March 2026 - Wine-GitLab - list.winehq.org

[PATCH v17 0/1] MR10368: vbscript: Implement Eval, Execute, and ExecuteGlobal.
by Francis De Brabandere (＠francisdb) March 31, 2026

March 31, 2026

Eval compiles the string as an expression and returns the result. ExecuteGlobal compiles and executes the string at global scope. Execute compiles and executes the string in the calling scope. When called from within a Sub or Function, Eval and Execute use the calling scope for variable resolution, accessing local variables and arguments. When called from global scope, they fall back to exec_global_code. ExecuteGlobal always executes in global scope. The calling scope is passed through script_ctx_t.caller_exec, which exec_script picks up and stores on exec_ctx_t.caller. The interpreter's lookup_identifier checks the caller's locals, arguments, and dynamic vars when running FUNC_GLOBAL code with a caller context set. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=49908 Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53962 -- v17: vbscript: Implement Eval, Execute, and ExecuteGlobal. https://gitlab.winehq.org/wine/wine/-/merge_requests/10368

3 3

[PATCH v8 0/1] MR10363: vbscript: Support assignment to chained array index expressions.
by Francis De Brabandere (＠francisdb) March 31, 2026

March 31, 2026

Handle expressions like x(0)(1) = 5 where the left side of an assignment is a call expression whose inner expression is also a call (not a direct member). The compiler now evaluates the inner expression as a read, then emits a new OP_assign_call opcode to assign to the resulting array element or default dispatch property. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53877 _Side note: https://bugs.winehq.org/show_bug.cgi?id=58318 is a duplicate of that bug_ -- v8: vbscript: Support assignment to chained array index expressions. https://gitlab.winehq.org/wine/wine/-/merge_requests/10363

2 1

Re: [PATCH v9 0/2] MR10460: msvcrt: Add support for %Z printf specifier.
by Piotr Caban (＠piotr) March 31, 2026

March 31, 2026

Piotr Caban (@piotr) commented about dlls/msvcrt/printf.h: > { > int r, ret; > > + /* UNICODE_STRING passes the number of bytes. */ > + if(len > 0 && flags->Format == 'Z') > + len /= sizeof(wchar_t); I think the code is easier to follow if pf_output_format_wstr always takes length in characters. Please change the caller instead. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10460#note_134549

1 0

Re: [PATCH v9 0/2] MR10460: msvcrt: Add support for %Z printf specifier.
by Piotr Caban (＠piotr) March 31, 2026

March 31, 2026

Piotr Caban (@piotr) commented about dlls/msvcrt/tests/printf.c: > { "% d", " 1", 0, INT_ARG, 1 }, > { "%+ d", "+1", 0, INT_ARG, 1 }, > { "%S", "wide", 0, PTR_ARG, 0, 0, 0, L"wide" }, > + { "%Z", "ansi", 0, PTR_ARG, 0, 0, 0, &(ANSI_STRING){ 4, 4, (char *)"ansi" } }, This is not portable, please define ANSI_STRING/UNICODE_STRING structures separately and use pointers to them here. Please also add %Z format test in swprintf test (to show that it uses ANSI_STRING). Please also check the function behavior with ucrtbase - current implementation will fail there. Its behavior probably changes based on \_CRT_INTERNAL_PRINTF_LEGACY_WIDE_SPECIFIERS or \_CRT_INTERNAL_PRINTF_LEGACY_MSVCRT_COMPATIBILITY flag. Please also add tests to ucrtbase. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10460#note_134548

1 0

[PATCH v3 0/1] MR10521: Certsetcertificatecontextproperty cert ncrypt key handle prop id
by Benoît Legat (＠blegat) March 31, 2026

