Ros-dev October 2011

ros-dev@reactos.org

31 participants
27 discussions

Re: [ros-dev] [ros-diffs] [akhaldi] 54136: [CMD] * Reduce the scope of some variables.
by Alex Ionescu 14 Oct '11

14 Oct '11

Why? It doesn't change anything...and... This is not our coding standard. Best regards, Alex Ionescu On Fri, Oct 14, 2011 at 1:50 PM, <akhaldi(a)svn.reactos.org> wrote: > Author: akhaldi > Date: Fri Oct 14 17:50:16 2011 > New Revision: 54136 > > URL: http://svn.reactos.org/svn/reactos?rev=54136&view=rev > Log: > [CMD] > * Reduce the scope of some variables. > > Modified: > trunk/reactos/base/shell/cmd/assoc.c > trunk/reactos/base/shell/cmd/dir.c > trunk/reactos/base/shell/cmd/filecomp.c > trunk/reactos/base/shell/cmd/start.c > trunk/reactos/base/shell/cmd/ver.c > > Modified: trunk/reactos/base/shell/cmd/assoc.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/base/shell/cmd/assoc.c?rev… > > ============================================================================== > --- trunk/reactos/base/shell/cmd/assoc.c [iso-8859-1] (original) > +++ trunk/reactos/base/shell/cmd/assoc.c [iso-8859-1] Fri Oct 14 17:50:16 > 2011 > @@ -206,8 +206,6 @@ > INT CommandAssoc (LPTSTR param) > { > > - LPTSTR lpEqualSign = NULL; > - > /* print help */ > if (!_tcsncmp (param, _T("/?"), 2)) > { > @@ -221,7 +219,7 @@ > PrintAllAssociations(); > else > { > - lpEqualSign = _tcschr(param, _T('=')); > + LPTSTR lpEqualSign = _tcschr(param, _T('=')); > if(lpEqualSign != NULL) > { > LPTSTR fileType = lpEqualSign + 1; > > Modified: trunk/reactos/base/shell/cmd/dir.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/base/shell/cmd/dir.c?rev=5… > > ============================================================================== > --- trunk/reactos/base/shell/cmd/dir.c [iso-8859-1] (original) > +++ trunk/reactos/base/shell/cmd/dir.c [iso-8859-1] Fri Oct 14 17:50:16 > 2011 > @@ -1260,13 +1260,11 @@ > LPDIRSWITCHFLAGS lpFlags) /* [IN] The flags that > we will use to sort */ > { > LPWIN32_FIND_DATA lpTemp; /* A temporary pointer */ > - int First, Last, Temp; > BOOL Way; > > if (i < j) > { > - First = i; > - Last = j; > + int First = i, Last = j, Temp; > Way = TRUE; > while (i != j) > { > > Modified: trunk/reactos/base/shell/cmd/filecomp.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/base/shell/cmd/filecomp.c?… > > ============================================================================== > --- trunk/reactos/base/shell/cmd/filecomp.c [iso-8859-1] (original) > +++ trunk/reactos/base/shell/cmd/filecomp.c [iso-8859-1] Fri Oct 14 > 17:50:16 2011 > @@ -209,7 +209,6 @@ > TCHAR path[MAX_PATH]; > TCHAR fname[MAX_PATH]; > TCHAR directory[MAX_PATH]; > - UINT longestfname = 0; > SHORT screenwidth; > > /* expand current file name */ > @@ -277,6 +276,7 @@ > hFile = FindFirstFile (path, &file); > if (hFile != INVALID_HANDLE_VALUE) > { > + UINT longestfname = 0; > /* Get the size of longest filename first. */ > do > { > > Modified: trunk/reactos/base/shell/cmd/start.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/base/shell/cmd/start.c?rev… > > ============================================================================== > --- trunk/reactos/base/shell/cmd/start.c [iso-8859-1] (original) > +++ trunk/reactos/base/shell/cmd/start.c [iso-8859-1] Fri Oct 14 17:50:16 > 2011 > @@ -48,7 +48,6 @@ > TCHAR szFullCmdLine [CMDLINE_LENGTH]; > PROCESS_INFORMATION prci; > STARTUPINFO stui; > - INT i = 0; > #ifdef UNICODE > DWORD dwCreationFlags = CREATE_NEW_CONSOLE | > CREATE_UNICODE_ENVIRONMENT; > #else > @@ -213,7 +212,7 @@ > /* Parsing the command that gets called by start, and it's > parameters */ > { > BOOL bInside = FALSE; > - > + INT i; > /* find the end of the command and put the arguments in > param */ > for (i = 0; Rest[i]; i++) > { > > Modified: trunk/reactos/base/shell/cmd/ver.c > URL: > http://svn.reactos.org/svn/reactos/trunk/reactos/base/shell/cmd/ver.c?rev=5… > > ============================================================================== > --- trunk/reactos/base/shell/cmd/ver.c [iso-8859-1] (original) > +++ trunk/reactos/base/shell/cmd/ver.c [iso-8859-1] Fri Oct 14 17:50:16 > 2011 > @@ -28,8 +28,6 @@ > VOID ShortVersion (VOID) > { > OSVERSIONINFO VersionInfo; > - unsigned RosVersionLen; > - LPTSTR RosVersion; > > ConOutResPrintf(STRING_CMD_SHELLINFO, _T(KERNEL_RELEASE_STR), > _T(KERNEL_VERSION_BUILD_STR)); > VersionInfo.dwOSVersionInfoSize = sizeof(OSVERSIONINFO); > @@ -37,6 +35,9 @@ > memset(VersionInfo.szCSDVersion, 0, > sizeof(VersionInfo.szCSDVersion)); > if (GetVersionEx(&VersionInfo)) > { > + LPTSTR RosVersion; > + unsigned RosVersionLen; > + > RosVersion = VersionInfo.szCSDVersion + > _tcslen(VersionInfo.szCSDVersion) + 1; > RosVersionLen = sizeof(VersionInfo.szCSDVersion) / > sizeof(VersionInfo.szCSDVersion[0]) - > (RosVersion - VersionInfo.szCSDVersion); > > >

