December 2022 - Wine-Bugs - list.winehq.org

[Bug 54228] CPython fails to import symlinked packages

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54228 --- Comment #1 from Sebastian Ernst <ernst(a)pleiszenburg.de> --- A quick "poor-man's regression test" based on Ubuntu packages actually puts the breaking change between staging 7.17 (working) and staging 7.18 (broken), i.e. between ca September 9th and 23rd 2022 (build dates). Thanks for keeping them all online. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54228] New: CPython fails to import symlinked packages

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54228 Bug ID: 54228 Summary: CPython fails to import symlinked packages Product: Wine-staging Version: 8.0-rc2 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: ernst(a)pleiszenburg.de CC: leslie_alistair(a)hotmail.com, z.figura12(a)gmail.com Distribution: --- Created attachment 73774 --> https://bugs.winehq.org/attachment.cgi?id=73774 complete, minimal reproducing example - Present in, at least: Wine Staging 7.22 and 8.0-rc2 - Not present: Wine Stable 7.0 - Wine builds: dl.winehq.org/wine-builds/ubuntu - System: Ubuntu 22.04 - Expected behavior: CPython <= 3.9 can import symlink-ed modules and their submodules - Actual behavior: CPython sees the folder and can navigate it, but fails to import its contents (submodules) - Reproduced by: https://gist.github.com/s-m-e/84d5fbf5663825a46d9ae98e4636d8ec (also attached) --- I have noticed this issue in Wine Staging 7.22 and 8.0-rc2. I believe it must have found its way into Wine Staging in the October to November 2022 time frame. This issue does not appear in Wine Staging until about September. It also does not appear in stable releases of Wine 7. Python's import mechanism was rewritten recently, so only versions prior to Python 3.10 appear to be affected. Many packages support down to 3.7 at the moment. I am heavily using CPython on top of Wine. When Python imports modules, it looks for them in certain paths, specified in `sys.path`. For development, it is common practise to symlink a package's folder from e.g. a git repo into one of the folders contained in `sys.path`, usually `site-packages`. Python then happily imports the module from there for e.g. testing. Many tools are built around this workflow. ``` Traceback (most recent call last): File "<string>", line 1, in <module> File "Z:\work\bar\__init__.py", line 1, in <module> from .subbar import funcbar ModuleNotFoundError: No module named 'bar.subbar' ``` Python's module import mechanism appears to follow the symlink correctly (into module "bar") but fails to "locate" stuff within the linked folder (submodule "bar.subbar"). Other Python code can at least correctly navigate the linked folder's contents, so I am suspecting a `stat` syscall of some kind to return unusual values. For a complete, minimal reproducing example see link above or attached script. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54221] vbscript: missing support for GetRef

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54221 --- Comment #10 from Jason Millard <jsm174(a)gmail.com> --- So this is probably the worst hack ever, but I leveraged VARIANT's wReserved2 to store the function index. static HRESULT Global_GetRef(BuiltinDisp *This, VARIANT *arg, unsigned args_cnt, VARIANT *res) { BSTR str; HRESULT hres; hres = to_string(arg, &str); if(FAILED(hres)) return hres; function_t **funcs = This->ctx->script_obj->global_funcs; size_t i, cnt = This->ctx->script_obj->global_funcs_cnt; for(i = 0; i < cnt; i++) { if(!wcsicmp(funcs[i]->name, str)) { IDispatch* disp = (IDispatch*) &This->ctx->script_obj->IDispatchEx_iface; DISPID dispid; hres = disp->lpVtbl->GetIDsOfNames(disp, 0, (LPOLESTR*)&funcs[i]->name, 1, 0, &dispid); if(FAILED(hres)) return hres; V_VT(res) = VT_DISPATCH; V_DISPATCH(res) = disp; res->n1.n2.wReserved1 = 0xFF; res->n1.n2.wReserved2 = i; IDispatch_AddRef(disp); return hres; } } return DISP_E_UNKNOWNNAME; } Then in do_icall: static HRESULT do_icall(exec_ctx_t *ctx, VARIANT *res, BSTR identifier, unsigned arg_cnt) { DISPPARAMS dp; ref_t ref; HRESULT hres; TRACE("%s %u\n", debugstr_w(identifier), arg_cnt); hres = lookup_identifier(ctx, identifier, VBDISP_CALLGET, &ref); if(FAILED(hres)) return hres; switch(ref.type) { case REF_VAR: case REF_CONST: if (V_VT(ref.u.v) == VT_DISPATCH && ref.u.v->n1.n2.wReserved1 == 0xFF) { vbstack_to_dp(ctx, arg_cnt, FALSE, &dp); ref.type = REF_FUNC; ref.u.f = ctx->script->script_obj->global_funcs[ref.u.v->n1.n2.wReserved2]; hres = exec_script(ctx->script, FALSE, ref.u.f, NULL, &dp, res); if(FAILED(hres)) return hres; break; } if(arg_cnt) return variant_call(ctx, ref.u.v, arg_cnt, res); This allows the following code to work: Sub test Debug.print "test call" End Sub Dim funcAlias set funcAlias = GetRef("test") If Not IsObject(GetRef("test")) Then Debug.print "getref test not object" funcAlias -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54191] Agarest 1 stopped showing Videos after implementing ASF Reader filter

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54191 --- Comment #3 from j-r(a)online.de --- Thanks for the quick response. Sorry for not searching more thoroughly for existing reports (Agarest has always been special when it comes to videos, so I wrongly assumed it was specific to the game). -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54221] vbscript: missing support for GetRef