March 31, 2026

This is part of my attempt to run the [Niko Home Control programming software](https://appdb.winehq.org/objectManager.php?sClass=application&iId… under Wine. This (.NET 8 / Electron app) is performing mTLS with a [Connected controller voor Niko Home Control II](https://products.niko.eu/nl-be/article/550-00003). Doing this, it calls `CertSetCertificateContextProperty` with `CERT_NCRYPT_KEY_HANDLE_PROP_ID`. Wine currently does not implement setting this property so it falls through to the default case, with the \``FIXME("%ld: stub\n", dwPropId);` that lead me to try to implement it. This patch seems to work for the Niko Home Control app so I added test to contribute it upstream. According to [the documentation](https://learn.microsoft.com/en-us/windows/win32/api/wincrypt…, `dwKeySpec` should be set to `CERT_NCRYPT_KEY_SPEC`. This value didn't seem to be defined yet in Wine so I added its definition. [Its value seems to be \`0xffffffff\`](https://android.googlesource.com/toolchain/mingw/+/refs/heads/main/mingw-w64-v6.x/mingw-w64-headers/include/wincrypt.h#3434). I added tests. Before the changes in `dlls/crypt32/cert.c`, running the tests with `wine 11.5-1` (current version on ArchLinux) the tests give 7 failures: ``` $ wine dlls/crypt32/tests/x86_64-windows/crypt32_test-stripped.exe cert 0024:fixme:crypt:add_cert_to_store Unimplemented add disposition 0 0024:fixme:crypt:add_cert_to_store Unimplemented add disposition 0 0024:fixme:crypt:Collection_release Unimplemented flags 2 0024:fixme:crypt:CertContext_SetProperty 78: stub cert.c:553: Test failed: CertSetCertificateContextProperty failed: 80092004 cert.c:559: Test failed: CertGetCertificateContextProperty failed: 80092004 cert.c:560: Test failed: Expected hCryptProv 0xBEEF1234, got 0 cert.c:562: Test failed: Expected dwKeySpec CERT_NCRYPT_KEY_SPEC, got 2 cert.c:569: Test failed: CertGetCertificateContextProperty failed: 80092004 cert.c:570: Test failed: Expected handle 0xBEEF1234, got 0 0024:fixme:crypt:CertContext_SetProperty 78: stub cert.c:587: Test failed: CertSetCertificateContextProperty failed: 80092004 0024:fixme:crypt:CertAddCertificateLinkToStore (00007FFFFE871940, 00007FFFFE8530D8, 00000001, 00007FFFFE1FFB50): semi-stub 0024:fixme:crypt:Collection_release Unimplemented flags 2 0024:fixme:crypt:ProvStore_release Unimplemented flags 2 0024:fixme:crypt:CRYPT_RegCloseStore Unimplemented flags: 00000002 0024:fixme:crypt:ProvStore_release Unimplemented flags 2 0024:fixme:crypt:CRYPT_RegCloseStore Unimplemented flags: 00000002 0024:fixme:bcrypt:BCryptOpenAlgorithmProvider algorithm L"_SHOULDNOTEXIST_" not supported 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:crypt:CryptVerifyCertificateSignatureEx unimplemented for NULL signer 0024:fixme:cryptasn:CRYPT_GetBuiltinEncoder Unimplemented encoder for lpszStructType OID 0 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:crypt:CertGetPublicKeyLength unimplemented for DH public keys 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 7 0020:cert: 649 tests executed (0 marked as todo, 0 as flaky, 7 failures), 0 skipped. ``` After applying my patches like so ``` sudo cp dlls/crypt32/x86_64-windows/crypt32.dll /usr/lib/wine/x86_64-windows/crypt32.dll sudo cp dlls/crypt32/crypt32.so /usr/lib/wine/x86_64-unix/crypt32.so ``` I get 0 failures: ``` $ wine dlls/crypt32/tests/x86_64-windows/crypt32_test-stripped.exe cert 0024:fixme:crypt:add_cert_to_store Unimplemented add disposition 0 0024:fixme:crypt:add_cert_to_store Unimplemented add disposition 0 0024:fixme:crypt:Collection_release Unimplemented flags 2 0024:fixme:crypt:CertAddCertificateLinkToStore (00007FFFFE872F70, 00007FFFFE854988, 00000001, 00007FFFFE1FFB50): semi-stub 0024:fixme:crypt:Collection_release Unimplemented flags 2 0024:fixme:crypt:ProvStore_release Unimplemented flags 2 0024:fixme:crypt:CRYPT_RegCloseStore Unimplemented flags: 00000002 0024:fixme:crypt:ProvStore_release Unimplemented flags 2 0024:fixme:crypt:CRYPT_RegCloseStore Unimplemented flags: 00000002 0024:fixme:bcrypt:BCryptOpenAlgorithmProvider algorithm L"_SHOULDNOTEXIST_" not supported 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:crypt:CryptVerifyCertificateSignatureEx unimplemented for NULL signer 0024:fixme:cryptasn:CRYPT_GetBuiltinEncoder Unimplemented encoder for lpszStructType OID 0 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 1 0024:fixme:crypt:CertGetPublicKeyLength unimplemented for DH public keys 0024:fixme:cryptasn:CryptDecodeObjectEx Unimplemented decoder for lpszStructType OID 7 0020:cert: 649 tests executed (0 marked as todo, 0 as flaky, 0 failures), 0 skipped. ``` Let me know if there is anything else I can do to help, this is my first contribution to this project. -- v3: crypt32: Implement CERT_NCRYPT_KEY_HANDLE_PROP_ID support https://gitlab.winehq.org/wine/wine/-/merge_requests/10521

3 3

[PATCH 0/1] MR10519: dlls/wineandroid.drv: Switch to AHardwareBuffer and raise minSdkVersion to 26
by Twaik Yont (＠twaik) March 31, 2026