2 1

RosBE-Unix 2.0-RC1
by Colin Finck 13 Oct '11

13 Oct '11

Hi all, A first release candidate of RosBE-Unix 2.0 is now available on http://svn.reactos.org/RosBE-Temp/. Compared to 1.5, it finally includes CMake 2.8.5 (with Jerome's patches), an updated GNU Make and new GMP and MPFR libraries. For the latter ones, I've added a patch to fix building under Mac OS X 10.7. Please try it out and tell me about your results. If there's anything that needs to be fixed before the release, please tell me in a reply to this mail. For the reference, the current CMake building commands in a ReactOS checkout are as follows: - ./configure.sh - cd output-i386-MinGW/host-tools - makex - cd ../reactos - makex bootcd I'll also set up this Build Environment on the Debug Buildslave as soon as possible. Happy building, Colin

5 6

Bugfixing mode
by Aleksey Bragin 11 Oct '11

11 Oct '11

Hello! As per discussion during the last status meeting, I proposed to set up a deadline for having a 0.3.14 Release Candidate available by the time of the next meeting. Olaf, Amine and others were kind enough to make an up-to-date list of critical bugs and regressions to be fixed. It's available here: http://www.reactos.org/wiki/Buglist (indeed this page should be in your web browser's Favorites!) . I want to ask all of you to concentrate on fixing regressions and critical/blocker bugs first, before going into fancy features or implementing something else. It's really important, we need to make 0.3.14 as the last release before switching to 0.4. We have all necessary prerequisites, and just need to sit down and fix regressions. I personally put off all of my ReactOS-related "fancy" projects because I want a solid trunk first, and I am working only on stuff which is urgently needed - fixing remaining LDR rewrite regressions, special pool to catch pool corruptions, etc. It's meaningless to work on something else now if regressions aren't fixed. Hopefully you will follow my advice and we will have a wonderful bugfixing week(s)! Thank you! WBR, Aleksey Bragin.

3 4

freeldr.ini now is overwritten by newly generated (livecd)
by Minas Abrahamyan 11 Oct '11

11 Oct '11

Hi all, Previously I used edited version of freeldr.ini to build livecd ISOs, it is in \output-i386\livecd\ Now freeldr.ini is becoming overwritten in make process by newly generated "standard" one, and it is overwriting my manually edited version. So 1) what is now the correct way is to have custom freeldr.ini in iso image? 2) For what are these new automagical erase-and-newly-generate scripts? Although I can now manually edit (in iso image) what has been done earlier automatically, replace the file, but, at my PoV something goes wrong. -- Best Regards, M.A.

