lpLVItem->mask & LVIF_TEXT ? debugtext_tn(lpLVItem->pszText, isW, 80) : 0, lpLVItem->cchTextMax, lpLVItem->iImage, lpLVItem->lParam, lpLVItem->iIndent); return buf; @@ -605,6 +605,7 @@ BOOL bReturn; item.iItem = iItem; + item.iSubItem = 0; item.mask = LVIF_PARAM; LISTVIEW_GetItemT(infoPtr, &item, TRUE, TRUE); @@ -865,7 +866,6 @@ } /* get item */ - ZeroMemory(&item, sizeof(item)); item.mask = LVIF_TEXT; item.iItem = idx; item.iSubItem = 0; @@ -1557,7 +1557,6 @@ if (adj_sel_only) return TRUE; /* set the selection on items */ - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.state = LVIS_SELECTED; lvItem.stateMask = LVIS_SELECTED; for(i = lower; i <= upper; i++) @@ -1576,7 +1575,7 @@ LVITEMW lvItem; INT index, i; - ZeroMemory(&lvItem, sizeof(lvItem)); + lvItem.state = 0; lvItem.stateMask = LVIS_SELECTED; remsel.lower = lower; @@ -1734,7 +1733,6 @@ INT oldFocus = infoPtr->nFocusedItem; LVITEMW lvItem; - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.state = LVIS_FOCUSED; lvItem.stateMask = LVIS_FOCUSED; LISTVIEW_SetItemState(infoPtr, nItem, &lvItem); @@ -1830,8 +1828,8 @@ if (nFirst == -1) nFirst = nItem; - item.stateMask = LVIS_SELECTED; item.state = LVIS_SELECTED; + item.stateMask = LVIS_SELECTED; /* FIXME: this is not correct LVS_OWNERDATA * See docu for LVN_ITEMCHANGED. Is there something similar for @@ -1889,8 +1887,6 @@ POINT ptItem; RECT rcSel; - item.stateMask = LVIS_SELECTED; - if ((uView == LVS_LIST) || (uView == LVS_REPORT)) { if (infoPtr->nSelectionMark == -1) @@ -1910,6 +1906,8 @@ nFirst = nLast = -1; } + item.stateMask = LVIS_SELECTED; + for (i = 0; i <= GETITEMCOUNT(infoPtr); i++) { if (nFirst > -1) @@ -2397,8 +2395,6 @@ lpSubItem = (LISTVIEW_SUBITEM *)COMCTL32_Alloc(sizeof(LISTVIEW_SUBITEM)); if (!lpSubItem) return FALSE; - ZeroMemory(lpSubItem, sizeof(LISTVIEW_SUBITEM)); - /* we could binary search here, if need be...*/ for (i = 1; i < hdpaSubItems->nItemCount; i++) { @@ -2692,7 +2688,6 @@ TRACE("(hdc=%x, nItem=%d)\n", hdc, nItem); /* get information needed for drawing the item */ - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.mask = LVIF_TEXT | LVIF_IMAGE | LVIF_STATE | LVIF_INDENT; lvItem.stateMask = LVIS_SELECTED | LVIS_STATEIMAGEMASK; lvItem.iItem = nItem; @@ -2874,7 +2869,6 @@ hdc, nItem, rcItem.left, rcItem.top, rcItem.right, rcItem.bottom); /* get information needed for drawing the item */ - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.mask = LVIF_TEXT | LVIF_IMAGE | LVIF_STATE; lvItem.stateMask = LVIS_SELECTED | LVIS_FOCUSED; lvItem.iItem = nItem; @@ -3752,7 +3746,7 @@ /* remove it from the selection range */ - ZeroMemory(&item,sizeof(item)); + item.state = LVIS_SELECTED; item.stateMask = LVIS_SELECTED; LISTVIEW_SetItemState(infoPtr,nItem,&item); @@ -3858,10 +3852,10 @@ else { ZeroMemory(&lvItemRef,sizeof(LISTVIEW_ITEM)); - ZeroMemory(&item,sizeof(item)); item.iItem = nItem; item.iSubItem = 0; item.mask = LVIF_PARAM | LVIF_STATE; + item.stateMask = ~0; ListView_GetItemW(infoPtr->hwndSelf, &item); lvItemRef.hdr.iImage = item.iImage; lvItemRef.state = item.state; @@ -3873,7 +3867,7 @@ dispInfo.item.mask = 0; dispInfo.item.iItem = nItem; dispInfo.item.state = lpItem->state; - dispInfo.item.stateMask = 0; + dispInfo.item.stateMask = 0; /* FIXME: why not copy the state mask in here? */ dispInfo.item.pszText = pszText; dispInfo.item.cchTextMax = textlenT(pszText, isW); dispInfo.item.iImage = lpItem->hdr.iImage; @@ -3940,11 +3934,10 @@ else { LVITEMW item; - ZeroMemory(&lvItemRef,sizeof(LISTVIEW_ITEM)); - ZeroMemory(&item, sizeof(item)); item.iItem = nItem; item.iSubItem = 0; item.mask = LVIF_PARAM | LVIF_STATE; + item.stateMask = ~0; ListView_GetItemW(infoPtr->hwndSelf, &item); lvItemRef.hdr.iImage = item.iImage; lvItemRef.state = item.state; @@ -3953,7 +3946,6 @@ } /* get information needed for drawing the item */ - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.mask = LVIF_TEXT; lvItem.iItem = nItem; lvItem.iSubItem = 0; @@ -3966,7 +3958,7 @@ dispInfo.item.mask = 0; dispInfo.item.iItem = nItem; dispInfo.item.state = lpItem->state; - dispInfo.item.stateMask = 0; + dispInfo.item.stateMask = 0; /* FIXME: why not copy the state mask in here? */ dispInfo.item.pszText = lvItem.pszText; dispInfo.item.cchTextMax = lstrlenW(lvItem.pszText); dispInfo.item.iImage = lpItem->hdr.iImage; @@ -4218,8 +4210,6 @@ if ((nItem >= -1) && (lpFindInfo != NULL)) { - ZeroMemory(&lvItem, sizeof(lvItem)); - if (lpFindInfo->flags & LVFI_PARAM) { lvItem.mask |= LVIF_PARAM; @@ -5131,7 +5121,6 @@ if (uView & LVS_REPORT) { - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.mask = LVIF_INDENT; lvItem.iItem = nItem; lvItem.iSubItem = 0; @@ -5544,9 +5533,9 @@ TRACE("(nItem=%d)\n", nItem); - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.mask = LVIF_TEXT; lvItem.iItem = nItem; + lvItem.iSubItem = 0; lvItem.cchTextMax = DISP_TEXT_SIZE; lvItem.pszText = szDispText; if (LISTVIEW_GetItemW(infoPtr, &lvItem, TRUE)) @@ -5603,10 +5592,10 @@ if ((nItem >= 0) && (nItem < GETITEMCOUNT(infoPtr))) { - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.iItem = nItem; - lvItem.stateMask = uMask; + lvItem.iSubItem = 0; lvItem.mask = LVIF_STATE; + lvItem.stateMask = uMask; if (LISTVIEW_GetItemW(infoPtr, &lvItem, TRUE)) uState = lvItem.state & uMask; } @@ -6352,8 +6341,6 @@ if ( !(lpItem = (LISTVIEW_ITEM *)COMCTL32_Alloc(sizeof(LISTVIEW_ITEM))) ) return -1; - ZeroMemory(lpItem, sizeof(LISTVIEW_ITEM)); - /* insert item in listview control data structure */ if ( (hdpaSubItems = DPA_Create(8)) ) nItem = DPA_InsertPtr(hdpaSubItems, 0, lpItem); @@ -6718,7 +6705,6 @@ } else { - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.iSubItem = iCol; lvItem.mask = LVIF_TEXT; lvItem.cchTextMax = DISP_TEXT_SIZE; @@ -6783,7 +6769,6 @@ SelectObject(hdc, old_font); /* restore the old font */ ReleaseDC(infoPtr->hwndSelf, hdc); - ZeroMemory(&lvItem, sizeof(lvItem)); lvItem.iSubItem = iCol; lvItem.mask = LVIF_TEXT; lvItem.cchTextMax = DISP_TEXT_SIZE; @@ -7190,14 +7175,13 @@ BOOL bResult = TRUE; LVITEMW lvItem; - TRACE("(nItem=%d, lpLVItem=%s)\n", - nItem, debuglvitem_t(lpLVItem, TRUE)); + TRACE("(nItem=%d, lpLVItem=%s)\n", nItem, debuglvitem_t(lpLVItem, TRUE)); - ZeroMemory(&lvItem, sizeof(lvItem)); + lvItem.iItem = nItem; + lvItem.iSubItem = 0; lvItem.mask = LVIF_STATE; lvItem.state = lpLVItem->state; lvItem.stateMask = lpLVItem->stateMask ; - lvItem.iItem = nItem; if (nItem == -1) { @@ -7227,24 +7211,18 @@ */ static BOOL LISTVIEW_SetItemTextT(LISTVIEW_INFO *infoPtr, INT nItem, LPLVITEMW lpLVItem, BOOL isW) { - BOOL bResult = FALSE; - LVITEMW lvItem; + LVITEMW lvItem; - TRACE("(nItem=%d, lpLVItem=%s, isW=%d)\n", - nItem, debuglvitem_t(lpLVItem, isW), isW); + TRACE("(nItem=%d, lpLVItem=%s, isW=%d)\n", nItem, debuglvitem_t(lpLVItem, isW), isW); - if ((nItem >= 0) && (nItem < GETITEMCOUNT(infoPtr))) - { - ZeroMemory(&lvItem, sizeof(LVITEMW)); - lvItem.mask = LVIF_TEXT; - lvItem.pszText = lpLVItem->pszText; + if ((nItem < 0) && (nItem >= GETITEMCOUNT(infoPtr))) return FALSE; + lvItem.iItem = nItem; lvItem.iSubItem = lpLVItem->iSubItem; - if(isW) bResult = ListView_SetItemW(infoPtr->hwndSelf, &lvItem); - else bResult = ListView_SetItemA(infoPtr->hwndSelf, &lvItem); - } - - return bResult; + lvItem.mask = LVIF_TEXT; + lvItem.pszText = lpLVItem->pszText; + + return LISTVIEW_SetItemT(infoPtr, &lvItem, isW); } /*** @@ -7489,7 +7467,6 @@ if (!infoPtr) return -1; SetWindowLongW(hwnd, 0, (LONG)infoPtr); - ZeroMemory(infoPtr, sizeof(LISTVIEW_INFO)); infoPtr->hwndSelf = hwnd; /* determine the type of structures to use */ @@ -8097,9 +8074,8 @@ bGroupSelect = (LISTVIEW_GetItemState(infoPtr, nItem, LVIS_SELECTED) == 0); - ZeroMemory(&item, sizeof(item)); + item.state = bGroupSelect ? LVIS_SELECTED : 0; item.stateMask = LVIS_SELECTED; - if(bGroupSelect) item.state = LVIS_SELECTED; LISTVIEW_SetItemState(infoPtr, nItem, &item); LISTVIEW_SetItemFocus(infoPtr, nItem);

