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

Re: [3/4 revised] msi: Add partial, expandable OLE automation support.
by James Hawkins 26 Feb '07

26 Feb '07

On 2/25/07, Misha Koshelev <mk144210(a)bcm.tmc.edu> wrote: > This patch adds OLE automation support to MSI. It does this by creating > a wrapper class, which implements any of the OLE automation classes as > long as a function is given which is called from within the wrapper > classes invoke method after appropriate error checking. Basic > functionality is implemented, and it is fairly straightforward to add > the rest as the methods are all already implemented as functions in the > DLL. > +/* + * If you would like to implement a new automation function/object, look towards the bottom of this + * file for the "meat and potatoes" section. + */ + +/* FIXME: I don't know how big this should be */ +#define MAX_MSI_STRING 1000 + +/* + * AutomationObject - "base" class for all automation objects so we don't have to repeat functions. Just + * need to implement Invoke function for each dispinterface that is called from within + * the AutomationObject Invoke function which performs error checking, and pass the new + * function to create_automation_object. + */ + +typedef interface AutomationObject AutomationObject; + +interface AutomationObject { + /* + * VTables - We provide IDispatch, IProvideClassInfo, IProvideClassInfo2, IProvideMultipleClassInfo + */ + const IDispatchVtbl *lpVtbl; + const IProvideClassInfoVtbl *lpvtblIProvideClassInfo; + const IProvideClassInfo2Vtbl *lpvtblIProvideClassInfo2; + const IProvideMultipleClassInfoVtbl *lpvtblIProvideMultipleClassInfo; + + /* Object reference count */ + LONG ref; + + /* Clsid for this class and it's appropriate ITypeInfo object */ + LPCLSID clsid; + ITypeInfo *iTypeInfo; + + /* The MSI handle of the current object */ + MSIHANDLE msiHandle; + + /* A function that is called from IDispatch::Invoke, specific to this type of object. By the + * time this function is called, basic error checking has been done in the AutomationObject + * Invoke function */ + HRESULT (STDMETHODCALLTYPE *funcInvoke)( + AutomationObject* This, + DISPID dispIdMember, + REFIID riid, + LCID lcid, + WORD wFlags, + DISPPARAMS* pDispParams, + VARIANT* pVarResult, + EXCEPINFO* pExcepInfo, + UINT* puArgErr); +}; You went a little comment crazy here (and elsewhere). You should tone it down a bit. Good code explains itself and requires little or no commenting. + /* + * Perform a sanity check on the parameters. + */ + if ( (This==0) || (ppvObject==0) ) + return E_INVALIDARG; + + /* + * Initialize the return parameter. + */ + *ppvObject = 0; + + /* + * Compare the riid with the interface IDs implemented by this object. + */ + if (IsEqualGUID(riid, &IID_IUnknown) || IsEqualGUID(riid, &IID_IDispatch) || IsEqualGUID(riid, This->clsid)) + *ppvObject = This; + else if (IsEqualGUID(riid, &IID_IProvideClassInfo)) + *ppvObject = (IProvideClassInfo*)&(This->lpvtblIProvideClassInfo); + else if (IsEqualGUID(riid, &IID_IProvideClassInfo2)) + *ppvObject = (IProvideClassInfo2*)&(This->lpvtblIProvideClassInfo2); + else if (IsEqualGUID(riid, &IID_IProvideMultipleClassInfo)) + *ppvObject = (IProvideMultipleClassInfo*)&(This->lpvtblIProvideMultipleClassInfo); + + /* + * Check that we obtained an interface. + */ + if ((*ppvObject)==0) + { + TRACE("() : asking for unsupported interface %s\n",debugstr_guid(riid)); + return E_NOINTERFACE; + } + + /* + * Query Interface always increases the reference count by one when it is + * successful + */ + IClassFactory_AddRef(iface); + + return S_OK; +} Way too many comments. It may seem like nitpicking, but the comments clutter up the code and don't provide any useful information. -- James Hawkins

1 0

Re: msi: Add full JScript/VBScript support and partial expandable OLE automation support. [PATCH 3/3]
by Robert Shearman 26 Feb '07

