Mark Jansen

[WIN32SS] Fix returning an uninitialized variable. Patch by Victor Martinez Calvo. CID 1341417, CORE-11027 #comment Thanks!
    • -1
    • +1
    /trunk/reactos/win32ss/user/ntuser/hook.c
[ACPPAGE] Implement custom compatibility mode selection + Expand paths CORE-10375
    • -0
    • +7
    /trunk/reactos/sdk/lib/atl/atlexcept.h
    • -0
    • +111
    /trunk/reactos/dll/shellext/acppage/CLayerStringList.hpp
[ACPPAGE] Update translations based on slayer. CORE-10375

Translators, please check your language, some text was changed!

    • -0
    • +46
    /trunk/reactos/dll/shellext/acppage/lang/sk-SK.rc
    • -0
    • +41
    /trunk/reactos/dll/shellext/acppage/lang/es-ES.rc
    • -0
    • +41
    /trunk/reactos/dll/shellext/acppage/lang/it-IT.rc
    • -0
    • +47
    /trunk/reactos/dll/shellext/acppage/lang/uk-UA.rc
    • -0
    • +39
    /trunk/reactos/dll/shellext/acppage/lang/no-NO.rc
    • -0
    • +39
    /trunk/reactos/dll/shellext/acppage/lang/bg-BG.rc
    • -0
    • +39
    /trunk/reactos/dll/shellext/acppage/lang/eu-ES.rc
    • -0
    • +43
    /trunk/reactos/dll/shellext/acppage/lang/sq-AL.rc
    • -0
    • +39
    /trunk/reactos/dll/shellext/acppage/lang/id-ID.rc
    • -0
    • +41
    /trunk/reactos/dll/shellext/acppage/lang/he-IL.rc
    • -0
    • +39
    /trunk/reactos/dll/shellext/acppage/lang/fr-FR.rc
  1. … 9 more files in changeset.
[SHELL32] Fix for BuildPathsList, this solves some problems showing while copying folders. Patch by 'Gleb'. CORE-8434
[WIN32SS] Add full name support in FontFamilyFillInfo. Patch by Katayama Hirofumi MZ. CORE-10876
    • -16
    • +30
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
[WIN32SS] Add extra logging to failed SURFACE_AllocSurface calls. CORE-13036
    • -3
    • +6
    /trunk/reactos/win32ss/gdi/eng/surface.c
[BROWSEUI] Port some IAutoComplete changes from wine + fix a memory leak. This allows the dropdown to show with ACO_AUTOSUGGEST and an empty input.
    • -36
    • +73
    /trunk/reactos/dll/win32/browseui/CAutoComplete.cpp
If we are changing this, might as well add spaces around the &

If we are changing this, might as well add spaces around the &

Can we test this?

Can we test this?

This should be covered by the test.

This should be covered by the test.

INVALID_FILE_ATTRIBUTES

INVALID_FILE_ATTRIBUTES

Fix copypasta error, spotted by Thomas.
    • -1
    • +1
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
[WIN32SS] Addendum to r74312 and r74309, fix some bad memory leaks. CORE-10876

Thanks Giannis and Thomas :)

    • -1
    • +3
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
If you need this for debugging, please indicate where you need it and how it would help. The current amount of spam in this file is already tremendous when uncommenting NDEBUG, and it makes output ...

If you need this for debugging, please indicate where you need it and how it would help.
The current amount of spam in this file is already tremendous when uncommenting NDEBUG, and it makes output already completely useless imo.

Will do in a separate commit.

Will do in a separate commit.

[WIN32SS] Fix charset in font enumeration. Patch by Katayama Hirofumi MZ. CORE-10876
    • -2
    • +0
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
[WIN32SS] Initial fixes for localized font enumeration. Patch by Katayama Hirofumi MZ. CORE-10876
    • -61
    • +96
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
[WIN32SS] Count number of faces added instead of fonts in IntGdiLoadFontsFromMemory.

This fixes displaying embedded fonts in firefox.

CORE-13046 CORE-13056

    • -9
    • +10
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
[WIN32SS] Cleanup fonts at process destruction + implement font memory reference counting.

Thanks to everyone involved in reviewing this code! (See CR-112)

CORE-13056

    • -0
    • +2
    /trunk/reactos/win32ss/gdi/ntgdi/init.c
    • binary
    /trunk/rostests/apitests/gdi32/Shadows_Into_Light.ttf
    • -0
    • +1
    /trunk/reactos/win32ss/gdi/ntgdi/text.h
    • -2
    • +1
    /trunk/reactos/win32ss/gdi/ntgdi/font.h
    • -61
    • +250
    /trunk/reactos/win32ss/gdi/ntgdi/freetype.c
    • -2
    • +10
    /trunk/reactos/win32ss/gdi/eng/engobjects.h
    • binary
    /trunk/rostests/apitests/gdi32/TTCTestV.ttc
[WIN32SS] Add missing \n to DPRINT1 Spotted by Thomas Schneider.
    • -1
    • +1
    /trunk/reactos/win32ss/gdi/eng/pdevobj.c
With an @ prefix is not found. I suspect that the @ prefix is only used when there are 2 fonts with the same face name in one package, where one font is vertical.

With an @ prefix is not found.
I suspect that the @ prefix is only used when there are 2 fonts with the same face name in one package, where one font is vertical.

NTSTATUS NTAPI suggests that this can be called from external code as well, it might make sense to assert on Index being in an acceptable range, since we index an array with it.

NTSTATUS NTAPI suggests that this can be called from external code as well,
it might make sense to assert on Index being in an acceptable range, since we index an array with it.

We don't care if this fails?

We don't care if this fails?

This is what windows font viewer reports: http://i.imgur.com/TqohWEv.png

This is what windows font viewer reports:
http://i.imgur.com/TqohWEv.png

Cannot commit this separately, it crashes the old code.

Cannot commit this separately, it crashes the old code.

New patch introducing SHARED_MEM to keep track of the memory. Add tests showing that we should count faces, not charsets! This allows to remove the firefox hack. Also introducing macro's to verify ...

New patch introducing SHARED_MEM to keep track of the memory.
Add tests showing that we should count faces, not charsets!
This allows to remove the firefox hack.
Also introducing macro's to verify locking, thanks to thomas!

I didnt test it, simply added then forgot about it. Ideally we should use TRACE, INFO, WARN macro's to use debug levels indeed. Let's save that for another time. For future reference: DbgChGdiFont

I didnt test it, simply added then forgot about it.
Ideally we should use TRACE, INFO, WARN macro's to use debug levels indeed.
Let's save that for another time.

For future reference: DbgChGdiFont

Function is recursive, we would need to pass it as arguments then.

Function is recursive, we would need to pass it as arguments then.

Tracked in CORE-13064

Tracked in CORE-13064