2 3

Patch - stub and workaround to make wineserver compile on cygwin
by Steven Edwards 22 Sep '02

22 Sep '02

This will allow you to run wineserver on cygwin. We still lack tons of stuff before we can have a WINE that really works on Cygwin but over the next few months/years++ as I am learning, I will keep working on this. Changelog: Wineserver - Stub get/set_thread_context in wineserver so it will build/run Change permission error to a Warning when running on Windows/Cygwin Index: context_i386.c =================================================================== RCS file: /home/wine/wine/server/context_i386.c,v retrieving revision 1.23 diff -u -r1.23 context_i386.c --- context_i386.c 14 Aug 2002 20:59:03 -0000 1.23 +++ context_i386.c 22 Sep 2002 03:19:41 -0000 @@ -460,9 +460,24 @@ file_set_error(); } -#else /* linux || __sun__ || __FreeBSD__ */ +#elif defined(__MINGW__) || defined(__CYGWIN__) || defined(_WINDOWS) + +/* retrieve a thread context */ +static void get_thread_context( struct thread *thread, unsigned int flags, CONTEXT *context ) +{ + printf("get_thread_context is a stub on WINDOWS!\n"); + return 0; +} + +static void set_thread_context( struct thread *thread, unsigned int flags, const CONTEXT *context ) +{ + printf("get_thread_context is a stub on WINDOWS!\n"); + return 0; +} + +#else /* linux || __sun__ || __FreeBSD__ || _WINDOWS */ #error You must implement get/set_thread_context for your platform -#endif /* linux || __sun__ || __FreeBSD__ */ +#endif /* linux || __sun__ || __FreeBSD__ || _WINDOWS */ /* copy a context structure according to the flags */ Index: request.c =================================================================== RCS file: /home/wine/wine/server/request.c,v retrieving revision 1.67 diff -u -r1.67 request.c --- request.c 16 Aug 2002 20:02:15 -0000 1.67 +++ request.c 22 Sep 2002 03:21:04 -0000 @@ -483,7 +483,11 @@ } if (!S_ISDIR(st->st_mode)) fatal_error( "%s is not a directory\n", name ); if (st->st_uid != getuid()) fatal_error( "%s is not owned by you\n", name ); +#if defined(__MINGW__) || defined(__CYGWIN__) || defined(_WINDOWS) + fprintf( stderr, "Warning: %s is accessible by other users\n", name ); +#else if (st->st_mode & 077) fatal_error( "%s must not be accessible by other users\n", name ); +#endif /* defined(__MINGW__) || defined(__CYGWIN__) || defined(_WINDOWS) */ } /* create the server directory and chdir to it */

