Commit graph

80691 commits

Author SHA1 Message Date
Hermès Bélusca-Maïto 6b7b0f0c7e
[WIN32SS:NTUSER] Few formatting fixes. 2021-10-22 21:44:52 +02:00
Justin Miller be223b9de7
[SDK] Allow ARM64 MSVC to complete configuration (#4045)
- Add some of the missing CMake adjustments to continue the configure and compile process with ARM64 MSVC
- Created quick stubs for the functions in SDK needed to finish the configuration process
- Put in an ARM64 option for spec2def

CORE-17518 CORE-17615
2021-10-22 18:52:32 +03:00
Serge Gautherie 4b0249e98a
[GITHUB] build.yml: Update LLVM version to 13 (#4054) 2021-10-22 18:17:06 +03:00
Eric Kohl 1ade494a70 [ADVAPI32] Fix several corner cases in RegOpenKey* functions
This fixes the remaining RegOpenKey* tests.
2021-10-21 14:45:56 +02:00
Katayama Hirofumi MZ a37d9a4e14
[IMM32] Rewrite ImmInstallIMEW (#4044)
- Add Imm32StrToUInt and Imm32UIntToStr helper functions.
- Add Imm32LoadImeVerInfo, Imm32GetRegImes, Imm32WriteRegIme, Imm32GetNextHKL, Imm32CopyFile helper functions.
- Add REG_IME structure for registered IMEs.
- Rewrite ImmInstallIMEW function.
- Improve ImmLoadLayout and Imm32LoadImeInfo functions.
CORE-11700
2021-10-21 10:28:04 +09:00
Konrad Dybcio d26409cc34
[SDK][MKISOFS] xconfig.h: Don't define VA_LIST_IS_ARRAY on Linux/ARM(64) (#4012)
For some bizarre reason ARM toolchains have some peculiar issues with various va_arg macros,
as this is not the first time one of these is problematic (though I don't remember the previous one,
it was a long, long time ago ™️).. They always seem to expect va_list as an argument though, so
let's make them happy and by extension make ros-tools compile on Linux/ARM(64) compile again!
2021-10-20 22:12:23 +03:00
Katayama Hirofumi MZ 396b49436c
[IMM32_APITEST] Fix himc testcase crash (#4036)
- Add NULL checks for pIC.
CORE-11700
2021-10-20 20:23:38 +09:00
James Tabor be8e5fc7b3 [User32-winetest] Remove ReactOS to do.
N0 longer needed.
2021-10-19 18:49:52 -05:00
James Tabor 83f86b8fdb [User32] Fix incorrect check.
Use Lo Word instead of Intersource check.
Fix wine user32 tests.
2021-10-19 15:29:51 -05:00
Katayama Hirofumi MZ 037c744eb1
[EXPLORER] Send WM_POPUPSYSTEMMENU asynchronously (#4047)
- Delete useless SendMessageTimeout:WM_NULL call.
- Add useful SendMessageCallbackW:WM_POPUPSYSTEMMENU call.
- In the callback function of SendMessageCallbackW, do PostMessageW:WM_NULL to properly handle the popup menu.
CORE-16353
2021-10-19 07:17:01 +09:00
Hervé Poussineau 46fbc6f432 [NTOS:PNP] Fix crash when removing a device without resources
This fixes commit 89fd2b86e4
2021-10-18 22:23:49 +02:00
Katayama Hirofumi MZ aa8fc872a0
[IMM32][IMM32_WINETEST] Re-implement ImmSetCompositionStringA/W (#4040)
- Modify ImmSetCompositionStringA and ImmSetCompositionStringW prototypes (removing const of two arguments).
- Add Imm32OpenICAndCS helper function.
- Implement Imm32SetCompositionStringAW function.
- Modify imm32_winetest (due to removal of const).
CORE-11700
2021-10-17 19:32:48 +09:00
James Tabor e98684ed8b [GDI32] Fix rectangle flag issues.
Fix CORE-17815. Remove dumb dumb code breakers.
2021-10-16 11:16:43 -05:00
Stanislav Motylkov b3b1bd66a0
[CONFIGURE][CMAKE] Add support for MSVC ARM64 target
Addendum to e4feaa1a and 51e2ab52. CORE-17518
2021-10-16 16:30:36 +03:00
Konrad Dybcio b7c1b95b45
[SDK] gcc.cmake: Only add -mstackrealign on i386 (#4016)
This option is only useful for the i386 architecture and breaks ARM builds.

Reviewed-by: Victor Perevertkin <victor.perevertkin@reactos.org>
Reviewed-by: Stanislav Motylkov <x86corez@gmail.com>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
2021-10-16 11:58:08 +03:00
Katayama Hirofumi MZ ea4c1a067b
[SETUP][REACTOS] Improve Japanese translation (#4033)
Improve the Japanese GUI of ReactOS Setup.
2021-10-16 07:51:40 +09:00
Konrad Dybcio 30c7c161fc
[SDK] host-tools.cmake: Allow cross-compilation from aarch64 (#4013)
Some compilers call it 'arm64', while others prefer 'aarch64'. It's a big mess,
thanks Arm Ltd...

