http://blog.gmane.org/gmane.comp.lib.gnustep.bugs hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14285 <pre>Update of bug #38955 (project gnustep): Status: None =&gt; Fixed Open/Closed: Open =&gt; In Test _______________________________________________________ Follow-up Comment #1: Thanks ... I think I fixed thjs in svn trunk ... please let me know. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38955&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Richard Frith-Macdonald 2013-05-24T14:35:24 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14284 <pre>Update of bug #38981 (project gnustep): Status: None =&gt; Fixed Open/Closed: Open =&gt; In Test _______________________________________________________ Follow-up Comment #1: Thanks, this should be fixed in trunk now ... needs some testing though. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38981&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Richard Frith-Macdonald 2013-05-24T14:34:44 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14283 <pre> On May 20, 2013, at 4:36 PM, Jiří Boldyš wrote: Hello Jiri, GNUstep does not work on Cygwin. If you'd like to use GNUstep on Windows, we recommend the MinGW installer: http://www.gnustep.org/experience/Windows.html _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Adam Fedor 2013-05-21T14:21:31 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14282 <pre>Hi, I am trying to install GNUstep and I am getting errors - logs are attached. Thanks for help. Jiri _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Jiří Boldyš 2013-05-20T22:36:01 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14281 <pre>2013/5/16, ganesh jaiswal &lt;ganesh.jn&lt; at &gt;gmail.com&gt;: gnustep-startup contains old gnustep packages. In the new packages, by default, gnustep isn't installed in /usr/GNUstep. I don't remember exactly where is this installed. But you can check the file /etc/GNUstep/GNUstep.conf where are listed all the paths used by GNUstep. Regards. </pre> German Arias 2013-05-17T06:05:03 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14280 <pre>URL: &lt;http://savannah.gnu.org/bugs/?38981&gt; Summary: NSMessagePortNameServer initialize method doesn't remove obsolete name references Project: GNUstep Submitted by: msilvax28 Submitted on: Mon 13 May 2013 09:08:51 PM GMT Category: Base/Foundation Severity: 3 - Normal Item Group: None Status: None Privacy: Public Assigned to: None Open/Closed: Open Discussion Lock: Any _______________________________________________________ Details: The logic in the +initialize method has logic that tries to remove old names. It tries to remove names assigned to the current pid and tries to see if other pids that have names are now dead. The logic as implemented removes no names, as it assumes the beginning of the first line of the name file is the PID (with a . suffix). However, the name files have this format: &lt;path to 'port' in file system&gt;\n &lt;owner pid&gt;\n for example: /tmp/GNUstepSecure1000/NSMessagePort/ports/10280.4 10280 I have code that I think fixes it, but I don't know enough about GNUstep to be sure. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38981&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Michael Silva 2013-05-13T21:08:52 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14279 <pre>URL: &lt;http://savannah.gnu.org/bugs/?38980&gt; Summary: NSConnection can cause thread deadlock Project: GNUstep Submitted by: msilvax28 Submitted on: Mon 13 May 2013 04:59:35 PM GMT Category: Base/Foundation Severity: 3 - Normal Item Group: Bug Status: None Privacy: Public Assigned to: None Open/Closed: Open Discussion Lock: Any _______________________________________________________ Details: There are cases in NSConnection when the use of NSRecursiveLocks between objects can lead to 2 threads deadlocking. The case I have seen is in gnustep-base v1.24.4. The NSConnection method locateLocalTarget: locks the following objects in this order IrefGate for self connection_table_gate IRefGate for another NSConnection In addition, the invalidate method locks objects in this order IrefGate connection_table_gate I have seen 1 lockup scenario, and fear for another occuring. Scenario: NSConnection 'A' has its locateLocalTarget: methods called. It locks the local IrefGate and proceeds into the method. Later in the method it locks connection_table_gate and begins to scan the members of the connection_table. Another NSConnection 'B' has invalidate called on another thread. It locks the local IrefGate and proceeds into the method. It later attempts to lock connection_table_gate but cannot as the thread running NSConnection 'A' has it. The thread blocks. Back on the thread for NSConnection 'A', while enumerating connection_table it attempts to lock NSConnection 'B' IrefGate. It cannot, because the thread running NSConnection 'B' has it. So it locks. The threads running NSConnections A and B are now deadlocked. I am also concerned that the same deadlock can happen if both Connection A and B enter locateLocalTarget at the same time where one connection gets its local IrefGate and connection_table_gate and the other gets only its local IRefGate. I don't have enough context on the software to recommend a solution, although I have 'tinkered' with something that seems to help (still testing). For information, in my test application it takes about 8-12 hours of continuous run before this condition occurs. