February 2014 - Wine-devel - list.winehq.org

Re: [PATCH 2/2] oleaut32: Fix VariantCopy() for VT_RECORD variants (resent)
by Alexandre Julliard 20 Feb '14

20 Feb '14

Nikolay Sivov <nsivov(a)codeweavers.com> writes: > --- > dlls/oleaut32/tests/vartest.c | 46 ++++++++++++++++++++++--- > dlls/oleaut32/variant.c | 79 ++++++++++++++++++++----------------------- > 2 files changed, 79 insertions(+), 46 deletions(-) It doesn't work here: ../../../tools/runtest -q -P wine -T ../../.. -M jscript.dll -p jscript_test.exe.so run && touch run.ok wine: Unhandled page fault on read access to 0x00000001 at address 0x686c7866 (thread 007f), starting debugger... Unhandled exception: page fault on read access to 0x00000001 in 32-bit code (0x686c7866). Register dump: CS:0023 SS:002b DS:002b ES:002b FS:0063 GS:006b EIP:686c7866 ESP:0032f6a0 EBP:0032f6d8 EFLAGS:00010202( R- -- I - - - ) EAX:0032f720 EBX:687a01ec ECX:0032f6f0 EDX:00000001 ESI:00000000 EDI:0032f720 Stack dump: 0x0032f6a0: 001d0ca8 000012e0 0032f6e8 686c6914 0x0032f6b0: 00000020 00000000 001d0da8 0032f758 0x0032f6c0: 4d430003 00000000 0032f6f0 687a01ec 0x0032f6d0: 558b6bc0 00000000 0032f748 686c8eed 0x0032f6e0: 00000018 00000000 00000000 686c8eed 0x0032f6f0: 003315c0 0032f720 00000080 0032f70c Backtrace: =>0 0x686c7866 SafeArrayGetRecordInfo+0x46(psa=<couldn't compute location>, pRinfo=<couldn't compute location>) [/home/julliard/wine/wine/dlls/oleaut32/../../include/oaidl.h:4140] in oleaut32 (0x0032f6d8) 1 0x686c8eed SAFEARRAY_CopyData+0x28c(psa=0x3315c0, dest=0x331e78) [/home/julliard/wine/wine/dlls/oleaut32/safearray.c:409] in oleaut32 (0x0032f748) 2 0x686c91e6 SafeArrayCopy+0x145(psa=<couldn't compute location>, ppsaOut=<couldn't compute location>) [/home/julliard/wine/wine/dlls/oleaut32/safearray.c:1409] in oleaut32 (0x0032f7a8) 3 0x5589c671 VBArrayConstr_value+0x140(ctx=0x1ba150, vthis=0x32f838, flags=0x4000, argc=0x1, argv=0x1e5b50, r=0x32fa38) [/home/julliard/wine/wine/dlls/jscript/vbarray.c:308] in jscript (0x0032f808) 4 0x5587810b invoke_value_proc.isra+0x6a() in jscript (0x0032f858) 5 0x55878a57 Function_invoke+0x66(func_this=0x1d0d60, jsthis=(nil), flags=0x4000, argc=0x1, argv=0x1e5b50, r=0x32fa38) [/home/julliard/wine/wine/dlls/jscript/function.c:351] in jscript (0x0032f8a8) 6 0x5586b089 jsdisp_call_value+0xd8(jsfunc=0x1d0d60, jsthis=(nil), flags=0x4000, argc=0x1, argv=0x1e5b50, r=0x32fa38) [/home/julliard/wine/wine/dlls/jscript/dispex.c:1061] in jscript (0x0032f8f8) 7 0x5586b1a1 disp_call_value+0xb0(ctx=0x1ba150, disp=0x1d0d60, jsthis=(nil), flags=0x4000, argc=0x1, argv=0x1e5b50, r=0x32fa38) [/home/julliard/wine/wine/dlls/jscript/dispex.c:1200] in jscript (0x0032f9f8) 8 0x5586ef84 interp_new+0xc3(ctx=<is not available>) [/home/julliard/wine/wine/dlls/jscript/engine.c:985] in jscript (0x0032fa58) 9 0x55874a94 exec_source+0x1e3(ctx=<is not available>, code=<is not available>, func=<is not available>, from_eval=<is not available>, ret=<is not available>) [/home/julliard/wine/wine/dlls/jscript/engine.c:2462] in jscript (0x0032fad8) 10 0x5587eb46 exec_global_code.isra+0x85() in jscript (0x0032fb18) 11 0x5587ee19 JScriptParse_ParseScriptText+0x2a8(iface=<couldn't compute location>, pstrCode=<couldn't compute location>, pstrItemName=<couldn't compute location>, punkContext=<couldn't compute location>, pstrDelimiter=<couldn't compute location>, dwSourceContextCookie=<couldn't compute location>, ulStartingLine=<couldn't compute location>, dwFlags=<couldn't compute location>, pvarResult=<couldn't compute location>, pexcepinfo=<couldn't compute location>) [/home/julliard/wine/wine/dlls/jscript/jscript.c:805] in jscript (0x0032fba8) -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: user32: If the listbox loses focus while holding capture, release it by essentially simulating a button up event.
by Nikolay Sivov 20 Feb '14

20 Feb '14

On 2/20/2014 15:35, Huw Davies wrote: > --- > dlls/user32/listbox.c | 1 + > dlls/user32/tests/listbox.c | 30 ++++++++++++++++++++++++++++++ > 2 files changed, 31 insertions(+) > > case WM_KILLFOCUS: > + LISTBOX_HandleLButtonUp( descr ); /* Release capture if we have it */ > descr->in_focus = FALSE; > descr->wheel_remain = 0; This isn't totally correct. It's fine for when you press and hold left button and then change focus with Tab for example, LBN_SELCHANGE does indeed happen in this case. But when you just Tab away from a list with no mouse commands you shouldn't send LBN_SELCHANGE.