Reviewed-by: Stanislav Motylkov <x86corez@gmail.com>
Reviewed-by: Serge Gautherie <reactos-git_serge_171003@gautherie.fr>
Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
2021-10-16 01:38:33 +03:00
Konrad Dybcio dff479c82f
[HHPCOMP] lz_nonslide.h: Define u_char on Linux (#4011)
Not defining the type breaks the compilation on (at least arm) Linux. Add a check for Linux-as-build-host to make it pass.

Signed-off-by: Konrad Dybcio <konrad.dybcio@somainline.org>
2021-10-16 01:30:22 +03:00
Katayama Hirofumi MZ 04cb13bc57
[IMM32] Refactor about reconversion (#4031)
- Delete Imm32ReconvertSize and Imm32ConvertReconvert helper functions.
- Add Imm32ReconvertAnsiFromWide and Imm32ReconvertWideFromAnsi helper functions.
CORE-11700
2021-10-15 22:46:58 +09:00
Katayama Hirofumi MZ fec35dc893
[IMM32] Improve IchWideFromAnsi and IchAnsiFromWide (#4028)
- Use LONG instead of DWORD for indexing and counting the characters.
- Use the sign for the loop condition.
- Add NUL check.
CORE-11700
2021-10-15 10:31:08 +09:00
Hervé Poussineau 89fd2b86e4 [NTOS:PNP] HACK: release resources when device is removed
CORE-17789
2021-10-14 23:39:31 +02:00
Hervé Poussineau 49358f3416 [NTOS:PNP] Fix resource conflict detection
Only resources of HAL were checked against conflicts, not those of PnP Manager

Let IoReportResourceForDetection() make a silent conflict check.
Otherwise IopCheckResourceDescriptor() will always return 'no conflict'.

CORE-17789
2021-10-14 23:39:31 +02:00
Hervé Poussineau ea8cbbd4ca [VIDEOPRT] Correctly return failure in IRP_MJ_CREATE when HwInitialize fails
CORE-17789
2021-10-14 23:39:31 +02:00
Hervé Poussineau 807af6a4d1 [VIDEOPRT] Correctly return failure when adapter can't be found
This was wrong in both call chains (PNP case or legacy case)
HwFindAdapter returns a VP_STATUS, while we were expecting a NTSTATUS.

CORE-17789
2021-10-14 23:39:31 +02:00
Hervé Poussineau 87b74f2901 [VIDEOPRT] Correctly report legacy resources in IRP_MN_FILTER_RESOURCE_REQUIREMENTS
It was broken since first commit 25eae4b916 (r54237)

CORE-17789
2021-10-14 23:39:31 +02:00
Hervé Poussineau 951dc66c66 [VGAMP] Correctly check and report legacy resources (VGA I/O ports and memory)
As we are a non PNP driver, the call chain will be
DriverEntry -> VidePortInitialize -> VideoPortFindAdapter -> HwFindAdapter.

If legacy resources are available, we will fail VGAFindAdapter,
so DriverEntry will fail, so vgamp.sys driver won't be used.

CORE-17789
2021-10-14 23:39:30 +02:00
Hervé Poussineau 6dd94572e5 [VBEMP] Correctly report legacy resources (VGA I/O ports and memory)
CORE-17789
2021-10-14 23:39:30 +02:00
Hervé Poussineau 3379de3924 [VBEMP] Fail HwInitialize (ie IRP_MJ_CREATE) when /NOVESA switch is present
In that case, we want a simple VGA driver to drive the graphic adapter.

CORE-17789
2021-10-14 23:39:30 +02:00
Hervé Poussineau ac13f37d0a [ISAPNP] Generate a name for PDO, as required 2021-10-14 23:39:30 +02:00
Hervé Poussineau a86c3794a6 [NTOS:IO] Remove final NULL char of PDO name before writing to registry
Otherwise, if a PDO has no name (bad!), you'll see two "(Default)" entries
in HKLM\HARDWARE\RESOURCEMAP\PnP Manager\PnpManager
2021-10-14 23:39:30 +02:00
Hervé Poussineau 500f00de6a [DEVMGR] Display current assigned resources, instead of boot resources 2021-10-14 23:39:30 +02:00
Doug Lyons cee171f5d2
[SHELL32] Bypass OpenWith for Delete and F2 (Rename) Keys (#4032) CORE-17810
Logspam started during 0.4.9-dev'ing when zipfldr got committed, logspam became much worse by
0.4.10-dev-599-g 932df378bf
But the actual bug of 'calling those functions when not needed' is much older than that.