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38980&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Michael Silva 2013-05-13T16:59:37 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14278 <pre>URL: &lt;http://savannah.gnu.org/bugs/?38955&gt; Summary: NSURLConnection does not release its delegate when it should Project: GNUstep Submitted by: cpulsford Submitted on: Thu 09 May 2013 10:15:23 PM GMT Category: Base/Foundation Severity: 3 - Normal Item Group: Bug Status: None Privacy: Public Assigned to: None Open/Closed: Open Discussion Lock: Any _______________________________________________________ Details: I am running on an Intel NUC running debian wheezy, with hand built gnustep 1.24. My GCC is version 4.7.2. This is sort of a corollary to this bug: https://savannah.gnu.org/bugs/?func=detailitem&amp;item_id=35686 I have written a wrapper around NSURLConnection called RPMHTTPConnection. In the attached test program, I alloc/init an RPMHTTPConnection, send a request, receive a callback, and then release the RPMHTTPConnection instance. However, -dealloc is never called in my RPMHTTPConnection class and I suspect that the underlying NSURLConnection is not releasing its delegate after the request finishes. If I am more eager about releasing the NSURLConnection, like releasing it in the method where RPMHTTPConnection calls its own delegate, the dealloc is called as I would expect. Today, May 9th 2013, I checked out the latest gnustep and I do not see _delegate being released after the connection finished or fails. When I run the attached program I get this output: NSURLConnectionBug[16514] got response from instance: 0x9b19e74, data length: 44074 When I run the equivalent program on my Mac I get this output: NSURLConnectionBug[9650] got response from instance: 0x9d8e22c, data length: 44074 NSURLConnectionBug[9650] deallocing instance: 0x9d8e22c _______________________________________________________ File Attachments: ------------------------------------------------------- Date: Thu 09 May 2013 10:15:23 PM GMT Name: NSURLConnectionBug.zip Size: 47kB By: cpulsford &lt;http://savannah.gnu.org/bugs/download.php?file_id=28056&gt; ------------------------------------------------------- Date: Thu 09 May 2013 10:15:23 PM GMT Name: RPMHTTPConnection.m Size: 7kB By: cpulsford &lt;http://savannah.gnu.org/bugs/download.php?file_id=28057&gt; _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38955&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Cameron Pulsford 2013-05-09T22:15:25 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14277 <pre> On 7 May 2013, at 21:15, Cameron Pulsford &lt;cpulsfor&lt; at &gt;me.com&gt; wrote: There should be no need ... the string is automatically converted to the file system encoding (the same thing as the default C-string encoding). The problem here seems to be that your system is using the default encoding (latin1) and presumably your system is actually using a utf-8 based encoding. Now, the default encoding should be automatically inferred from your operating system's nl_langinfo() function ... but may be set/overridden with the GNUSTEP_STRING_ENCODING environment variable. So you could try setting the environment variable to the encoding you want to use. Better still, you could run your code under gdb and set a breakpoint in the GSPrivateDefaultCStringEncoding() function (base/Source/Additions/Unicode.m) and step through and see if you can figure out why the automatic selection of encoding is not working (so we can improve that). _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Richard Frith-Macdonald 2013-05-08T06:37:01 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14276 <pre> On 7 May 2013, at 22:45, Kal &lt;b17c0de&lt; at &gt;gmail.com&gt; wrote: We would need to know what the trivial fix is ... presumably there's some problem including the system headers, since the compiler seems to think that PRIuPTR is an identifier, which it is not (it's a standard preprocessor constant which should have been replaced by a literal string before the compiler sees it). Is that a system specific issue? Do we need to define some other constant before including headers etc? </pre> Richard Frith-Macdonald 2013-05-08T06:42:15 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14275 <pre>URL: &lt;http://savannah.gnu.org/bugs/?38933&gt; Summary: [Thematic] copyright info support Project: GNUstep Submitted by: rmottola Submitted on: Wed 08 May 2013 07:54:49 AM GMT Category: Application Severity: 2 - Minor Item Group: Change Request Status: None Privacy: Public Assigned to: None Open/Closed: Open Discussion Lock: Any _______________________________________________________ Details: Among the theme information (version, Details...) Thematic should add a line for copyright/license information, like a regular project. This is useful for redistribution. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38933&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Riccardo mottola 2013-05-08T07:54:50 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14274 <pre>I think the problem here is that GNUstep has the impression that your file system is using Latin1, where it seems to be using UTF8. Fred On the road Am 07.05.2013 um 22:15 schrieb Cameron Pulsford &lt;cpulsfor&lt; at &gt;me.com&gt;: _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Fred Kiefer 2013-05-08T07:22:03 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14273 <pre>Hi, GNUstep.h doesn't compile with clang in C++11 mode. I get the following errors: /opt/GNUstep/Local/Library/Headers/GNUstepBase/GNUstep.h:347:19: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] PRIuPTR", %"PRIuPTR" } extends beyond size (%"PRIuPTR")", \ ^ /opt/GNUstep/Local/Library/Headers/GNUstepBase/GNUstep.h:347:53: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] PRIuPTR", %"PRIuPTR" } extends beyond size (%"PRIuPTR")", \ ^ /opt/GNUstep/Local/Library/Headers/GNUstepBase/GNUstep.h:356:30: error: invalid suffix on literal; C++11 requires a space between literal and identifier [-Wreserved-user-defined-literal] format: &lt; at &gt;"in %s, index %"PRIuPTR" is out of range", \ Can someone commit the trivial fix for this? Thanks! Kal </pre> Kal 2013-05-07T21:45:49 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14272 <pre>I am using debian wheezy with hand built gnustep 1.24 and the gcc 4.7.2 objc runtime. I am getting an exception from calling [[NSFileManager defaultManager] fileExistsAtPath:] with the following two arguments and presumably others: &lt; at &gt;"/home/user/GNUstep/Library/ApplicationSupport/Foo/mediaServer/Jeff Hamilton Trio-NS/J.Hamｉlton/Beatles.png" &lt; at &gt;"/home/user/GNUstep/Library/ApplicationSupport/Foo/mediaServer/Brad White &amp; Pierre Grill-Winter’s Journey.png" Here are those strings again converted with CFStringTransform to include the unicode special character names: &lt; at &gt;"/home/user/GNUstep/Library/ApplicationSupport/Foo/mediaServer/Jeff Hamilton Trio-NS/J.Ham\N{FULLWIDTH LATIN SMALL LETTER I}lton/Beatles.png" &lt; at &gt;"/home/user/GNUstep/Library/ApplicationSupport/Foo/mediaServer/Brad White &amp; Pierre Grill-Winter\N{RIGHT SINGLE QUOTATION MARK}s Journey.png" Internally, that method eventually calls [GSUnicodeString cStringUsingEncoding:NSISOLatin1StringEncoding] which raises this exception: &lt;NSException: 0x83a29bc&gt; NAME:NSCharacterConversionException REASON:Can't get cString from Unicode string. INFO:(null) I do not get this crash when running the same code on a mac. Is there any way I should be sanitizing strings before I pass them to the -fileExistsAtPath: method? _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Cameron Pulsford 2013-05-07T20:15:06 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14271 <pre>I don't have a FreeBSD, so I don't know how supposedly should work it. But try with the official released packages. I think is better if you write to people who made the packages for FreeBSD. Also, I don't know if those packages are updated. Or try with the official packages of gnustep. Installing these one by one. Regards. On 2013-05-01 20:32:48 -0600 Bartlomiej Mika &lt;bmika&lt; at &gt;icloud.com&gt; wrote: _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Germán Arias 2013-05-03T17:38:18 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14270 <pre>Hi! So when I run cd /usr/ports/devel/gnustep &amp;&amp; make install clean I get the following error: Code: configure: exit 1 (end of "config.log") *** [do-configure] Error code 1 Stop in /usr/ports/lang/gnustep-base. *** [build-depends] Error code 1 Stop in /usr/ports/devel/gnustep. Ok, so I got into /usr/ports/lang/gnustep-base and run the whole make and install and I then get the following error: Code: configure: exit 1 (end of "config.log") *** [do-configure] Error code 1 Stop in /usr/ports/lang/gnustep-base. I'm at a loss on what to do next. Any hints? I think this is a bug…. Thanks! _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Bartlomiej