Re: [Denemo-devel] Denemo warnings and code cleanup

[Richard Shann]

On Fri, 2013-05-17 at 14:11 +0200, Éloi Rivard wrote:
> In keymapio.c:58 create_command is called when the tooltip tag is
> added. Is there (still?) a reason to keep it there and force tooltip
> being the last tag, or can put it out of the loop ?

I cannot see a reason why it was  ever inside the loop. It makes it
dependent on the order in the xml file, which is commented on in the
code as being a bad thing (that is my comment).
The things to check are that a user can create a new command, can load a
command that is not in the default set, can save a command set with
additional commands and can load a command set. When a user has created
a command this should appear in the menu straight away, even if the user
has decided not to include it in their default commands.
There is a flag to indicate that the command set has been modified, but
this was partially deactivated by someone a few years ago and I haven't
got round to re-activating it; I am not sure what the current situation
is with it. (It will be a field in the singleton object DenemoRoot
Denemo; which contains all globals.)

Richard



> 
> 
> 
> 2013/5/17 Richard Shann <address@hidden>
>         yes - ok, I have done that.
>         Richard
>         On Thu, 2013-05-16 at 21:13 +0200, Éloi Rivard wrote:
>         > Do you think you can tell your editor to use whitespaces
>         instead of
>         > tabulations ? It produces stranges indentation in gedit,
>         anjuta or
>         > cgit.
>         >
>         >
>         >
>         > 2013/5/16 Richard Shann <address@hidden>
>         >         On Thu, 2013-05-16 at 17:23 +0200, Éloi Rivard
>         wrote:
>         >         > In figure.c, line 85.
>         >         > ((chord *) curObj->object)->figure has a chance to
>         be used
>         >         elsewhere ?
>         >
>         >         yes
>         >         >  Else, g_free(((chord *) curObj->object)->figure);
>         could do
>         >         the job,
>         >         > couldn't it ?
>         >
>         >         No, it is a GString *, the old value (if any) needed
>         >         de-allocating: I
>         >         have done this, in git now.
>         >
>         >         Richard
>         >
>         >         >
>         >         >
>         >         >
>         >         > 2013/5/16 Éloi Rivard <address@hidden>
>         >         >         All -Wall warnings are fixed in master,
>         plus some
>         >         -Wextra. I
>         >         >         will take care of other -Wextra warnings
>         later.
>         >         >
>         >         >
>         >         >
>         >         >         2013/5/15 Richard Shann
>         <address@hidden>
>         >         >                 On Wed, 2013-05-15 at 10:36 +0200,
>         Éloi
>         >         Rivard wrote:
>         >         >                 >         I have always been
>         daunted by this
>         >         task. As
>         >         >                 you have found
>         >         >                 >         trying to
>         >         >                 >         refactor the keymap code
>         it is a
>         >         high risk
>         >         >                 occupation. I am
>         >         >                 >         not sure if
>         >         >                 >         I can honestly say that
>         I have the
>         >         courage
>         >         >                 to tackle it.
>         >         >                 >         Perhaps if we
>         >         >                 >         created a map of the
>         file, that
>         >         would be a
>         >         >                 start...
>         >         >                 >
>         >         >                 > You mean separate functions in
>         the
>         >         categories you
>         >         >                 proposed ?
>         >         >
>         >         >
>         >         >                 I meant a list of line number
>         ranges for the
>         >         different
>         >         >                 categories. I
>         >         >                 hope they are virtually all
>         contiguous.
>         >         >
>         >         >                 Richard
>         >         >
>         >         >
>         >         >
>         >         >
>         >         >
>         >         >
>         >         >         --
>         >         >         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. »
>         >
>         
>         
>         
> 
> 
> 
> -- 
> Éloi Rivard - address@hidden
>         
> « On perd plus à être indécis qu'à se tromper. »
>