I do intend to port this improvement back into older rls-branches.
2021-10-14 23:03:53 +02:00
Victor Perevertkin 55f1f3000a
[CMAKE] Remove /fallback compile flag from clang-cl builds
This option is not needed anymore and was removed from LLVM 13
2021-10-14 20:30:27 +03:00
James Tabor 63e7d6c29e [Win32K] Fix typo
Fix a GERRRR typo.
2021-10-14 09:59:53 -05:00
Katayama Hirofumi MZ 6e356a9b6c [NOTEPAD] Improve Japanese translation 2021-10-14 13:08:38 +09:00
Katayama Hirofumi MZ bfb7bd05a4
[IMM32] Re-implement ImmGetCompositionStringA/W (#4026)
- Re-implement ImmGetCompositionStringA and ImmGetCompositionStringW functions.
- Add Imm32CompStrAnsiToWide, Imm32CompStrWideToAnsi, Imm32CompAttrWideToAnsi, Imm32CompAttrAnsiToWide, Imm32CompClauseAnsiToWide, and Imm32CompClauseWideToAnsi helper functions.
CORE-11700
2021-10-14 08:07:06 +09:00
Mark Jansen 143b2a5035
[RAPPS] Speed up installed application loading
Only load in details when we are going to display them
2021-10-13 19:27:40 +02:00
Wu Haotian 0ed04e3640
[TASKMGR] Make performance graph grid scroll (#3581)
The grids are now rolling along with performance graphs.
2021-10-13 19:14:29 +03:00
Wu Haotian a4ab9a1e19
[TASKMGR] Formatting only (#3581) 2021-10-13 19:14:20 +03:00
Victor Perevertkin 43f1d91687
[NTOS:PNP] Fix resource conflict detection
Previous code did not detect equal resource ranges as conflicting.
Thanks Hervé Poussineau for pointing this out!

Meanwhile, simplify the code to make it more readable.
2021-10-13 00:00:25 +03:00
Chan Chilung 025f58a051
[NOTEPAD] Fix access key for "New Window" menu item (#4024) 2021-10-12 15:34:34 +03:00
Katayama Hirofumi MZ d14d3dedcb
[IMM32] Move Imm(Get|Set)CompositionStringA/W (#4021)
- Add new file compstr.c.
- Move ImmGetCompositionStringA, ImmGetCompositionStringW, ImmSetCompositionStringA, and ImmSetCompositionStringW functions into compstr.c.
- Half-implement ImmGetCompositionStringA/W.
CORE-11700
2021-10-12 13:10:01 +09:00
Chan Chilung 570cedf176
[SETUPLIB][INPUT.CPL] Move zh-MO to ChineseTraditionalFonts group (#4010)
- The use of Traditional Chinese character is far more than Simplified one according to Wikipedia
- The code page used on zh-MO is 950, which is Traditional Chinese's one
- zh-MO is described as Traditional Chinese here: https://support.microsoft.com/en-us/windows/microsoft-traditional-chinese-ime-ef596ca5-aff7-4272-b34b-0ac7c2631a38#ID0EBBD=Microsoft_ChangJie/Quick
2021-10-11 16:24:21 +03:00
Serge Gautherie 51e2ab52af [CMAKE] msvc.cmake: Add arm64 CMAKE_ASM_MASM_COMPILER option support
CORE-17518
2021-10-11 15:24:30 +03:00
Serge Gautherie e4feaa1a69 [CMAKE] CMakeLists.txt: Add arm64 to ARCH allowed values
CORE-17518
2021-10-11 15:20:34 +03:00
Katayama Hirofumi MZ 9078e34f3c
[SHELL32] Don't smartass-renaming on LNK files (#4020)
Renaming .LNK files had a bug that the selection won't include the extension-like filename part.
CORE-17807
2021-10-11 13:33:43 +09:00
Konrad Dybcio 3c36cde874
[SDK] config.cmake: Fix generic mtune flag (#4017) 2021-10-11 03:46:12 +03:00
Robert Naumann 38b9688d45 [IERNONCE] Add German translation 2021-10-11 00:08:19 +02:00
Robert Naumann 4e5301d09f [INETCPL] Update German translation 2021-10-11 00:07:39 +02:00
Joachim Henze f1d4a83bf7 [NOTEPAD] Properly fix accelerator issue in es-ES.rc CORE-17809
In 0.4.14-dev-1321-g d6e0fe78e7
we introduced the following compiler warning for MSVC2010 release build:

reactos/base/applications/notepad/lang/es-ES.rc(12) : warning RC4204 : ASCII character not equivalent to virtual key code

That warning was muted without investing too much love by
0.4.15-dev-2469-g d0bb775774

Today we apply the proper fix that brings us back in sync with other languages.
I do intend to port the proper fix back to 0.4.14RC as well.
2021-10-10 21:29:05 +02:00