26 Feb '07

Misha Koshelev wrote: > +/* Macros to get pointer to AutomationObject) from the other VTables. */ > Typo with extra ")" here. > +HRESULT WINAPI SessionImpl_Invoke( > + AutomationObject* This, > + DISPID dispIdMember, > + REFIID riid, > + LCID lcid, > + WORD wFlags, > + DISPPARAMS* pDispParams, > + VARIANT* pVarResult, > + EXCEPINFO* pExcepInfo, > + UINT* puArgErr) > +{ > + WCHAR szString[MAX_MSI_STRING]; > + DWORD dwLen = MAX_MSI_STRING; > + IDispatch *iDispatch = NULL; > + MSIHANDLE msiHandle; > + LANGID langId; > + UINT ret; > + INSTALLSTATE iInstalled, iAction; > Some error checking here would be good, like you do in AutomationObject_Invoke. > + > + switch (dispIdMember) > + { > ... > +/* > + * AutomationObject - "base" class for all automation objects so we don't have to repeat functions. Just > + * need to implement Invoke function for each dispinterface and pass the new function > + * to create_automation_object. > + */ > + > +typedef struct { > + /* > + * VTables - We provide IDispatch, IProvideClassInfo, IProvideClassInfo2, IProvideMultipleClassInfo > + */ > + const IDispatchVtbl *lpVtbl; > + const IProvideClassInfoVtbl *lpvtblIProvideClassInfo; > + const IProvideClassInfo2Vtbl *lpvtblIProvideClassInfo2; > + const IProvideMultipleClassInfoVtbl *lpvtblIProvideMultipleClassInfo; > + > + /* Object reference count */ > + LONG ref; > + > + /* Clsid for this class and it's appropriate ITypeInfo object */ > + LPCLSID clsid; > + ITypeInfo *iTypeInfo; > + > + /* The MSI handle of the current object */ > + MSIHANDLE msiHandle;; > + > + /* A function that is called from IDispatch::Invoke, specific to this type of object */ > + LPVOID funcInvoke; > +} AutomationObject; > You shouldn't need to expose the implementation details of AutomationObject outside of automation.c. > + > +/* This is the function that one needs to call to create an automation object. */ > +extern HRESULT create_automation_object(MSIHANDLE msiHandle, IUnknown *pUnkOuter, LPVOID *ppObj, REFIID clsid, LPVOID funcInvoke); > It is dangerous to use LPVOID as the type to pass a function in. > + > +/* We need to expose these functions because our IActiveScriptSite calls it */ > +extern HRESULT WINAPI LoadTypeInfo(IDispatch *iface, ITypeInfo **pptinfo, REFIID clsid, LCID lcid); > +extern HRESULT WINAPI SessionImpl_Invoke(AutomationObject*,DISPID,REFIID,LCID,WORD,DISPPARAMS*,VARIANT*,EXCEPINFO*,UINT*); > + > +/* Disp Ids > + * (not complete, look in msiserver.idl to add more) */ > +typedef enum { > + RecordDispId_StringData=1, > + RecordDispId_IntegerData, > + RecordDispId_SetStream, > + RecordDispId_ReadStream, > + RecordDispId_FieldCount, > + RecordDispId_IsNull, > + RecordDispId_DataSize, > + RecordDispId_ClearData, > + RecordDispId_FormatText > +} RecordDispId; > + > +typedef enum { > + ViewDispId_Execute=1, > + ViewDispId_Fetch, > + ViewDispId_Modify, > + ViewDispId_Close, > + ViewDispId_ColumnInfo, > + ViewDispId_GetError > +} ViewDispId; > + > +typedef enum { > + DatabaseDispId_DatabaseState=1, > + DatabaseDispId_SummaryInformation, > + DatabaseDispId_OpenView, > + DatabaseDispId_Commit, > + DatabaseDispId_PrimaryKeys, > + DatabaseDispId_Import, > + DatabaseDispId_Export, > + DatabaseDispId_Merge, > + DatabaseDispId_GenerateTransform, > + DatabaseDispId_ApplyTransform, > + DatabaseDispId_EnableUIPreview, > + DatabaseDispId_TablePersistent, > + DatabaseDispId_CreateTransformSummaryInfo > +} DatabaseDispId; > + > +typedef enum { > + SessionDispId_Installer=1, > + SessionDispId_Property, > + SessionDispId_Language, > + SessionDispId_Mode, > + SessionDispId_Database, > + SessionDispId_SourcePath, > + SessionDispId_TargetPath, > + SessionDispId_DoAction, > + SessionDispId_Sequence, > + SessionDispId_EvaluateCondition, > + SessionDispId_FormatRecord, > + SessionDispId_Message, > + SessionDispId_FeatureCurrentState, > + SessionDispId_FeatureRequestState, > + SessionDispId_FeatureValidStates, > + SessionDispId_FeatureCost, > + SessionDispId_ComponentCurrentState, > + SessionDispId_ComponentRequestState, > + SessionDispId_SetInstallLevel, > + SessionDispId_VerifyDiskSpace, > + SessionDispId_ProductProperty, > + SessionDispId_FeatureInfo, > + SessionDispId_ComponentCost > +} SessionDispId; > See dlls/oleaut32/tests/tmarshal_dispids.h and dlls/oleaut32/tests/tmarshal.idl for how to use DISPIDs properly. > +typedef struct { > + IActiveScriptSite lpVtbl; > + AutomationObject *session; > You don't actually access this as anything other than an IDispatch object, so you might as well change to "IDispatch *session". > +LPCWSTR read_script_from_file(LPCWSTR szFile, INT type) > Should be static. > +/* JScript or VBScript? */ > +LPCWSTR progid_from_type(INT type) > Should also be static. > + > +/* > + * Call a script. This is our meat and potatoes. > + * - Currently, since the function is relatively new, it will always end up returning S_OK. > + * Think of it like a bonus feature, we can run the script - great. If we have a problem, > + * we are no worse off than if this function had not been called. > + */ > +DWORD call_script(MSIHANDLE hPackage, INT type, LPCWSTR filename, LPCWSTR function, LPCWSTR action) > +{ > + LPCWSTR script = NULL, progId = NULL; > + HRESULT hr; > + IActiveScript *iActiveScript = NULL; > + IActiveScriptParse *iActiveScriptParse = NULL; > I haven't seen this variable notation style before. In Hungarian notation "p" is used instead of "i" to show a pointer to an object exposing the interface declared. -- Rob Shearman

