September 2009 - Wine-Devel

Running 13 months locales

by Nikolay Sivov

Hi. Working on monthcal control I've realized that some locales could have 13 months calendars. How could I guess that I'm running such locale? After a brief look I didn't find anything useful in GetLocaleInfo().

Sept. 26, 2009

Re: sti: test behavior of creation with the A/W functions versus STI_VERSION_FLAG_UNICODE

by Detlef Riekenberg

On Fr, 2009-09-25 at 16:35 +0200, Damjan Jovanovic wrote: > * sti: test behavior of creation with the A/W functions versus > STI_VERSION_FLAG_UNICODE Using the name "sti_tests" is bad. The short "sti" is enough. You forgot the license notice We usually cast the result of GetProcAddress() as (void *). That is not perfect, but much better to read. -- By by ... Detlef

Sept. 26, 2009

Re: [PATCH 3/4] mshtml: Add helper function get_url_components

by Detlef Riekenberg

On Fr, 2009-09-25 at 15:30 -0500, Andrew Eikum wrote: > +static HRESULT get_url_components(HTMLLocation *This, URL_COMPONENTSW > *url) > +{ > + const WCHAR *doc_url; > + if(!InternetCrackUrlW(doc_url, 0, 0, url)) { > + FIXME("InternetCrackUrlW failed\n"); Such a FIXME does not help to find the reason. Dumping doc_url helps, but might add personal data to the log -- By by ... Detlef

Sept. 26, 2009

Re: [PATCH 2/4] mshtml: Add test suite for IHTMLLocation

by Detlef Riekenberg

On Fr, 2009-09-25 at 15:30 -0500, Andrew Eikum wrote: > + todo_wine ok(SUCCEEDED(hres), "get_hash failed: 0x%08x\n", > hres); > + if(SUCCEEDED(hres)){ > + if(str_pass) > + ok(strcmp_wa_null(str, exp) == 0, "expected retrieved > hash to be L\"%s\", was: %s\n", exp, wine_dbgstr_w(str)); > + else > + todo_wine ok(strcmp_wa_null(str, exp) == 0, "expected > retrieved hash to be L\"%s\", was: %s\n", exp, wine_dbgstr_w(str)); > + } > +} > + Your Patch is hard to read. Please split the long lines (3 parts are easy) The whole patch can be split in smaller patches. > +static void perform_test(const struct location_test* test) In the case of a test failure or a crash, there is no chance to know, which test-set failed. A suffix (protocol name or line number) at the start of every ok() can be used to make the normal output as silent as possible. Using a trace before you call perform_tests() works also. That produce a bit noise in the result file, but the test binary is smaller. -- By by ... Detlef

Sept. 26, 2009

Re: kernel32: check parameter of CheckRemoteDebuggerPresent with tests

by Alexandre Julliard

Stefan Leichter <Stefan.Leichter(a)camline.com> writes: > + SetLastError(0xdeadbeef); > + bret = pCheckRemoteDebuggerPresent(INVALID_HANDLE_VALUE,NULL); > + ok(!bret , "expected CheckRemoteDebuggerPresent to fail\n"); > + ok(ERROR_INVALID_PARAMETER == GetLastError(), > + "expected error ERROR_INVALID_PARAMETER, got %d/%x\n",GetLastError(), GetLastError()); It doesn't make sense to pass INVALID_HANDLE_VALUE as a process handle, it's only for files. You probably want GetCurrentProcess(). -- Alexandre Julliard julliard(a)winehq.org

Sept. 26, 2009

Re: RFC: Sharing a wineprefix

by Vit Hrachovy

Hi Paul, as usual admin of home PC server with few users sharing installed games, I'm really missing feature similar to this in Wine. You've got my vote and should you need testing, I'm glad to help. Regards Hark

Sept. 26, 2009

Re: cppcheck sept 18 redux

by Detlef Riekenberg

On Di, 2009-09-22 at 07:46 -0700, Chris Robinson wrote: > so dmW->dmFormName is the same as &(*dmW).dmFormName, When the target is an array, the address of the first member is returned: dmW->fmFormName is: &(dmW->dmFormName[0]) similar: char buffer[32]; printf("buffer at %p\n", buffer); which is only a shortcut for: printf("buffer[0] at %p\n", &buffer[0]); > However, since GCC will remove deadcode and it's simple to see the dereference > isn't needed, it just optimizes it away. No, removing deadcode must be enabled. You can use every compiler. > but I'm not so > sure that it's guaranteed by the C standard. Is it? Yes. -- By by ... Detlef

