Ros-dev March 2017

ros-dev@reactos.org

39 participants
31 discussions

Re: [ros-dev] New ideas added to GSoC Ideas list
by Ștefan Fulea 15 Mar '17

15 Mar '17

Marți, 14 martie 2017 12:00:02 +0000, Mark Jansen <learn0more+ros(a)gmail.com> a scris: > How about a better way to translate ros? > For example integrating .po files with our rc files (possibly needs a > preprocess step or something tho), > Or creating a resource editor that allows multiple files to be edited > at once? > Please don't push for .po resources, for these are not better. As for improving the process, I doubt it'd repay the investment. After the initial translation effort, the further maintenance requires very little.

5 6

Regarding Smart Commits
by Colin Finck 15 Mar '17

15 Mar '17

Hi all! Some people have noticed that Smart Commits (#resolve, #comment, etc.) have stopped working after the JIRA/FishEye upgrade. This is due to the new per-user OAuth authentication between both systems, which means that every committer needs to authorize FishEye once to impersonate as your JIRA account. FishEye only tells you this and the URL to fix it after the first failed Smart Commit, but here it is: https://code.reactos.org/plugins/servlet/applinks/oauth/login-dance/authori… Click that, confirm and all subsequent Smart Commits should work. Make sure you are logged into JIRA and FishEye before. Cheers, Colin

1 0

Re: [ros-dev] New ideas added to GSoC Ideas list
by Alex Ionescu 13 Mar '17

13 Mar '17

I would like to see support for GPT partititioning :) Best regards, Alex Ionescu On Sun, Mar 12, 2017 at 3:51 PM, Javier Agustìn Fernàndez Arroyo < elhoir(a)gmail.com> wrote: > i dont think anyone has any problems on downloading a 100MB ISO file.... > but... what about any kind of minimal network install ISO file support? > > On Sun, Mar 12, 2017 at 11:34 PM, Thomas Mueller <mueller6723(a)twc.com> > wrote: > >> > Victor thought also about adding registry hive healing. >> > Hermès. >> >> > -----Message d'origine----- >> > De : Ros-dev [mailto:ros-dev-bounces@reactos.org] De la part de Colin >> Finck >> > Envoyé : dimanche 12 mars 2017 17:27 >> > À : 'ReactOS Development List' >> > Objet : [ros-dev] New ideas added to GSoC Ideas list >> >> > Hi all! >> >> > Daniel and me collected some additional ideas for GSoC today, which >> I've added to our Wiki Ideas list: >> > https://reactos.org/wiki/Google_Summer_of_Code_2017_Ideas >> >> > In particular: >> > * Fundamental WiFi components >> > * USBXHCI driver for supporting USB 3.x controllers >> > * Bluetooth Stack >> > * WebKit-based MSHTML implementation >> >> > I'm open for comments and suggestions! >> > We can still add ideas until March 20, so let's give students a large >> pool to draw from. >> >> > Cheers, >> >> > Colin >> >> Would GPT partitioning support for ReactOS be appropriate for GSoC? >> >> Also, the ability to build or cross-build ReactOS and install directory >> to a FAT32 partition, mounted on a directory, without having to burn to CD >> and boot/install from there: would that be appropriate? >> >> Tom >> >> >> >> >> _______________________________________________ >> Ros-dev mailing list >> Ros-dev(a)reactos.org >> http://www.reactos.org/mailman/listinfo/ros-dev >> > > > _______________________________________________ > Ros-dev mailing list > Ros-dev(a)reactos.org > http://www.reactos.org/mailman/listinfo/ros-dev > >

2 1

Work on dxg.sys
by Sebastian Gąsiorek 10 Mar '17

10 Mar '17

Hi, Couple of days ago I have started dxg.sys functions implementation. There is some progress but it would be good to write tests for this api. Dxg mostly is just wrapper between driver and win32k, so win32k api tests should be enough. I saw some tests in rostests\dxtest\win32kdxtest but can't figure out how to enable them in build. Could some help me with this and add proper cmake files? :) Best regards, Sebastian

4 4

CLT, one seat left
by Daniel Reimer 09 Mar '17

09 Mar '17

After some strike related problems we lost one of our supporters @ Chemnitzer Linux Tage (https://chemnitzer.linux-tage.de/2017/de) SO if someone wants to join. Here we are, waiting for ya. Start is on THIS saturday and ending is THIS sunday evening. Yes, quite a bit early I ask here, but... sadly the problems showed up today. We might even have a hotel room left if our hotel does not cancel it that close to booking date. Otherwise we find a way, too. So, first come first served. Sell your soul/come to us ^^

1 0

Re: [ros-dev] [ros-diffs] [mjansen] 74063: [NTDLL] Fix RtlDecodeSystemPointer for usage inside ntdll. CORE-10368
by Michael Fritscher 06 Mar '17

06 Mar '17

Hi, I don't know - but this looks somehow ... surprising. Best regards, Michael > /* > * @implemented > + */ > +PVOID > +NTAPI > +RtlDecodeSystemPointer(IN PVOID Pointer) > +{ > + return RtlEncodeSystemPointer(Pointer); > +} > + > +/* > + * @implemented