2 2

Re: wined3d: Use application provided fog range when fog is enabled.
by Stefan Dösinger 26 Feb '07

26 Feb '07

Am Montag 26 Februar 2007 02:03 schrieb Vitaliy Margolen: > This should fix fog issues with PoP SoT. > --- > dlls/wined3d/state.c | 6 ++++-- > 1 files changed, 4 insertions(+), 2 deletions(-) Just wanted to add that this would fix the fixed function Z vs specular color alpha selection with table and vertex fog. Vertex shaders seem to be treated as transformed vertices. That said, the way the fixed function pipeline uses was developed by looking at what games need. The patch Vitaly sent looks good to me, but if there are any regressions I will have to try to write a unit test(though it will be tricky due to the pixel exactness thingy)

1 0

Re: [2/2] user32: do not call NULL message callback
by Robert Shearman 25 Feb '07

25 Feb '07

Fabian Bieler wrote: > @@ -8894,23 +8894,20 @@ static void test_nullCallback(void) > HWND hwnd; > MSG msg; > > - if(0) > - { > - hwnd = CreateWindowExA(0, "TestWindowClass", "Test overlapped", WS_OVERLAPPEDWINDOW, > - 100, 100, 200, 200, 0, 0, 0, NULL); > - ok (hwnd != 0, "Failed to create overlapped window\n"); > + hwnd = CreateWindowExA(0, "TestWindowClass", "Test overlapped", WS_OVERLAPPEDWINDOW, > + 100, 100, 200, 200, 0, 0, 0, NULL); > + ok (hwnd != 0, "Failed to create overlapped window\n"); > > - ShowWindow(hwnd, SW_SHOW); > - UpdateWindow( hwnd ); > + ShowWindow(hwnd, SW_SHOW); > + UpdateWindow( hwnd ); > > - SendMessageCallbackA(hwnd,WM_NULL,0,0,NULL,0); > + SendMessageCallbackA(hwnd,WM_NULL,0,0,NULL,0); > > - while (PeekMessage( &msg, 0, 0, 0, PM_REMOVE )) DispatchMessage( &msg ); > + while (PeekMessage( &msg, 0, 0, 0, PM_REMOVE )) DispatchMessage( &msg ); > > - ok(1, "\n"); > + ok(1, "\n"); > > - DestroyWindow(hwnd); > - } > + DestroyWindow(hwnd); > } > > START_TEST(msg) > You might as well add the test and fix the bug in one patch since the second patch is no shorter due to all of the whitespace changes. -- Rob Shearman