1 0

Re: wine status page
by Brian Vincent 22 Sep '02

22 Sep '02

> that's ok. I've just uploaded your file to winehq > it's up (& running). thanx for your good work If someone wants to forward me the raw HTML I'll edit it to include WWN threads. -brian

4 4

Re: WineLib and DLLs (Winedump)
by steve.lustbader＠philips.com 22 Sep '02

22 Sep '02

One step at a time... Now I've successfully included my stubbed dll into the wine tree. When I try to link to it, however, I still get undefined references to what should be in the dll. I added my stubbed dll to the my_other_dll_DLLS section of the Makefile.in, just like it were a normal wine dll. Is that all I need to do to link to it? When I run nm on my dll in /usr/local/lib/wine, it only lists entries like: 000131a4 T __wine_stub_mydll_dll_13 It doesn't list any of the exported functions that are listed in the spec file that winedump generated. Does wine automagically match up those stub statements with the exported symbols? Does it matter that the dll is exporting C++ classes, not straight C functions? -Steve Hi Steve, > I upgraded my autoconf, which solved my first problem. Now, > everything runs fine until I get to the make install step, when > I get this error: <snip> The attached patch to winedump corrects the over-enthusiatic insertion of generated DLL entries into dlls/Makefile.in, and renames the MODULE from <dll> to <dll>.dll as per the current build process. You can manually remove the offending lines youself, edit dlls/Makefile.in and remove the 2 lines referring to your DLL after the targets "uninstall::" and "install::". Then append".dll" to the MODULE = line in dlls/<yourdll>/Makefile.in. Alternately, apply the patch and regenerate. I've just tested the process from start to finish and it works for me now. Let me know if you find any other issues. I've sent the patch to wine-patches/Alexandre. > Thanks for making those changes to the README, hopefully they'll get > committed soon so other can avoid the problems I had. np :-) Jon (See attached file: winedump_fix.diff)

2 1

Re: notepad vs write
by Sean Young 21 Sep '02

21 Sep '02

Dmitry Timoshkov <dmitry(a)baikal.ru> wrote: > "Francois Gouget" <fgouget(a)free.fr> wrote: > > The problem is that write will handle RTF and Word documents correctly, > > while notepad is just a text editor and will only display garbage... > > > > (yes, wordpad will probably have problems with complex word documents, > > but it should work with simple ones) > > I believe that Write is able to handle only old Word 5.x/6.x format files > (and of course RTFs). RTFs should not produce too much trouble (just feed > them to a richedit control). Regarding Word 5.x format here is a snipped > from my very old DOS code which handled those files (*.doc,*.wri) well > (it completely ignores formatting though): I've written a .wri to RTF converter, which is available here: http://www.msxnet.org/word2rtf/write.c It should be almost complete. The output is almost the same as the microsoft converter (including bugs, if desired). Now I have enough information to write write32.cnv, but is there any need for this? Sean

2 2

regression in recent cvs (transparent window) ?
by T'aZ 21 Sep '02

21 Sep '02

i've recently compiled wine cvs, and now, the content of a window from an app is not shown, it's the bug http://bugs.winehq.com/show_bug.cgi?id=1033 cmorgan on irc said me i might post here so that "maybe someone knows if it was their patch that did it" :) -- T'aZ *Just because you are paranoid, it doesn't mean they are not after you.*

