Re: [bug-gettext] Glade msgctxt patch

[Miguel Ángel Arruga Vivas]

Hi Daiki,

Daiki Ueno <address@hidden> writes: 
> Well, sorry, I don't like it.  I didn't strongly suggest to use
> arglist_parser_* functions, but just asked if there is any possibility
> to avoid having a new function like split_glib_syntax_function, just to
> split a string with '|' as a seperator.

I actually prefer arglist_parser functions family, because It eases a
little bit GtkBuilder implementation, but there is no real problem,
because It also adds lots of unnecessary instructions and allocations to
the code path.

But before rebasing it, do you prefer "switch (type) { case
GLADE1: ... }" to hook functions? Because get_shapes is one of them. ;-)

> Your initial attempt was rather better, but I no longer have the
> original patch (because it has never posted here and your github repo
> was rebased), so I recovered it from my memory and modified a bit.

If you mean the one with split_glib_syntax_string I could send it, the
other one is at
<http://lists.gnu.org/archive/html/bug-gettext/2013-01/msg00022.html>
but your patch seems better. My local repo already has every patch I
have done, feel free to ask for old (and probably buggy) versions.

> (I know that the patch has small code dupes in xgettext.c and x-glade.c,
> which I think acceptable.  And also it doesn't support GtkBuilder-style
> context attribute yet, but I'd like to get things done step by step.)

Of course, It will be done step by step because also helps a lot with
bugs traceability. The dupes are not a problem.

> Anyway, if there is no serious issue, I'll push it to git.

No problem seen.

Also, for the tests patch, should I split msgctxt extraction test
(xgettext-glade-5) and atkproperty test case (xgettext-glade-4) in two
commits?

Best regards,
Miguel