1 0

Re: a big fan (now with bad news)
by Joris Huizer 25 Feb '07

25 Feb '07

Aww... yes, I'm seeing that too According the the wine application database (http://appdb.winehq.org/appview.php?iVersionId=469) some settings help it get a bit further, but it reports: "Microsoft Internet Explorer has encountered a problem and needs to close. We are sorry for the inconvenience." (With a "debug" button) With these messages: fixme:shell:StopWatchMode () stub! fixme:shdocvw:IEWinMain "" 10 fixme:ole:CoResumeClassObjects stub err:module:DelayLoadFailureHook failed to delay load urlmon.dll.CreateURLMoniker wine: Call from 0x7b842138 to unimplemented function urlmon.dll.CreateURLMoniker, aborting fixme:imm:ImmDisableIME (-1): stub fixme:advapi:RegisterEventSourceW ((null),L"Microsoft Internet Explorer"): stub fixme:advapi:ReportEventA (0xcafe4242,0x0001,0x0000,0x000003e8,(nil),0x0005,0x00000050,0x632f6ec0,0x632f6a84): stub fixme:advapi:ReportEventW (0xcafe4242,0x0001,0x0000,0x000003e8,(nil),0x0005,0x00000050,0x161400,0x632f6a84): stub err:eventlog:ReportEventW L"iexplore.exe" err:eventlog:ReportEventW L"6.0.2800.1106" err:eventlog:ReportEventW L"" err:eventlog:ReportEventW L"0.0.0.0" err:eventlog:ReportEventW L"00000000" fixme:advapi:DeregisterEventSource (0xcafe4242) stub fixme:wininet:InternetGetConnectedState always returning LAN connection. Hitting "debug" crashes as an unimplemented function in urlmon is called - that seems weird, as it should be using the native version of urlmon... oh well I guess you need to use ies4linux to get a working IE for now. I had it installed as some program thought it needed it - hadn't noticed it didn't actually run Jeffrey Sabarese <core5fedora(a)gmail.com> wrote: Hi Joris, Thanks again for corresponding with me on this. (i realize there's much text here, so please take you time, get to it when you can...) for the link, just think: "this dude is such a 'NOVICE'!! he really should take more 'NOTES'!!!..." get it? novicenotes? hee hee then a text search (upper right) of WinE ie will get you there, undoubtedly. but that's not why i am writing. i wanted to let you know that the "genuine" installer, http://www.microsoft.com/windows/ie/ie6/downloads/critical/ie6sp1/default.m… did not succeed in placing a useable software on the system. it seemed to "install" just fine, but launching it: [ wine '/home/user/.wine/drive_c/Program Files/Internet Explorer/iexplore.exe' ] brings up a window w/ nothing but "white space". as if it "wants" to launch, but doesn't quite get there. when ie was installed via "ies4linux", i DID experience a similar activity, but only if i launched IE via WinE application link, for example, if i linked to IE from... say EditPlus, as an external browser, i would get the same: a window w/ nothing in it-- as you might see after an app has crashed, the window's still up, but no screen-updating, so you get a "frozen" image, or white screen. so, the only time i've ever seen it look "right" myself was using the "non-standard" ies4linux launcher, as i've illustrated specifically, here some time ago (showing the 3 versions of IE in that package): http://digisquirrel.blogspot.com/2006/08/fiddlers-three.html please understand that i want to install it the proper, WinE developers' recommended way, so this is why i continue to research the subject (and continue to probably annoy you, for which i apologize). What might i be doing wrong? i noticed in my blog entry-- (a different one this time) http://novicenotes.com/2006/08/04/ie_on_linux_wine/ where i discussed DCOM98.EXE . i haven't actively installed that component this time-- so unless it came down w/ the MS installer, might i need to have that installed as well? any other feedback you have on this subject is appreciated. perhaps you have URL w/ the preferred, WinE dev's method? (assuming it's "okay [legal]" to do the MS installer) thank you for your generous help!! --Jeff Sabarese PS. I apologize if it's been an invasion-- i thought i mailed to the "list" last time, but it looks like i sent just to you last time? (and i know i sent only to you this time, so if you wish to forward to the list, please do so. i have a real problem w/ being not-concise [as you've noticed, i'm sure], so i tried to spare the other members for now. sorry i can't seem to shorten things more than i do). On 2/24/07, Joris Huizer wrote: > I think I don't have it anymore, though I'd like to > see - though I'm pretty sure you got it covered well, > like before. > > regards, > > Joris > > > > --- Jeffrey Sabarese wrote: > > > hey there! thanks for the valuable lesson learned > > here! > > > > btw-- i changed my web log info , per > > recommendations i received in > > this mail conversation. if you still have the link > > and care to see > > it-- you'll see that i've now got the wineHQ URL > > there, and a > > recommendation about IE6 should install properly > > :) > > > > On 2/23/07, Joris Huizer > > wrote: > > > > > > --- Jeffrey Sabarese > > wrote: > > > > > > > > > > > I tried [ yum remove wine ], and then [yum > > install > > > > wine] which went > > > > without a hitch, but there's no change in my > > Primary > > > > User (the broken > > > > one) > > > > HOWEVER, WinE continues to run perfectly under > > other > > > > user accounts, and ROOT. > > > > > > > > > > The wine configuration is in ~/.wine > > > Remove/rename it for a fresh start. Wine will > > > automatically recreate the ~/.wine directory with > > > required contents for normal functionality (but > > then > > > you need to do installation of every program > > again) > > > > > > > > > > > > > > > ____________________________________________________________________________________ > > > Yahoo! Music Unlimited > > > Access over 1 million songs. > > > http://music.yahoo.com/unlimited > > > > > > > > > > > > > > > -- > > /* ************************************************ > > */ > > Quanta+ Web Development IDE: > > http://quanta.kdewebdev.org/ > > Quanta+ Wiki: > > http://tinyurl.com/3dzqxk > > Other Quanta Tips: > > http://invisiblepixels.org/progs/quanta.html > > /* ************************************************ > > */ > > > > > > > ____________________________________________________________________________________ > Looking for earth-friendly autos? > Browse Top Cars by "Green Rating" at Yahoo! Autos' Green Center. > http://autos.yahoo.com/green_center/ > -- /* ************************************************ */ Quanta+ Web Development IDE: http://quanta.kdewebdev.org/ Quanta+ Wiki: http://tinyurl.com/3dzqxk Other Quanta Tips: http://invisiblepixels.org/progs/quanta.html /* ************************************************ */ --------------------------------- Need Mail bonding? Go to the Yahoo! Mail Q&A for great tips from Yahoo! Answers users.

