|
| From: | Éloi Rivard |
| Subject: | Re: [Denemo-devel] Testsuite, step 1 |
| Date: | Thu, 14 Nov 2013 12:56:19 +0100 |
Hmm, I had to throw the whole git clone away and start with a fresh git
clone. (Something bad must have happened, because git pull said I had
merge conflicts - but this is a clean user and I haven't done anything
as that user except change to the branch and do the build
stuff: ./autogen.sh, ./configure --prefix and make, make install.)
Having done this, Command Center no longer crashes - the search routine
bug is still present when tooltip search is unchecked.
I have played about a bit with this version and seen nothing bad so far.
At the moment I cannot get it to playback, I think because of
permissions for that user. I'll look further at that.
Richard
On Thu, 2013-11-14 at 10:31 +0100, Éloi Rivard wrote:
> According to your segfault, it seems to fail at line 2019 that is:
>
> if(!idx)
>
> How can this line produce a segfault ? I simply checks if a pointer is
> NULL, are you sure you are compiling the very last version ?
>
>
>
> 2013/11/14 Éloi Rivard <address@hidden>
> Actually here idx is a pointer, if *idx is -1 then the
> following line will return NULL too. It is strange because I
> do not have this segfault anymore since my fix.
>
>
>
> 2013/11/14 Richard Shann <address@hidden>
> On Thu, 2013-11-14 at 09:50 +0100, Éloi Rivard wrote:
> > The segfault is fixed now fixed.
>
>
> I did a fresh git clone as a separate user and then
> checked out and
> built the branch. Trying to open the Command Center I
> get the same
> segfault:
>
> ** (denemo:28410): WARNING **: Error executing
> lilypond. Perhaps Lilypond is not installed or its
> path is not correctly configured.
> [New Thread 0x7fffe3087700 (LWP 28420)]
>
> Program received signal SIGSEGV, Segmentation fault.
>
> get_menu_label (name=0x36fc360 "Built-in, see file
> denemoui.xml for position")
> at kbd-custom.c:2019
> 2019 command_row* row =
> g_hash_table_lookup(Denemo.map->commands, *idx);
>
> (gdb) where
> #0 get_menu_label (
> name=0x36fc360 "Built-in, see file denemoui.xml
> for position")
> at kbd-custom.c:2019
> #1 0x000000000043f304 in get_menu_position
> (menupath=<optimized out>)
>
> at kbd-custom.c:2032
> #2 0x000000000043f5a0 in
> keymap_change_binding_view_on_command_selection (
> selection=<optimized out>, model=<optimized out>,
> path=<optimized out>,
> path_currently_selected=<optimized out>,
> data=""> >
> at kbd-custom.c:2117
>
> Looking at your last commit on that branch I see
>
> + if(!idx)
> + return name;
>
> idx have -1 as the not found value, is this your
> problem? Did you test
> it - does it work for you?
>
> Richard
>
> > It might be related to the search bug. I just tested
> and with the
> > checkbox unchecked, it seems to search in the names
> only. Can you
> > re-check ?
> >
> >
> > 2013/11/13 Richard Shann <address@hidden>
> > This is probably not related, but I notice
> that searching in
> > the Command
> > Center is only working now for searching
> tooltips. The other
> > search
> > (with the checkbox unchecked) was intended
> to be a search for
> > the
> > command name (originally), but I can't
> understand what it is
> > actually
> > doing at present.
> >
> > Richard
> >
> > On Wed, 2013-11-13 at 17:40 +0000, Richard
> Shann wrote:
> > > Eloi, I checked out the command-hash-table
> branch
> > ran ./autogen.sh and re-built.
> > > Running the program I open View->Command
> Center and
> > immediately get:
> > >
> > > Program received signal SIGSEGV,
> Segmentation fault.
> > > get_menu_label (name=0x3a79c10 "Built-in,
> see file
> > denemoui.xml for position")
> > > at ../../denemo/src/kbd-custom.c:2019
> > > 2019 command_row* row =
> > g_hash_table_lookup(Denemo.map->commands,
> *idx);
> > >
> > > where gives
> > >
> > > #0 get_menu_label (
> > > name=0x3a79c10 "Built-in, see file
> denemoui.xml for
> > position")
> > > at ../../denemo/src/kbd-custom.c:2019
> > > #1 0x000000000043f304 in
> get_menu_position
> > (menupath=<optimized out>)
> > > at ../../denemo/src/kbd-custom.c:2032
> > > #2 0x000000000043f5a0 in
> >
> keymap_change_binding_view_on_command_selection (
> > > selection=<optimized out>,
> model=<optimized out>,
> > path=<optimized out>,
> > > path_currently_selected=<optimized
> out>, data=""> > > > at ../../denemo/src/kbd-custom.c:2117
> > >
> > > Happens every time.
> > >
> > > Richard
> > >
> > >
> > >
> > > On Wed, 2013-11-13 at 17:02 +0100, Éloi
> Rivard wrote:
> > > > Hi,
> > > >
> > > > I have commited on the
> command-hash-table branch. On this
> > branch
> > > > denemo can be launched with no graphical
> interface (for
> > example on a
> > > > server). This is very useful for Travis
> for example. So
> > now Travis
> > > > launches denemo at each commit with the
> command : denemo
> > -n -a
> > > > "(d-Quit)". I would like to test some
> more complex
> > command, but for
> > > > now a lot of commands are dependent from
> the GUI.
> > > >
> > > >
> > > > I had to move a lot of things to make
> denemo less
> > dependent towards
> > > > gtk.
> > > >
> > > > It would be great if you can test this
> branch, so I can
> > merge it and
> > > > tackle the second step.
> > > >
> > > >
> > > > When it is merged, I will try to set up
> a test-suite using
> > Glib, and
> > > > progressively add some tests. The first
> one should be to
> > load and save
> > > > a file for example.
> > > >
> > > >
> > > > --
> > > > Éloi Rivard - address@hidden
> > > >
> > > > « On perd plus à être indécis qu'à se
> tromper. »
> > > >
> > > >
> _______________________________________________
> > > > Denemo-devel mailing list
> > > > address@hidden
> > > >
> https://lists.gnu.org/mailman/listinfo/denemo-devel
> > >
> > >
> > >
> > >
> _______________________________________________
> > > Denemo-devel mailing list
> > > address@hidden
> > >
> https://lists.gnu.org/mailman/listinfo/denemo-devel
> >
> >
> >
> >
> >
> >
> > --
> > Éloi Rivard - address@hidden
> >
> > « On perd plus à être indécis qu'à se tromper. »
> >
>
>
>
>
>
>
> --
> Éloi Rivard - address@hidden
>
> « On perd plus à être indécis qu'à se tromper. »
>
>
>
>
> --
> Éloi Rivard - address@hidden
>
> « On perd plus à être indécis qu'à se tromper. »
>
| [Prev in Thread] | Current Thread | [Next in Thread] |