2 2

Re: [PATCH 2/2] oleaut32: Fix VariantCopy() for VT_RECORD variants (resent)
by Marvin 20 Feb '14

20 Feb '14

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://newtestbot.winehq.org/JobDetails.pl?Key=5039 Your paranoid android. === wvistau64 (64 bit vartest) === No test summary line found === w2008s64 (64 bit vartest) === No test summary line found === w7pro64 (64 bit vartest) === No test summary line found

1 0

Re: [PATCH 1/2] oleaut32/tests: Test for VariantClear() on VT_RECORD variant (try2)
by Marvin 20 Feb '14

20 Feb '14

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://newtestbot.winehq.org/JobDetails.pl?Key=5038 Your paranoid android. === wvistau64 (64 bit vartest) === No test summary line found === w2008s64 (64 bit vartest) === No test summary line found === w7pro64 (64 bit vartest) === No test summary line found

1 0

Re: msi: Dont crash on installation of services without component KeyPath
by Hans Leidekker 20 Feb '14

20 Feb '14

On Wed, 2014-02-19 at 14:49 +0900, Aric Stewart wrote: > diff --git a/dlls/msi/action.c b/dlls/msi/action.c > index 07322fe..a004225 100644 > --- a/dlls/msi/action.c > +++ b/dlls/msi/action.c > @@ -5794,27 +5794,30 @@ static UINT ITERATE_InstallService(MSIRECORD *rec, LPVOID param) > goto done; > } > key = MSI_RecordGetString(row, 6); > - file = msi_get_loaded_file(package, key); > - msiobj_release(&row->hdr); > - if (!file) > - { > - ERR("Failed to load the service file\n"); > - goto done; > - } > - > - if (!args || !args[0]) image_path = file->TargetPath; > - else > + if (key) > { > - int len = strlenW(file->TargetPath) + strlenW(args) + 2; > - if (!(image_path = msi_alloc(len * sizeof(WCHAR)))) > + file = msi_get_loaded_file(package, key); > + msiobj_release(&row->hdr); > + if (!file) > { > - ret = ERROR_OUTOFMEMORY; > + ERR("Failed to load the service file\n"); > goto done; > } > > - strcpyW(image_path, file->TargetPath); > - strcatW(image_path, szSpace); > - strcatW(image_path, args); > + if (!args || !args[0]) image_path = file->TargetPath; > + else > + { > + int len = strlenW(file->TargetPath) + strlenW(args) + 2; > + if (!(image_path = msi_alloc(len * sizeof(WCHAR)))) > + { > + ret = ERROR_OUTOFMEMORY; > + goto done; > + } > + > + strcpyW(image_path, file->TargetPath); > + strcatW(image_path, szSpace); > + strcatW(image_path, args); > + } > } > service = CreateServiceW(hscm, name, disp, GENERIC_ALL, serv_type, > start_type, err_control, image_path, load_order, > @@ -5831,7 +5834,7 @@ static UINT ITERATE_InstallService(MSIRECORD *rec, LPVOID param) > WARN("failed to set service description %u\n", GetLastError()); > } > > - if (image_path != file->TargetPath) msi_free(image_path); > + if (image_path && image_path != file->TargetPath) msi_free(image_path); > done: > CloseServiceHandle(service); > CloseServiceHandle(hscm); Should we still create the service if the installer doesn't specify the image path?

4 8

Re: [PATCH 2/2] oleaut32: Fix VariantCopy() for VT_RECORD variants
by Marvin 20 Feb '14

20 Feb '14

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://newtestbot.winehq.org/JobDetails.pl?Key=5036 Your paranoid android. === wvistau64 (64 bit vartest) === No test summary line found === w2008s64 (64 bit vartest) === No test summary line found === w7pro64 (64 bit vartest) === No test summary line found

1 0

Re: [1/2] shell32: Correct create flags when SEE_MASK_NO_CONSOLE isn't specified (resend)
by Alexandre Julliard 20 Feb '14

20 Feb '14

Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> writes: > Hi, > > > Changelog: > shell32: Correct create flags when SEE_MASK_NO_CONSOLE isn't specified There's more to it than that, now it's popping up consoles all over the place during make test. -- Alexandre Julliard julliard(a)winehq.org

2 2

Re: [PATCH 1/2] oleaut32/tests: Test for VariantClear() on VT_RECORD variant
by Marvin 20 Feb '14

20 Feb '14

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://newtestbot.winehq.org/JobDetails.pl?Key=5035 Your paranoid android. === wvistau64 (64 bit vartest) === No test summary line found === w2008s64 (64 bit vartest) === No test summary line found === w7pro64 (64 bit vartest) === No test summary line found

1 0

Re: msi: Expand REG_EXPAND_SZ keys when resolving AppSearch entries
by Marvin 20 Feb '14

20 Feb '14

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://newtestbot.winehq.org/JobDetails.pl?Key=5034 Your paranoid android. === build (build) === Patch failed to apply

1 0

Re: (try 3)winemac.drv: Improve positioning of IME client window based on caret
by Ken Thomases 20 Feb '14

20 Feb '14

On Feb 20, 2014, at 12:28 AM, Aric Stewart wrote: > --- > dlls/winemac.drv/ime.c | 15 +++++---------- > 1 file changed, 5 insertions(+), 10 deletions(-) Looks good to me. -Ken

1 0