2 1

Re: msi: Add handlers for JScript/VBScript actions that call one script function. [PATCH 2/3]
by Robert Shearman 25 Feb '07

25 Feb '07

Misha Koshelev wrote: > All > actual script implementation is left (for the next patch) in the > call_script function. Patches should attempt to be atomic changes that compile and pass all tests without another patch needing to be applied on top. I suggest you implement automation.c and the associated header file in patch 2 and combine this patch with the rest of patch 3. > +static UINT HANDLE_CustomType37_38(MSIPACKAGE *package, LPCWSTR source, > + LPCWSTR target, const INT type, LPCWSTR action) > +{ > ... > + if (GetTempFileNameW(fmt, f1, 0, tmp_file) == 0) > ... > + info = do_msidbCustomActionTypeScript( package, type, tmp_file, NULL, action ); > + > + return wait_thread_handle( info ); > +} > + > +static UINT HANDLE_CustomType5_6(MSIPACKAGE *package, LPCWSTR source, > + LPCWSTR target, const INT type, LPCWSTR action) > +{ > ... > + r = store_binary_to_temp(package, source, tmp_file); > ... > + info = do_msidbCustomActionTypeScript( package, type, tmp_file, target, action ); > + > + return wait_thread_handle( info ); > +} > + > +static UINT HANDLE_CustomType21_22(MSIPACKAGE *package, LPCWSTR source, > + LPCWSTR target, const INT type, LPCWSTR action) > +{ > ... > + info = do_msidbCustomActionTypeScript( package, type, file->TargetPath, target, action ); > + > + return wait_thread_handle( info ); > +} > + > +static UINT HANDLE_CustomType53_54(MSIPACKAGE *package, LPCWSTR source, > + LPCWSTR target, const INT type, LPCWSTR action) > +{ > ... > + if (GetTempFileNameW(fmt, f1, 0, tmp_file) == 0) > ... > + info = do_msidbCustomActionTypeScript( package, type, tmp_file, NULL, action ); > + > + return wait_thread_handle( info ); > +} > It would seem simpler here to just open the file here for the one case that has an external file and just pass in the LPWSTR script for all cases. -- Rob Shearman