2 2

Buildslave disk space problems fixed
by Colin Finck 05 Mar '17

05 Mar '17

Hi all! We have run out of space on our temporary universal Build and ISO machine HotelLux today. This is why TestCD creation has been failing. I'm currently moving files away from the main disk to a temporary storage. You should already be able to redo the failed builds. Within this month, Aleksey and me will move the ISO storage back to our actual storage server, upgraded with larger disks. This one should be sufficient for quite some time. Sorry for the inconveniences! Cheers, Colin

1 0

Re: [ros-dev] [ros-diffs] [hbelusca] 74068: [WS2_32] - Fix pointers verification in WSALookupServiceNextW and WSALookupServiceNextA; - In WSALookupServiceNextA: Perform extra checks to see whether we failed the first local ...
by Thomas Faber 05 Mar '17

05 Mar '17

On 2017-03-05 01:40, hbelusca(a)svn.reactos.org wrote: > + > + /* > + * Check whether we actually just retrieved the needed buffer size > + * because our previous local allocation did fail. If so, allocate > + * a new buffer and retry again. > + */ > + if ( (!UnicodeQuerySet) && (*lpdwBufferLength >= sizeof(WSAQUERYSETW)) && > + (ErrorCode == SOCKET_ERROR) && (GetLastError() == WSAEFAULT) ) Seems like you'd want to check WSAGetLastError when looking for WSA* error codes. > + { > + /* Allocate the buffer we'll use */ > + UnicodeQuerySet = HeapAlloc(WsSockHeap, 0, UnicodeQuerySetSize); > + if (UnicodeQuerySet) > + { > + /* Call the Unicode Function */ > + ErrorCode = WSALookupServiceNextW(hLookup, > + dwControlFlags, > + &UnicodeQuerySetSize, > + UnicodeQuerySet); > + } > + /* > + * Otherwise the allocation failed and we > + * fall back into the error checks below. > + */ > + } > + > if (ErrorCode == ERROR_SUCCESS) > { > - /* Not convert to ANSI */ > + /* Now convert back to ANSI */ > ErrorCode = MapUnicodeQuerySetToAnsi(UnicodeQuerySet, > lpdwBufferLength, > lpqsResults); > - if (ErrorCode != ERROR_SUCCESS) SetLastError(ErrorCode); > + if (ErrorCode != ERROR_SUCCESS) > + SetLastError(ErrorCode); > } > else > { > @@ -499,10 +542,11 @@ > } > > /* If we had a local buffer, free it */ > - if (UnicodeQuerySet) HeapFree(WsSockHeap, 0, UnicodeQuerySet); > + if (UnicodeQuerySet) > + HeapFree(WsSockHeap, 0, UnicodeQuerySet); HeapFree(NULL) is fine by the way. > > /* Return to caller */ > - return ErrorCode == ERROR_SUCCESS ? ErrorCode : SOCKET_ERROR; > + return (ErrorCode == ERROR_SUCCESS) ? ErrorCode : SOCKET_ERROR; > }

2 1

Re: [ros-dev] [ros-diffs] [ekohl] 74054: [NTOS:OB] Create a custom security descriptor for the global DosDevices directory.
by Thomas Faber 04 Mar '17

04 Mar '17

On 2017-03-04 17:02, ekohl(a)svn.reactos.org wrote: > + Dacl = ExAllocatePool(PagedPool, AclLength); Would you mind using ExAllocatePoolWithTag for new code please, with no exceptions? (I want to #define ExAllocatePool do_not_use_this_function at some point) Thanks! -Thomas

2 1

Re: [ros-dev] [ros-diffs] [hbelusca] 73895: [NTGDI]: Few fixes & hacks for NtGdiGetGlyphIndicesW, see CORE-12825: - (Fix) Check for integer count overflow (per Thomas suggestion), - (Hack#1) Signal the particular calling ca...
by Thomas Faber 03 Mar '17

03 Mar '17

On 2017-02-24 02:52, hbelusca(a)svn.reactos.org wrote: > @@ -4419,12 +4422,31 @@ > LPCWSTR UnSafepwc = pwc; > LPWORD UnSafepgi = pgi; > > - if ((!UnSafepwc) && (!UnSafepgi)) return cwc; > - > - if ((UnSafepwc == NULL) || (UnSafepgi == NULL)) > + /* Check for integer overflow */ > + if (cwc & 0x80000000) // (INT_MAX + 1) == INT_MIN > + return GDI_ERROR; > + > + if (!UnSafepwc && !UnSafepgi) > + return cwc; > + > + if (!UnSafepwc || !UnSafepgi) > { > DPRINT1("UnSafepwc == %p, UnSafepgi = %p\n", UnSafepwc, UnSafepgi); > - return -1; > + return GDI_ERROR; > + } > + > + // TODO: Special undocumented case! > + if (!pwc && !pgi && (cwc == 0)) Why the sudden switch away from using the Unsafe version? > + { > + DPRINT1("ERR: NtGdiGetGlyphIndicesW with (!pwc && !pgi && (cwc == 0)) is UNIMPLEMENTED!\n"); > + return 0; > + } > + > + // FIXME: This is a hack!! (triggered by e.g. Word 2010). See CORE-12825 > + if (cwc == 0) > + { > + DPRINT1("ERR: NtGdiGetGlyphIndicesW with (cwc == 0) is UNIMPLEMENTED!\n"); > + return GDI_ERROR; > } > > dc = DC_LockDc(hdc); > >

3 2

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev March 2017