September 2007 - Wine-Devel

Re: setupapi(8/22): Implement SetupDiCreateDeviceInfoW
by Alexandre Julliard Sept. 20, 2007

Sept. 20, 2007

"Juan Lang" <juan.lang(a)gmail.com> writes: > +static DWORD SETUPDI_DevNameToDevID(LPCWSTR devName) > +{ > + LPCWSTR ptr; > + DWORD devNameLen = lstrlenW(devName), devInst = 0; > + BOOL valid = TRUE; > + > + TRACE("%s\n", debugstr_w(devName)); > + for (ptr = devName; valid && *ptr && ptr - devName < devNameLen; ) > + { > + if (isdigitW(*ptr)) > + { > + devInst <<= 8; This won't do what you want. You'd need *= 10 if the id is really stored in decimal, though 0000 looks more like it should be in hex. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: winex11.drv/wintab32[3/5]: make wintab strings in winex11.drv Unicode, implement WTInfoW
by Alexandre Julliard Sept. 20, 2007

Sept. 20, 2007

"Mikolaj Zalewski" <mikolajz(a)google.com> writes: > @@ -305,8 +312,8 @@ void X11DRV_LoadTabletInfo(HWND hwnddefault) > hwndTabletDefault = hwnddefault; > > /* Do base initializaion */ > - strcpy(gSysContext.lcName, "Wine Tablet Context"); > - strcpy(gSysDevice.NAME,"Wine Tablet Device"); > + MultiByteToWideChar(CP_UTF8, 0, "Wine Tablet Context", -1, gSysContext.lcName, WT_MAX_NAME_LEN); > + MultiByteToWideChar(CP_UTF8, 0, "Wine Tablet Device", -1, gSysDevice.NAME, WT_MAX_NAME_LEN); You should define Unicode constants directly, there's no need to convert at run-time. > @@ -695,10 +716,12 @@ int X11DRV_AttachEventQueueToTablet(HWND hOwner) > X11DRV_expect_error(data->display,Tablet_ErrorHandler,NULL); > for (cur_loop=0; cur_loop < gNumCursors; cur_loop++) > { > + char cursorNameA[WT_MAX_NAME_LEN]; > int event_number=0; > > + WideCharToMultiByte(CP_UNIXCP, 0, gSysCursor[cur_loop].NAME, -1, cursorNameA, WT_MAX_NAME_LEN, NULL, NULL); > for (loop=0; loop < num_devices; loop ++) > - if (strcmp(devices[loop].name,gSysCursor[cur_loop].NAME)==0) > + if (strncmp(devices[loop].name, cursorNameA, WT_MAX_NAME_LEN)==0) This doesn't look right; if the name overflows the buffer we shouldn't simply match against a truncated string. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: Alexandre Julliard : wordpad.exe: We can now store binary files in the repository.
by Markus Amsler Sept. 20, 2007

Sept. 20, 2007

So we can also add ttf fonts and get rid of the fontforge dependency? Markus

2 1

Re: kernel32: Don't crash accessing an invalid handle in GlobalSize [try3]
by Dmitry Timoshkov Sept. 20, 2007

Sept. 20, 2007

"James Hawkins" <truiken(a)gmail.com> wrote: > This version compares the pointer value to 64000 instead of using a > try/catch block, as suggested by Alexandre. > - if(pintern->Magic==MAGIC_GLOBAL_USED) > + if(pintern > (PGLOBAL32_INTERN)64000 && pintern->Magic==MAGIC_GLOBAL_USED) > { > if (!pintern->Pointer) /* handle case of GlobalAlloc( ??,0) */ It should be either 65535 (0xffff), or a test for HIWORD(pintern) being 0. Also the cast should be to (char *) on both sides rather than (PGLOBAL32_INTERN). -- Dmitry.