1 0

Re: HRGN
by Michael Stefaniuc 21 Sep '02

21 Sep '02

On Fri, Sep 20, 2002 at 05:15:10PM -0400, Dimitrie O. Paun wrote: > Oddly enough, this seem to work... STOP! It dosn't work: [...] ../../windows/painting.c:433: switch quantity not an integer [...] make[2]: *** [../../windows/painting.o] Error 1 make[2]: Leaving directory `/home/michi/work/wine/dlls/user' make[1]: *** [user] Error 2 make[1]: Leaving directory `/home/michi/work/wine/dlls' make: *** [dlls] Error 2 You picked one of the 3 handles that dosn't even compile when switched to a void* (the other two are HICON, HINSTANCE). None of the to be converted handles are so simple to convert, because if it would have been so simple, it would have already been done ;). Before i start working on the conversion of a handle i do an automated compile run for every remaining DECLARE_OLD_HANDLE saving the output of the build. Here some statistic from a compile run from 3 days ago: camus:/scr/wine.test$ for i in wine.H*; do > echo -n "$i: " > diff -u wine/make.out $i/make.out | grep -c ^+ > done wine.HBITMAP: 664 wine.HBRUSH: 428 wine.HDC: 941 wine.HFONT: 460 wine.HICON: 163 wine.HINSTANCE: 200 wine.HKEY: 423 wine.HMENU: 322 wine.HPEN: 216 wine.HRGN: 220 As you can the there are at least 160 lines of warnings to look at. bye michael > > ChangeLog > Convert HRGN to void * > > Index: include/windef.h > =================================================================== > RCS file: /var/cvs/wine/include/windef.h,v > retrieving revision 1.74 > diff -u -r1.74 windef.h > --- include/windef.h 16 Sep 2002 22:47:05 -0000 1.74 > +++ include/windef.h 20 Sep 2002 18:25:30 -0000 > @@ -89,7 +89,7 @@ > DECLARE_HANDLE(HMONITOR); > DECLARE_HANDLE(HPALETTE); > DECLARE_OLD_HANDLE(HPEN); > -DECLARE_OLD_HANDLE(HRGN); > +DECLARE_HANDLE(HRGN); > DECLARE_HANDLE(HRSRC); > DECLARE_HANDLE(HTASK); > DECLARE_HANDLE(HWINEVENTHOOK); -- Michael Stefaniuc Tel.: +49-711-96437-199 System Administration Fax.: +49-711-96437-111 Red Hat GmbH Email: mstefani(a)redhat.com Hauptstaetterstr. 58 http://www.redhat.de/ D-70178 Stuttgart

2 2

Re: winedump doc
by Dimitrie O. Paun 21 Sep '02

21 Sep '02

On September 20, 2002 04:13 pm, Eric Pouech wrote: > this patch adds winedump documentation > (Alexandre, you can delete tools/winedump/README if you want) Sorry, but I don't agree with this patch. Unix docs for command line utilities come in the form of man pages (or info pages). Why put in .sgml? Who's gonna read it there? Who's gonna maintain it there? It should go into a file: programs/winedump/windump.man.in or similar... -- Dimi.

2 2

Re: documentation dir
by Dimitrie O. Paun 20 Sep '02

20 Sep '02

On September 18, 2002 04:10 am, Eric POUECH wrote: > IMO, most of the documentation (even the internals description, > status of DLL) should be made in the DocBook form I don't think this is right. Documenetation that is meant for the user should go into .sgml, no disagreement here. Documentation that is meant for the ongoing developement of a DLL (something like the TODO list, or a few comments on the internal structure, or some tips, etc.) should go as plain text as close as possible to the code that it's documenting. I can not stress how important that is. I find this "developer" documentation useful beyon words: as I read through the code, it gives me ideas on what to work, I keep it _updated_. On the other hand, I've never even read 50% of the .sgml documentation even once. But even if I do, it will never have the "current" status of the one embedded in the/close to the code. As I said before, this thing has been tried numerous times before, and everybody reached the conclusion that the documentation pertaining to code should be as close to the code that it's documenting (see javadoc et. Co.). -- Dimi.

3 5

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