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54221 --- Comment #9 from Jason Millard <jsm174(a)gmail.com> --- (In reply to Nikolay Sivov from comment #8) > We have some code in vbdisp.c that looks related and good start to figure > out how this works. Okay, I will dig in. This is where I'm at now: static HRESULT Global_GetRef(BuiltinDisp *This, VARIANT *arg, unsigned args_cnt, VARIANT *res) { BSTR str; HRESULT hres; hres = to_string(arg, &str); if(FAILED(hres)) return hres; function_t **funcs = This->ctx->script_obj->global_funcs; size_t i, cnt = This->ctx->script_obj->global_funcs_cnt; for(i = 0; i < cnt; i++) { if(!wcsicmp(funcs[i]->name, str)) { IDispatch* disp = (IDispatch*) &This->ctx->script_obj->IDispatchEx_iface; DISPID dispid; hres = disp->lpVtbl->GetIDsOfNames(disp, 0, (LPOLESTR*)&funcs[i]->name, 1, 0, &dispid); if(FAILED(hres)) return hres; V_VT(res) = VT_DISPATCH; V_DISPATCH(res) = disp; IDispatch_AddRef(disp); return hres; } } return DISP_E_UNKNOWNNAME; } -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54221] vbscript: missing support for GetRef

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54221 --- Comment #8 from Nikolay Sivov <bunglehead(a)gmail.com> --- We have some code in vbdisp.c that looks related and good start to figure out how this works. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54221] vbscript: missing support for GetRef

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54221 --- Comment #7 from Jason Millard <jsm174(a)gmail.com> --- Sorry, it took me a while to understand what you were telling me. In windows, I did this: m_pScript->SetScriptState(SCRIPTSTATE_CONNECTED); EXCEPINFO exception = {}; VARIANT var; WCHAR* wzText = new WCHAR[1024]; swprintf_s(wzText, 1024, L"GetRef(\"Table1_KeyDown\")"); m_pScriptParse->ParseScriptText(wzText, 0, 0, 0, CONTEXTCOOKIE_NORMAL, 0, SCRIPTTEXT_ISEXPRESSION, &var, &exception); Looking at var it's definitely a dispatch into vbscript.dll I tried to look at the typeinfo with: IDispatch* disp = V_DISPATCH(&var); ITypeInfo* ti; HRESULT hr = disp->GetTypeInfo(0, LOCALE_USER_DEFAULT, &ti); but HR comes back with DISP_E_BADINDEX -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54227] New: Altium Designer 20.x - 22.x slowness in 2D schematics editor - possibly GDI font rendering problem

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54227 Bug ID: 54227 Summary: Altium Designer 20.x - 22.x slowness in 2D schematics editor - possibly GDI font rendering problem Product: Wine-staging Version: 7.22 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown Assignee: wine-bugs(a)winehq.org Reporter: kepszlok(a)gmail.com CC: leslie_alistair(a)hotmail.com, z.figura12(a)gmail.com Distribution: --- Created attachment 73773 --> https://bugs.winehq.org/attachment.cgi?id=73773 Wine's output This is a problem that affects all Altium Designer versions from 20.x to current, with all wine-staging versions of 7.xx branch. See AD compatibility reports: 20.x https://appdb.winehq.org/objectManager.php?sClass=version&iId=38478 21.x https://appdb.winehq.org/objectManager.php?sClass=version&iId=40170 22.x https://appdb.winehq.org/objectManager.php?sClass=version&iId=40683 All of them shows slowness in the 2D schematics editor, while the 3D PCB editor works fine. Using the schematics editor, this message appears in constantly: 0118:fixme:d2d:d2d_device_context_DrawGeometry Ignoring stroke style 0000000001647F70. (the schematics editor itself working fine, just unusably slow) Possible reason: "probably because X2 uses GDI font rendering by default and lacks option to turn it off" -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 54226] New: widl: implement `cstruct_out` option to generate C bindings compatible with small struct return values

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=54226 Bug ID: 54226 Summary: widl: implement `cstruct_out` option to generate C bindings compatible with small struct return values Product: Wine Version: unspecified Hardware: x86-64 OS: Windows Status: UNCONFIRMED Severity: normal Priority: P2 Component: tools Assignee: wine-bugs(a)winehq.org Reporter: alvin(a)alvinhc.com According to https://devblogs.microsoft.com/oldnewthing/20220113-00/?p=106152, the previous C bindings are incompatible with the C++ COM ABI when the method returns a small struct for MSVC. Their solution is to add the `/cstruct_out` flag to the MIDL compiler (https://learn.microsoft.com/en-us/windows/win32/midl/-cstruct-out, added in the Win11 SDK according to comments) to generate new C bindings compatible with the C++ COM ABI. This is similar to the (long-standing) GCC issue with its mingw-w64 MS ABI (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64384) which has been worked around in widl/mingw-w64 by having special wrapper functions in the C++ bindings, guarded by `WIDL_EXPLICIT_AGGREGATE_RETURNS`. I think widl should follow suit to implement a `--cstruct_out` flag to do the same thing for the C bindings. This does not need to be guarded by `WIDL_EXPLICIT_AGGREGATE_RETURNS` because both GCC and MSVC has the same issue and will require the same fix. (I do not know if we need some kind of guard for the mingw-w64 headers to revert to the old function declarations, but given that the old bindings are practically broken perhaps it doesn't matter.) (I have not checked what `/cstruct_out` on MIDL actually does to the C bindings.) -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022

[Bug 53826] Attempting to install / update battle.net fails with Qt error

by WineHQ Bugzilla

https://bugs.winehq.org/show_bug.cgi?id=53826 --- Comment #11 from Kohen Didi <kohen.d(a)gmail.com> --- Happens for me as well, didn't try the patch but tried the workaround and it works, how is this unconfirmed? -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.

Dec. 28, 2022