2 1

Handling New IDLs with widl
by Roy Sept. 19, 2007

Sept. 19, 2007

Howdy, I'm looking for advice on how to integrate a new idl into wine. I'm implementing bits and using the widl compiler on the IDL I wrote. My current setup places the IDL in include/ with the build system generating a header file that it posits there. So far, so good. I now need to generate bits_i.c and link that into my dll located in dlls/bits. Adding: ---- IDL_I_SRCS = $(TOPSRCDIR)/include/bits.idl ---- to dlls/bits/Makefile.in solves this problem but drops the bits_i.c into include/. This is rather messy. Any advice on if I should (1) moving my IDL into dlls/bits/ and do everything locally or (2) dig into Make.rules.in and update the rule for *_i.c generation to allow the generated file to exist locally with the .idl sitting in include. The second idea is a little tricky since the current rule is careful to rename the generated .i file with the -U option: ---- .idl_i.c: $(WIDL) $(IDLFLAGS) -u -U $@ $< ---- which is a great idea, but uses the full path to the original idl when renaming the file. Probably good for the general case, but not for me. Thanks for the advice! -Roy

2 1

Re: wineboot: Pass non-zero timeouts to SendMessageTimeout, otherwise the call won't timeout.
by Alexandre Julliard Sept. 19, 2007

Sept. 19, 2007

Robert Shearman <rob(a)codeweavers.com> writes: > Set the default for WM_QUERYENDSESSION to be 5s which should be enough > time for users to respond to any message boxes and WM_ENDSESSION to a > lower 1s. If you are going to timeout on the user interaction you need to popup a confirmation dialog of some kind to allow users to force killing the app; you can't kill it just because the user didn't click on a "do you want to save your file?" message box fast enough. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: widl [2/2]: Implement pointer descriptions for complex structures
by Robert Shearman Sept. 19, 2007

Sept. 19, 2007

Dan Hipschman wrote: > var_t *field, > } > else if (is_ptr(type)) > { > - print_file(file, 2, "0x8,\t/* FC_LONG */\n"); > + unsigned char fc = (cont->type == RPC_FC_BOGUS_STRUCT > + ? RPC_FC_POINTER > + : RPC_FC_LONG); > + print_file(file, 2, "0x%x,\t/* %s */\n", fc, string_of_type(fc)); > *tfsoff += 1; > } > else if (!write_base_type(file, type, tfsoff)) > I sent a patch yesterday which does a similar thing to this chunk. I think we need to agree on which approach we're going to take so we can let Alexandre know which patch to commit. -- Rob Shearman

2 4

Re: winsock: Increase amount of iovec's to 32
by Alexandre Julliard Sept. 19, 2007

Sept. 19, 2007

Maarten Lankhorst <m.b.lankhorst(a)gmail.com> writes: > I don't know why WS_MSG_MAXIOVLEN is used here, that seems to be a > message identifier rather then the actual size of MAXIOVLEN, so I added > a define for MAXIOVLEN and made it 32. MSG_MAXIOVLEN is not a message identifier, it's really supposed to be an iovec size. Do you have a test case showing that Windows uses 32 instead? -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: include: add a macro to declare critical sections as we declare them in Wine - initialized and with a debug
by Alexandre Julliard Sept. 19, 2007

Sept. 19, 2007

"Mikolaj Zalewski" <mikolajz(a)google.com> writes: > Would such a patch be accepted? No, debug.h is about the libwine debugging API, critical sections have nothing to do with that. -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: winex11.drv: When minimizing a top level window move it outside of the visible desktop area
by Alexandre Julliard Sept. 19, 2007

Sept. 19, 2007

Dmitry Timoshkov <dmitry(a)codeweavers.com> writes: > Changelog: > winex11.drv: When minimizing a top level window move it outside of > the visible desktop area. This will break apps that expect to paint their icon window, and probably also break the positioning of the icon title. It also doesn't work even for standard icons in desktop mode. -- Alexandre Julliard julliard(a)winehq.org

1 0