1 0

Success with enhanced cd's... kind of
by Dan Kegel 24 Feb '07

24 Feb '07

I noticed that my copy of "No!" by "They Might Be Giants" is an "enhanced cd", so I inserted it into my Ubuntu Dapper laptop. Ubuntu only noticed the audio tracks, so I had to actually use the commandline to mount the data tracks: sudo mount -oro,loop -tiso9660 /dev/cdrom /mnt After that it went fine; wine was able to run the flash app on the disc. Looks like others have complained about this, e.g. http://ubuntuforums.org/showthread.php?p=2143161 but I didn't see any bug reports about it, so I filed https://launchpad.net/ubuntu/+bug/87663 - Dan -- Wine for Windows ISVs: http://kegel.com/wine/isv

2 1

RE: xcopy try 2
by Ann & Jason Edmeades 24 Feb '07

24 Feb '07

Hi Detlef, (again) >Configure.ac - I included that because another 'regular' developer who >added a dll recently added it (See:) > http://www.winehq.org/pipermail/wine-patches/2007-February/036243.html) >I'll leave it in for that reason unless there's a big problem with it (I >always thought you were supposed to send in configure.ac but not configure) I stand corrected - tools\make_makefiles does that update for me as well as the others :-) Jason

1 0

Re: xcopy try 2
by Detlef Riekenberg 24 Feb '07

24 Feb '07

On Fr, 2007-02-23 at 16:44 +0000, Ann & Jason Edmeades wrote: Looks nice, but a view over the code result in some more comments: --- a/configure.ac +++ b/configure.ac Not needed, when sending the Patch (and that line causes already an offset, when patching wine) "tools/make_makefiles" takes care of all required changes and when Alexandre commit your Patch, the changes from make_makefiles and autoconf are included in the commit. +EXTRADEFS = -DUNICODE There is no need for that define. Remove that line to get a warning for all ANSI-Functions, that missed the "W" +C_SRCS = \ + xcopy.c Spaces are wrong before the filename. You must use TAB in makefiles +/* Prototypes */ This can be avoided completly, when you rearange your functions (so main() is then the last function in the file) + printf("Invalid number of parameters - Use xcopy /? for help\n"); The text should be read from the resources, so it can be localized . You should convert the UNICODE-Message with WideCharToMultiByte. When the output-handle is the console, printing UNICODE-messages avoid converting the message twice (xcopy and kernel32) > + if (*argvW[0] != '/') { > + if (suppliedsource[0] == 0x00) { > + lstrcpyW(suppliedsource, *argvW); > + } else { > + lstrcpyW(supplieddestination, *argvW); > + } This is wrong. tested with: "xcopy file1 file2 file3" Your code would copy file1 to file3, while w2k fails with "Unzulässige Parameteranzahl" (Invalid number of parameters) I suggest to initialize supplieddestination as empty string and copy the arg to supplieddestination only, when it is emtpy. When supplieddestination was not empty, quit with the error. You need to fill supplieddestination with the default "." before /* Trace out the supplied information */, when it is still empty at that location. + if (flags & OPT_SIMULATE) { + printf("%d file(s) would be copied\n", filesCopied); + } else { + printf("%d file(s) copied\n", filesCopied); + } As Above: Format-Pattern from the resource and WriteFile() + if (GetFullPathName(suppliedsource, MAX_PATH, "W" missing: => GetFullPathNameW (More Functions in various locations and WIN32_FIND_DATA => WIN32_FIND_DATAW) I did not test the code very much, but I think, you did. (ToDo-Reminder: "xcopy file1 name_not_found" => is "name_not_found" a file or a directory) Thanks for your work. It's pretty good as start! -- By by ... Detlef

3 2

New wintest executable is there (but there are some issues due to introduction of skip)
by Paul Vriens 24 Feb '07

24 Feb '07

Hi, as of today there's a new winetest executable available at the usual place: http://www.astro.gla.ac.uk/users/paulm/WRT/CrossBuilt/ There is however an issue with the generated reports. It looks like this is introduced when the 'skip' thingy was added to the test infrastructure. I'm currently going through the perl scripts but have no fix yet. Most likely we have to patch the 'dissect' script at line 158: } elsif (s/^($test: (\d+) tests executed, (\d+) marked as todo, (\d+) failures?\.)\r?$/$1/) { The previous lines were like: advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31 crypt: 69 tests executed, 0 marked as todo, 0 failures. advapi32:crypt done (0) now it's: advapi32:crypt start dlls/advapi32/tests/crypt.c 1.31 crypt: 69 tests executed (0 marked as todo, 0 failures), 0 skipped. advapi32:crypt done (0) Anybody? Cheers, Paul.

3 3