Re: Dia goes Gtk2, anyone else with it ?

Le Sat, Jun 08, 2002, à 02:17:50PM +0200, Hans Breuer a écrit:

I'll take care of that this week-end, using James' work as a basis.

Nice to hear that.

it's almost done now. There was some remaining #hell in UML objects; and I
got rid of charconv.[ch] entirely.

(crossing fingers for the linker stage)

one thing: libxml2 forced me to change all instances of 
#include "tree.h"  ---> #include "libxml/tree.h"

(same for a couple other include files). I hope that'll be OK with your
project settins.

app/render_libart.c probably broken on X11

Actually, obsolete. We'll get rid of it.
What do you mean is obsolete :-) X11, libart ? or our nice libart
renderer, which has some X11 specific tweaking which I don't
understand nor use on win32 ...


app/sheet.c:207 move message_notice out of loop
app/ps_utf8.c : not yet build on win32

What's wrong with it ?

still some #ifdefs and charconv_*.

yep, done converting to glib

app/render_eps.c : some further utf8 cleaning needed


tabs are evil. XIM still seems applicable, where GTK2 doesn't already take
care of it !

AFAIK the Gtk2 way to do it has changed. So all within HAVE_XIM
will currently break with Gtk2.

Indeed (fixing while answering your mail). For now, I'll just disable the
XIM case (but I'll leave #hell HAVE_XIM until the Asian guys complain)

charconv is still needed, at least to recover the pre 0.90 dia files (damn
Nope. It all can be done with glib2 - in fact it already is.

eventually saw that.
There are some widgets in Gtk1.2 which were considered so broken,
that a complete re-write appeared to be appropriate in Gtk2. 
They are GtkText (replacement GtkTextBuffer/GtkTextView) and GtkTree 
(replaced with GtkTreeView GtkTreeModel). To still use the old
widgets one needs to define GTK_ENABLE_BROKEN.

I see
What's the status of the UTF8 #hell (UNICODE_WORK_IN_PROGRESS and

Almost all vanished already, beside the places noted above.
But please let us next time do such stuff on it's own
branch to keep patch size small :-)

Hmmm... ;-)

        -- Cyrille


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]