mailRe: r2863 - /branches/test_suite/test_suite/unit_tests/unit_test_runner.py


Others Months | Index by Date | Thread Index
>>   [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Header


Content

Posted by Edward d'Auvergne on November 21, 2006 - 16:55:
On 11/22/06, Gary S. Thompson <garyt@xxxxxxxxxxxxxxx> wrote:
Edward d'Auvergne wrote:

> Are you having a play with the unittest GUI?


I was just having a go to see that it worked. The version I was using, however, is under the wrong license (I stole it from zope I think) so it was just a test of the code/concept and will most probably disappear again.

> If so, is it any good?
> Pity it's in Tk though,

Why the problem with tk? Tk is nice (mostly)  and is present in all
python distributions (note the prejudice, I am an old tcl/tk programmer)

There's nothing wrong with Tk. Wish was the first widget set I played with, and I do think it is very useful for certain applications. Tk does need to be modernised though, it desperately needs font antialiasing! And the Tk widgets could do with a lot of polish.

I ported an early version of relax (0.2 which was called 'mf' back
then) to use the Gtk widget set.  I called it 'xmf' but unfortunately
I never added it to the relax repository.  It was pretty pathetic -
not destroying flexibility while implementing a GUI is a very hard
job.  Qt support in Python was and still is light years ahead of Tk,
Gtk, etc.  Implementing a complex GUI for relax would significantly
benefit from using the advanced Qt bindings.  For 3D opengl work, for
example assembling data pipes (when runs are renamed to pipes) to
create the equivalent of a script, the stable Qt bindings would be
better for sanity of the developers :)

Edward



Related Messages


Powered by MHonArc, Updated Tue Nov 21 17:20:45 2006