Sept. 26, 2009

Re: RFC: Sharing a wineprefix

by Paul Chitescu

On Saturday 26 September 2009 02:19:53 Avery Pennarun wrote: > On Fri, Sep 25, 2009 at 3:35 PM, Paul Chitescu <paulc(a)voip.null.ro> wrote: > > Required changes I've identified so far: > > - The R/W lock to the prefix > > - wineserver -k as admin should terminate all users' instances > > - wineserver needs to be able to load Registry hives R/O > > - wineboot needs to be able to create just the user's profile and > > ignore pending renames for restricted users > > - Report to application the detected type of user: Administrator / > > Restricted - Some winecfg settings need to be migrated from HKCU to HKLM > > so the admin can enforce sane values or provide proper defaults or > > because they are logically per prefix > > - The user Registry hive and the temporary directory for the > > restricted users need to be stored in their profile directory > > In my particular case, what I would like is in some ways actually simpler: > > - Define WINEPREFIX and WINEUSERPREFIX. > > - For all file accesses: > - If accessing read-only: try reading from WINEUSERPREFIX; if it's > not there, use WINEPREFIX instead. > - If accessing read-write: try reading from WINEUSERPREFIX; if it's > not there, copy from WINEPREFIX instead. > > If the admin updates WINEPREFIX, then I can just wipe out my personal > WINEUSERPREFIX to see his changes. > > You might want to be configurable to have certain paths (the HKLM > registry and Program Files folders, among others) not ever use the > WINEUSERPREFIX, but I don't need that. > > I think this feature would be really handy for supporting shared > installations. One big advantage of this method over others is that > it should work even if multiple users are doing destructive things to > the same files; they won't see each others' changes. This makes it > great for test environments (which is one of the things I'd definitely > use it for). > > Have fun, > > Avery What you're proposing here is essentially an overlay file system. The best would be to use a real overlay file system driver as: - Doesn't require complex coding in Wine - Can offer features not easy to implement like local deletion (file or directory exist in the main prefix but we don't want them in the "copy") - It offers an unified view between Wine and the host operating system - Can be tested outside Wine - The developers of the filesystem already banged their heads against the many ways such an overlay system may fail. You should also know that any such simplistic approach can fail in interesting and hardly to debug ways unless the user's prefix is wiped (which you already noted) which is unacceptable in most scenarios. Paul

Sept. 26, 2009

Re: user32: Add check for the need to redraw static control after STM_SETIMAGE and STM_SETICON

by Nikolay Sivov

Ilya Shpigor wrote: > Hi, > > This patch fix the redrawing static controls after the processing STM_SETIMAGE > and STM_SETICON messages with errors. > --- + if ((style & SS_TYPEMASK) != SS_ICON) return (HICON)-1; ... + if ((style & SS_TYPEMASK) != SS_BITMAP) return (HBITMAP)-1; ... + if ((style & SS_TYPEMASK) != SS_ENHMETAFILE) return (HENHMETAFILE)-1; --- Add tests for that please (MSDN tells about 0 return value on error).

Sept. 26, 2009

Re: RFC: Sharing a wineprefix

by Avery Pennarun

On Fri, Sep 25, 2009 at 3:35 PM, Paul Chitescu <paulc(a)voip.null.ro> wrote: > Required changes I've identified so far: > - The R/W lock to the prefix > - wineserver -k as admin should terminate all users' instances > - wineserver needs to be able to load Registry hives R/O > - wineboot needs to be able to create just the user's profile and ignore > pending renames for restricted users > - Report to application the detected type of user: Administrator / Restricted > - Some winecfg settings need to be migrated from HKCU to HKLM so the admin > can enforce sane values or provide proper defaults or because they are > logically per prefix > - The user Registry hive and the temporary directory for the restricted users > need to be stored in their profile directory In my particular case, what I would like is in some ways actually simpler: - Define WINEPREFIX and WINEUSERPREFIX. - For all file accesses: - If accessing read-only: try reading from WINEUSERPREFIX; if it's not there, use WINEPREFIX instead. - If accessing read-write: try reading from WINEUSERPREFIX; if it's not there, copy from WINEPREFIX instead. If the admin updates WINEPREFIX, then I can just wipe out my personal WINEUSERPREFIX to see his changes. You might want to be configurable to have certain paths (the HKLM registry and Program Files folders, among others) not ever use the WINEUSERPREFIX, but I don't need that. I think this feature would be really handy for supporting shared installations. One big advantage of this method over others is that it should work even if multiple users are doing destructive things to the same files; they won't see each others' changes. This makes it great for test environments (which is one of the things I'd definitely use it for). Have fun, Avery

Sept. 25, 2009