May 2011 - Wine-devel - list.winehq.org

Re: tools: Upgrade the config.guess/config.sub scripts
by Austin English 08 May '11

08 May '11

2011/5/7 André Hentschel <nerv(a)dawncrow.de>: > from git://git.sv.gnu.org/config.git without the whitespace changes > --- > tools/config.guess | 17 ++++++++++------- > tools/config.sub | 49 +++++++++++++++++++++++++++++++++++-------------- > 2 files changed, 45 insertions(+), 21 deletions(-) > > diff --git a/tools/config.guess b/tools/config.guess > index 4c8f032..1f28e79 100755 > --- a/tools/config.guess > +++ b/tools/config.guess > @@ -92,7 +92,7 @@ if test $# != 0; then > exit 1 > fi > > -trap 'exit 1' HUP INT TERM > +trap 'exit 1' 1 2 15 > > # CC_FOR_BUILD -- compiler used by this script. Note that the use of a > # compiler to aid in system detection is discouraged as it requires > @@ -106,7 +106,7 @@ trap 'exit 1' HUP INT TERM > > set_cc_for_build=' > trap "exitcode=\$?; (rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null) && exit \$exitcode" 0 ; > -trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" HUP INT PIPE TERM ; > +trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" 1 2 13 15 ; > : ${TMPDIR=/tmp} ; > { tmp=`(umask 077 && mktemp -d "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n "$tmp" && test -d "$tmp" ; } || > { test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir $tmp) ; } || Looks like http://bugs.winehq.org/show_bug.cgi?id=16622 has re-emerged: http://git.savannah.gnu.org/gitweb/?p=config.git;a=commitdiff;h=2ec80f0dba7… -- -Austin

1 0

Re: riched20:tests Add conformance test for EM_FINDWORDBREAK function.
by André Hentschel 08 May '11

08 May '11

Am 08.05.2011 05:14, schrieb James McKenzie: > > + if (winetest_debug > 1) { > test_WM_CHAR(); > test_EM_FINDTEXT(); > test_EM_GETLINE(); > @@ -7090,6 +7368,8 @@ START_TEST( editor ) > test_WM_GETDLGCODE(); > test_zoom(); > test_dialogmode(); > + } > + test_EM_FINDWORDBREAK(); > Hi, i guess you didn't meant to include that in your patch. -- Best Regards, André Hentschel

2 1

Re: ntdll: Make RtlReAllocateHeap() preserve V-bit information when run under Valgrind
by Marcus Meissner 08 May '11

08 May '11

On Sat, May 07, 2011 at 06:33:08PM +0200, Bart Van Assche wrote: > Hello, > > A patch I submitted about five weeks ago had the status "Pending" on the patch status page during about four weeks. I can't see this patch anymore on the patch status page today. Anyone who can tell me what happened or which steps I should take next ? > > The patch itself can be found here: http://www.winehq.org/pipermail/wine-patches/2011-April/100570.html If you got no reply or indication on a patch within a week, you can consider it lost ;) To do then: - Ask yourself if the patch is good (tm) - Ask by e-mail or in IRC (#winehackers) on what could be improved with it. >From a brief look ... I think your patch seems to be doing more than one thing at once. Please seperate multiple changes into multiple small patches. Ciao, Marcus

1 0

RFC: doc for message contexts additions in PO files
by Frédéric Delanoy 08 May '11

08 May '11

Hi guys, I've put together the following doc for the introduction of message contexts. It's still a bit raw but I'd improve it for the wiki version. Could you please have a look and help proofread it? I'd like to ensure I didn't miss an important step. Thanks in advance. Frédéric Delanoy PS: BTW what is the reference version? en_US.po or en.po? ********* Adding one/more message contexts (ADD LINK TO EXTERNAL DOC) (QUICK MC EXPLANATION) Prerequisite: enable maintainer mode using ./configure --enable-maintainer-mode (you might have to add some development packages; check with your distribution for infos). 1. Locate strings to be "altered" Say you have in your localized po file: #: file1.rc:123 file2.rc:456 msgid "ambiguous word" msgstr "translation" and you want to disambiguate, you'd do the following 2. Find appropriate translations - check the respective rc entries (in file1.rc line 123) FOO "ambiguous word" (in file2.rc line 456) BAR "ambiguous word" and find their meaning/context (e.g. using some code browsing tools, like cscope to find references; running the program is recommanded, ...) 3. Add the messages contexts - The following syntax is used: "translation" -> "#msgctxt#your context#translation" - You end up with sthg like FOO "#msgctxt#context #ambiguous word#" Similarly BAR "#msgctxt#context 2#ambiguous word#" 3. Run 'make depend && make' 4. Edit your po file to add the new translations 5. Ideally, remove the "#, fuzzy" automatically inserted in the En_US.po reference file (and En.rc ). 6. Run 'make' again and test your changes.

2 1

Re: gdi32: Remove a set but unused variable
by Gerald Pfeifer 08 May '11

08 May '11

I admit I am surprised that Commit: 46988651d91fbea7f611287e5d8320239e72108d URL: http://source.winehq.org/git/wine.git/?a=commit;h=46988651d91fbea7f611287e5… Author: Nicolas Le Cam <niko.lecam at gmail.com> Date: Fri Apr 29 23:57:09 2011 +0200 gdi32: Remove a set but unused variable. has been applied, for I had proposed this very patch a year ago when Alexandre had rejected it with It seems pretty clear that it should be used, not removed. [ http://www.winehq.org/pipermail/wine-devel/2010-May/083517.html ] I tried to take that into account by updating the patch accordingly http://www.winehq.org/pipermail/wine-patches/2010-May/088317.html which, unexplicably to me, broke the tests so it was not applied: http://www.winehq.org/pipermail/wine-devel/2010-May/083850.html Now my original patch was re-submitted by Nicolas and applied. Was Alexandre's analysis incorrect in the original case, did he just fail to catch the mistake this time around, or is it something else? Gerald