1 1

[ros-diffs] [rharabien] 54006: [WIN32K] - Rewrite keyboard input processing to match windows behavior -
by James Tabor 11 Oct '11

11 Oct '11

Hi! Well if anyone cares! This commit message is a lie! In IntKeyboardInput I added support for KF_DLGMODE and KF_MENUMODE ! BTW is used! In Dialog.c DF_DIALOGACTIVE! So someone will fix it right? Okay, James

1 0

CMake adoption
by caemyr＠myopera.com 09 Oct '11

09 Oct '11

You might know me as a stout supporter of CMake transition, but a single experience shattered my certainity. Today i tried to resolve a backtrace for a bug report, done with a cmake iso. The results were horrendous. The backtrace was lenghty, but on rbuild such task would take around 15 mintues for me, having all files extracted and prepped log.. At the present moment, cmake bt resolving is a total mess and practically prevents anyone doing such log. I might not be a programmer, but was somewhat skilled in my testers duties, also as some recall, i loved doing manual traces by raddr2line. With cmake - i hate it: 1. the whole process takes minutes more per frame: you need the base address, add the offset and type in the addr2line command 2. base address from kdbg: mod is not reliable, as modules might be relocated. addr2line doesn't know about that... 3. C++ function names are not demangled. For those modules that are written in c++ you will only get source lines, unless you check them up manually. 4. Being somewhat experienced, i see no way how i`m to explain newcomer how to do a backtrace... on the other hand, do you expect anyone to waste half an hour doing so? This HAS to be resolved before rbuild is put to retirement. Amine was talking about getting back to rsym - sure, why not? Anything that works will be better than current situation. -- With best regards Caemyr

5 6

Re: [ros-dev] [ros-diffs] [jimtabor] 54044: [Win32k] - Implement get atom name for user32.
by Timo Kreuzer 07 Oct '11

07 Oct '11

Am 07.10.2011 15:56, schrieb jimtabor(a)svn.reactos.org: > > +DWORD > +APIENTRY > +NtUserGetAtomName( > + ATOM nAtom, > + PUNICODE_STRING pBuffer) > +{ > + DWORD Ret; > + UNICODE_STRING CapturedName = {0}; > + UserEnterShared(); > + _SEH2_TRY > + { > + CapturedName = ProbeForReadUnicodeString(pBuffer); > + Ret = IntGetAtomName((RTL_ATOM)nAtom, CapturedName.Buffer, (ULONG)CapturedName.Length); > + } > + _SEH2_EXCEPT(EXCEPTION_EXECUTE_HANDLER) > + { > + Ret = 0; > + } > + _SEH2_END > + UserLeave(); > + return Ret; > +} IntGetAtomName must not be called from inside SEH, if it throws you'll be left with a locked atom table. Timo

1 0

Re: [ros-dev] [ros-diffs] [cgutman] 54016: [HAL] - Implement support for scatter/gather DMA - Tested with Intel's NDIS 5.1 PRO/1000 driver (but more testing would be helpful to weed out bugs) [NDIS] - Report NDIS 5.1 now si...
by Colin Finck 06 Oct '11

06 Oct '11

cgutman(a)svn.reactos.org wrote: > - Implement support for scatter/gather DMA This commit deserves a huge LIKE button, thanks! Will try it out as soon as I have access to my ReactOS laptop again. Finally a chance for getting its Ethernet port to work :-) - Colin

5 4

[ros-diffs] [gadamopoulos] 53751: [comctl32]
by James Tabor 06 Oct '11

06 Oct '11

Hi!! This is wrong, 8^( + class.style |= CS_GLOBALCLASS; + class.hInstance = COMCTL32_hModule; The point in THEMING_Initialize was to get the current global classes and make locals for each process. So that process could have it. Not the whole system! Note from a msdn blog, "If you pass the CS_GLOBALCLASS flag when registering the class, then the window manager will ignore the instance handle when looking for your class. All of the USER32 classes are registered as global." I spent a week now looking through it all, ReactOS is handling the class registration correctly. My best guess, if themes is active do as the original code had it before and use IsThemeActive like it is setup. Keep class.hInstance = COMCTL32_hModule; and don't use CS_GLOBALCLAS so you can find it again. Thanks, James

2 2

Proposal for deletion
by Pierre Schweitzer 05 Oct '11

05 Oct '11

Hi, today, after looking at the daily source analysis we have, I found out that all that stuff could be improved by lightening trunk. We have huge amount of unused code (even not linked in CMake) rotting in trunk. I'm mainly talking about icu4ros. AFAIK it was an attempt from KJK for I even don't recall what. Here is my proposal: we do a branch for this. And then, we remove it from trunk. This will make things lighter, without loosing code. Any objections/comments? Regards, Pierre

3 3

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev October 2011