Geany GTK 2.8 build test failed at: waf build
See http://nightly.geany.org/misc/build_gtk28_stderr.log for details.
Last error message: ../../gtk28_test/src/highlighting.c: In function 'styleset_common_init': ../../gtk28_test/src/highlighting.c:532: warning: implicit declaration of function 'g_hash_table_remove_all' default/src/highlighting_3.o: In function `styleset_common_init': highlighting.c:(.text+0xdb2): undefined reference to `g_hash_table_remove_all' collect2: ld returned 1 exit status Build failed -> task failed (err #1): {task: cxx_link about_3.o,build_3.o,callbacks_3.o,dialogs_3.o,document_3.o,editor_3.o,encodings_3.o,filetypes_3.o,geanyentryaction_3.o,geanymenubuttonaction_3.o,geanyobject_3.o,geanywraplabel_3.o,highlighting_3.o,interface_3.o,keybindings_3.o,keyfile_3.o,log_3.o,main_3.o,msgwindow_3.o,navqueue_3.o,notebook_3.o,plugins_3.o,prefix_3.o,prefs_3.o,printing_3.o,project_3.o,queue_3.o,sciwrappers_3.o,search_3.o,socket_3.o,stash_3.o,symbols_3.o,templates_3.o,toolbar_3.o,tools_3.o,treeviews_3.o,ui_utils_3.o,utils_3.o,vte_3.o -> geany}
Hi all.
This is definitely introduced in r3955 when my patch for reloading color schemes was applied. Here is a patch replacing problematic g_hash_table_remove_all() with g_hash_table_destroy(). The former is available since 2.12 only.
When I wrote the code, I did not take required GTK version into account. Sorry for inconvenience.
The patch is against r3969. Alternative link: http://pastie.org/pastes/546469 .
Best regards, Eugene.
Geany GTK 2.8 build test failed at: waf build
See http://nightly.geany.org/misc/build_gtk28_stderr.log for details.
Last error message: ../../gtk28_test/src/highlighting.c: In function 'styleset_common_init': ../../gtk28_test/src/highlighting.c:532: warning: implicit declaration of function 'g_hash_table_remove_all' default/src/highlighting_3.o: In function `styleset_common_init': highlighting.c:(.text+0xdb2): undefined reference to `g_hash_table_remove_all' collect2: ld returned 1 exit status Build failed -> task failed (err #1): {task: cxx_link about_3.o,build_3.o,callbacks_3.o,dialogs_3.o,document_3.o,editor_3.o,encodings_3.o,filetypes_3.o,geanyentryaction_3.o,geanymenubuttonaction_3.o,geanyobject_3.o,geanywraplabel_3.o,highlighting_3.o,interface_3.o,keybindings_3.o,keyfile_3.o,log_3.o,main_3.o,msgwindow_3.o,navqueue_3.o,notebook_3.o,plugins_3.o,prefix_3.o,prefs_3.o,printing_3.o,project_3.o,queue_3.o,sciwrappers_3.o,search_3.o,socket_3.o,stash_3.o,symbols_3.o,templates_3.o,toolbar_3.o,tools_3.o,treeviews_3.o,ui_utils_3.o,utils_3.o,vte_3.o -> geany}
http://nightly.geany.org/ _______________________________________________ Geany-devel mailing list Geany-devel@uvena.de http://lists.uvena.de/cgi-bin/mailman/listinfo/geany-devel
On Wed, 15 Jul 2009 10:21:13 +0400 Eugene Arshinov earshinov@gmail.com wrote:
This is definitely introduced in r3955 when my patch for reloading color schemes was applied. Here is a patch replacing problematic g_hash_table_remove_all() with g_hash_table_destroy(). The former is available since 2.12 only.
When I wrote the code, I did not take required GTK version into account. Sorry for inconvenience.
No problem.
The patch is against r3969.
Thanks! Applied.
Geany GTK 2.8 build test failed at: waf build
See http://nightly.geany.org/misc/build_gtk28_stderr.log for details.
Last error message: ../../gtk28_test/src/highlighting.c: In function 'styleset_common_init': ../../gtk28_test/src/highlighting.c:532: warning: implicit declaration of function 'g_hash_table_remove_all' default/src/highlighting_3.o: In function `styleset_common_init': highlighting.c:(.text+0xdb2): undefined reference to `g_hash_table_remove_all'
This goes to show how useful the nightly build is, thanks Enrico ;-)
Regards, Nick
On Wed, 15 Jul 2009 10:21:13 +0400, Eugene wrote:
Hi all.
This is definitely introduced in r3955 when my patch for reloading color schemes was applied. Here is a patch replacing problematic g_hash_table_remove_all() with g_hash_table_destroy(). The former is available since 2.12 only.
When I wrote the code, I did not take required GTK version into account. Sorry for inconvenience.
As Nick said, don't worry. That's why we set up the build tests (I also like to do such mistakes :).
Regards, Enrico