1 0

Re: Jerome Leclanche : wine.desktop: Remove the nonexistent application/ x-win-lnk MIME type.
by Francois Gouget 07 May '11

07 May '11

On Thu, 5 May 2011, Alexandre Julliard wrote: > Module: wine > Branch: master > Commit: db03d247158ee44c06935b5e23fbf37485d8c8bf > URL: http://source.winehq.org/git/wine.git/?a=commit;h=db03d247158ee44c06935b5e2… > > Author: Jerome Leclanche <adys.wh(a)gmail.com> > Date: Wed May 4 19:01:33 2011 +0100 > > wine.desktop: Remove the nonexistent application/x-win-lnk MIME type. I'm not sure this patch is correct: on my Debian Testing machine application/x-win-lnk is defined in /usr/share/mime/packages/kde.xml which comes from the kdelibs5-data. I also found a reference to this package in the kdebase-runtime Fedora 11 runtime. So this MIME type may not be standard (heck it has the 'x-' prefix anyway) but KDE seems to know / be using it. -- Francois Gouget <fgouget(a)free.fr> http://fgouget.free.fr/ Research is the transformation of money to knowledge. Innovation is the transformation of knowledge to money. -- Dr. Hans Meixner

4 4

Huge frustration
by Roland Kaeser 07 May '11

07 May '11

Hello Please take the following not personal but the huge pent-up frustration constrains me to say a few words. I've just tested one of my "longest waiting" apps to get working on linux: CorelDraw X3/X4. And I had to see that it fails already with the same errors. I tooke a short look into the bug database to see all the related bugs: http://bugs.winehq.org/show_bug.cgi?id=11687 open (not even confirmed) since more than 3 years http://bugs.winehq.org/show_bug.cgi?id=14123 open since 2008-06-25 http://bugs.winehq.org/show_bug.cgi?id=3599 open (new) since 2005-10-15 http://bugs.winehq.org/show_bug.cgi?id=3611 open (new) since 2005-10-17 I know we had this discussion already a lot of times but after six acc. three years its time for me to say something again. I know also thats a lot more fun developing directx routines instead of fixing old bugs but sometimes You should also think about the "other" generality which uses the computer not just to play games. It would be understandeable to me if its written in the bug that the fix of this bug will consume a lot of resources and its impossible to fix this, but just to ignore it is not "the proper way to behave". I'm sure that there are many more of the above bugs which nobody has has ever took seriously a lock at it. Has anybody ever imagined how many apps depending on such bugs which prevents them to get running on Linux? Maybe all of the above bugs are only related to CorelDraw but its also possible that dozens or hundrets of other apps also fail with the same problems but nobody has ever wrote a bug report or saw it in the bug database and thought that it will be fixed cause its already reported? Just take a look at: http://www.inside-it.ch/frontend/insideit?_d=_article&site=ic&news.id=22515 English version: http://arstechnica.com/civis/viewtopic.php?f=14&t=1122655 The most of their problems where related to special windows apps running only in goverment institutions and where not able to run on Linux (wine). How many of these problems whould not even ever have been occured if such small bugs like the above whoud be fixed which makes all of those apps runnable on linux. Sorry to say that but Your politics just to develop things which makes fun or are interesting is also part of the above problem. Regards Roland

5 4

Re: [1/2] mmsystem.dll16: fix MCI_STATUS mapping for digitalvideo
by Damjan Jovanovic 06 May '11

06 May '11

On Thu, May 5, 2011 at 9:23 PM, Eric Pouech <eric.pouech(a)orange.fr> wrote: > Le 05/05/2011 15:30, Damjan Jovanovic a écrit : > > Changelog: > * mmsystem.dll16: fix MCI_STATUS mapping for digitalvideo > > Damjan Jovanovic > > > Hi Damjan, > > I wonder if storing the device type (or at least information to select the > two types of mapping) in the allocated structure for mapping instead of > calling for the device type at each unmap operation wouldn't be a preferable > solution > A+ > > -- > Eric Pouech Hi Eric It might be even more preferable to store the device type in a local variable. The allocated structure is way too complex already. Damjan Jovanovic

1 0

Re: [PATCH try4] wininet: Added support for persistent HTTP connections
by Alexandre Julliard 06 May '11

06 May '11

Jacek Caban <jacek(a)codeweavers.com> writes: > + /* FIXME: do the check also for SSL connections */ > + if(!netconn->useSSL) { > + res = NETCON_recv(netconn, &b, 1, MSG_PEEK|MSG_DONTWAIT, &len); > + if((res == ERROR_SUCCESS && len == 1) || res == WSAEWOULDBLOCK) > + break; This doesn't build with winsock (MSG_DONTWAIT is not defined). I'm also not sure what is the purpose of that check. If it's to check if remote has closed the connection, you need to do that by handling the error on the first send. -- Alexandre Julliard julliard(a)winehq.org

2 1

Re: [PATCH try4] wininet: Added support for persistent HTTP connections
by Alexandre Julliard 06 May '11

06 May '11

Jacek Caban <jacek(a)codeweavers.com> writes: > try4: Check idle connection before using it in an attempt to catch > closed connection sooner, avoiding some request resends. It hopefully > fixes observed problem with tests. It does, but there's a new problem: it sometimes crashes in the collect_connections thread during the winhttp tests, because wininet has been unloaded while the thread is still running. Probably the thread needs to grab a library reference, though a collection mechanism that doesn't require a separate thread would be even better. I can still put the patch in for now, since the failure is not very frequent. -- Alexandre Julliard julliard(a)winehq.org

1 0