March 31, 2026

Switching the Android backend to `AHardwareBuffer` drops the legacy `gralloc` path in favor of the supported and stable NDK API. According to https://apilevels.com/, Android 8.0+ devices already account for about 94.8% of the ecosystem, and the real share is likely higher today, so removing the old `gralloc` code path is a reasonable trade-off. `AHardwareBuffer` does not provide a way to construct a buffer from an existing `native_handle_t`, so buffer transfer is implemented by fully duplicating it through the existing socket-based code path, even for same-process cases. This keeps the implementation consistent with the cross-process path. Function loading is done via `dlopen("libandroid.so")` in `device.c` (instead of expected location in `init.c`). This works around an NDK issue where symbols marked with `__INTRODUCED_IN` cannot be safely referenced when targeting lower API levels, even in `typeof` expressions (I reported it to upstream). To make this work, `__ANDROID_UNAVAILABLE_SYMBOLS_ARE_WEAK__` is used, although this is not ideal. Moving these definitions into shared headers would require enabling this macro for the entire module, which could lead to unexpected runtime linker behavior during development in the future. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10519

2 3

[PATCH v4 0/1] MR9038: Draft: msi: implement MsiProvideAssembly[AW]()
by Ratchanan Srirattanamet (＠ratchanan.sr) March 31, 2026

March 31, 2026

TODO: - make awstring version of MsiProvideComponent() and use that. - Actually implement MsiProvideAssemblyA() using the above. - Write conformance test? Dependency: - Have ACTION_PublishAssemblies() record feature (https://gitlab.winehq.org/wine/wine/-/merge_requests/9037) - Have MsiEnumFeatures() fixed. Currently it seems to just look for features at the wrong place and thus is useless for this purpose. (https://gitlab.winehq.org/wine/wine/-/merge_requests/9041) Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=58578 -- v4: [WIP] msi: implement MsiProvideAssembly[AW]() https://gitlab.winehq.org/wine/wine/-/merge_requests/9038

4 6

[PATCH v8 0/3] MR9992: Draft: Introduce ImeToAsciiEx() user driver function for use in winemac
by Marc-Aurel Zent (＠mzent) March 31, 2026

March 31, 2026

-- v8: winemac: Implement and use macdrv_ImeToAsciiEx(). win32u: Move IME processing to ImeToAsciiEx. win32u: Introduce new ImeToAsciiEx user driver call. https://gitlab.winehq.org/wine/wine/-/merge_requests/9992

3 6

Re: [PATCH v6 0/1] MR10518: wscript: Implement error messages, usage output, and //nologo banner.
by Jacek Caban (＠jacek) March 31, 2026

March 31, 2026

Jacek Caban (@jacek) commented about programs/wscript/main.c: > + print_string(L"Usage: CScript scriptname.extension [option...] [arguments...]"); > + print_string(L""); > + print_string(L"Options:"); > + print_string(L" //B Batch mode: Suppresses script errors and prompts from displaying"); > + print_string(L" //D Enable Active Debugging"); > + print_string(L" //E:engine Use engine for executing script"); > + print_string(L" //H:CScript Changes the default script host to CScript.exe"); > + print_string(L" //H:WScript Changes the default script host to WScript.exe (default)"); > + print_string(L" //I Interactive mode (default, opposite of //B)"); > + print_string(L" //Job:xxxx Execute a WSF job"); > + print_string(L" //Logo Display logo (default)"); > + print_string(L" //Nologo Prevent logo display: No banner will be shown at execution time"); > + print_string(L" //S Save current command line options for this user"); > + print_string(L" //T:nn Time out in seconds: Maximum time a script is permitted to run"); > + print_string(L" //X Execute script in debugger"); > + print_string(L" //U Use Unicode for redirected I/O from the console"); Similar question, should it be stored in resources so it can be translated? -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10518#note_134531

1 0

Re: [PATCH v6 0/1] MR10518: wscript: Implement error messages, usage output, and //nologo banner.
by Jacek Caban (＠jacek) March 31, 2026

March 31, 2026

Jacek Caban (@jacek) commented about programs/wscript/main.c: > if(!filename) { > - WINE_FIXME("No file name specified\n"); > + print_string(L"Input Error: There is no script file specified."); > return 1; > } > + > res = GetFullPathNameW(filename, ARRAY_SIZE(scriptFullName), scriptFullName, &filepart); > if(!res || res > ARRAY_SIZE(scriptFullName)) > return 1; > > + if(GetFileAttributesW(scriptFullName) == INVALID_FILE_ATTRIBUTES) { > + WCHAR buf[MAX_PATH + 64]; > + swprintf(buf, ARRAY_SIZE(buf), L"Input Error: Can not find script file \"%s\".", scriptFullName); > + print_string(buf); > + return 1; > + } This should probably be part of the error handling when opening the file. We could move reading the script file earlier if needed. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10518#note_134532

1 0