Mika 2013-05-02T02:32:48 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14269 <pre>Follow-up Comment #2, bug #38859 (project gnustep): Hi Matt, thanks for the advice. Removing the EOAccess_LIBRARIES_DEPEND_UPON line is sufficient, so the ADDITIONAL_NATIVE_LIBS settings in the preamble are working. However, I don't know whether this would affect other platforms, so no patch yet: The Target_LIBRARIES_DEPEND_UPON were added in r35397, which just says "Add flags": 2012-08-10 German A. Arias &lt;german&lt; at &gt;xelalug.org&gt; * EOInterface/GNUmakefile: * EOAccess/GNUmakefile: * EOAdaptors/SQLiteAdaptor/GNUmakefile.in: * EOAdaptors/PostgreSQLAdaptor/config.mak.in: * EOControl/GNUmakefile: * EOModeler/GNUmakefile: Add flags to compile with new linkers. I guess we need to ask German what "new linkers" there are, and how removing Target_LIBRARIES_DEPEND_UPON might break them? Thanks, Graham. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38859&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Graham Lee 2013-05-01T09:20:05 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14268 <pre>Follow-up Comment #1, bug #38859 (project gnustep): the following entries in Makefile.preamble used to link this with the appropriate linker command line for the platform, figuring out whats gone awry there would be the right place to start ADDITIONAL_NATIVE_LIBS += EOControl ADDITIONAL_NATIVE_LIB_DIRS+=../EOControl (apologies if this gets sent twice) _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38859&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> matt rice 2013-05-01T02:17:34 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14267 <pre>URL: &lt;http://savannah.gnu.org/bugs/?38859&gt; Summary: GDL2 build failures on Apple platforms Project: GNUstep Submitted by: leeg Submitted on: Tue 30 Apr 2013 02:32:54 PM GMT Category: gdl2 Severity: 3 - Normal Item Group: Bug Status: None Privacy: Public Assigned to: None Open/Closed: Open Discussion Lock: Any _______________________________________________________ Details: Hi, I'm building GDL2 from r36452 on apple-apple-apple. The build is fine, but linking fails because it tries to use Linux-style -lFramework instead of Apple-style -framework Framework flags. I can patch the GNUmakefiles to solve my problem (example below), but a patch should really support both linker styles and I don't know how I'd do that. Thanks, Graham. -EOAccess_LIBRARIES_DEPEND_UPON = -lEOControl $(OBJC_LIBS) +EOAccess_LIBRARIES_DEPEND_UPON = -framework EOControl $(OBJC_LIBS) _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?38859&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ </pre> Graham Lee 2013-04-30T14:32:55 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14266 <pre>Follow-up Comment #1, bug #35671 (project gnustep): I don't know how precise compatibility with Mac OS X/Cocoa should be for this bug, but the behaviour proposed in the patch doesn't match it: 'Wrong error Error Domain=NSCocoaErrorDomain Code=516 "The file “foobar” couldn’t be saved in the folder “tmp” because a file with the same name already exists." UserInfo=0x7f85d940ab80 {NSFilePath=/tmp/foobar, NSUnderlyingError=0x7f85d940a9d0 "The operation couldn’t be completed. File exists"}' but error code 516 is not EEXIST and the error is in the Cocoa error domain. The _underlying_ error is in NSPOSIXErrorDomain and has code 17 (EEXIST). The matching Foundation error (from Foundation/FoundationErrors.h) is this: NSFileWriteFileExistsError NS_ENUM_AVAILABLE(10_7, 5_0) = 516, // Write error (file exists) which is what NSFileManager returns in Foundation. I think it'd be reasonable for GNUstep-base to follow that example. _______________________________________________________ Reply to this item at: &lt;http://savannah.gnu.org/bugs/?35671&gt; _______________________________________________ Message sent via/by Savannah http://savannah.gnu.org/ _______________________________________________ Bug-gnustep mailing list Bug-gnustep&lt; at &gt;gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnustep </pre> Graham Lee 2013-04-29T15:51:36 http://permalink.gmane.org/gmane.comp.lib.gnustep.bugs/14265 <pre>On 2013-04-29 00:02:22 -0600 ganesh jaiswal &lt;ganesh.jn&lt; at &gt;gmail.com&gt; wrote: From where downloaded you these packages? Download these separately from here: http://ftpmain.gnustep.org/pub/gnustep/core/gnustep-make-2.6.4.tar.gz http://ftpmain.gnustep.org/pub/gnustep/core/gnustep-base-1.24.4.tar.gz http://ftpmain.gnustep.org/pub/gnustep/core/gnustep-gui-0.23.1.tar.gz http://ftpmain.gnustep.org/pub/gnustep/core/gnustep-back-0.23.0.tar.gz And install them one by one. Hope this help. </pre> Germán Arias 2013-04-29T06:59:34 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.lib.gnustep.bugs