The feature where you click on graph elements and are taken to the
correct place in the editing tree (in 0.8-cvs) is great. But it'd be
better if you didn't have to click on the exact pixel to get to the
data points bit. It should be more specific too, like taking you to
the title if you click on i
-- Forwarded message --
From: Jamie C <[EMAIL PROTECTED]>
Date: Jun 19, 2005 7:36 PM
Subject: Re: [Veusz-discuss] Windows binary
To: Jeremy Sanders <[EMAIL PROTECTED]>
On 6/13/05, Jeremy Sanders <[EMAIL PROTECTED]> wrote:
> On Sun, 12 Jun 2005, Jamie C
That patch totally breaks it, but using 1. instead of 75. is perfect. Fixed!
Installing on windows? Depends how you want to do it. In theory I can
just zip my python directory and send it to you, that plus a couple of
DLLs would be quick and easy. Alternatively you can go through all the
stuff in:
> Jamie - can you try out the new code in cvs on Windows??
>
> Thanks
>
> Jeremy
Well, no exceptions this time, but it still does a tiny graph in a big
page. Increasing the page size scales the graph but still leaves a
huge margin. So it's not fixed that problem.
Error message on close:
Error w
You need an "import os" (or "import os.path") in commandinterpreter.py
or running GPL() doesn't work on windows (because you use os.path to
open COPYING). Porting to other platforms really is good for finding
bugs...
Also, it comes out in a narrow column instead of wrapping nicely, and
it has a few
ROTECTED]> wrote:
> > On Tue, 24 May 2005, J. Graham wrote:
> >
> > > On Tue, 24 May 2005, Jamie C wrote:
> > >
> > >> Had a quick go at it on my machine.
> >
> > > It looks like there's a few evaluation options or a port of the free unix
> > settingdb gives an exception on close, but other than that and the
> > weird bug where the graph is postage-stamp sized
>
> That's probably because it can't find the geometry settings.
Makes sense.
> I've had a look at moving SettingsDb to QSettings and it seems pretty
> easy... Well I can ge
http://www-xray.ast.cam.ac.uk/~jc/veusz.png
settingdb gives an exception on close, but other than that and the
weird bug where the graph is postage-stamp sized it seems to pretty
much work.
I dealt with the issue where you do "import directory.pythonfile" by
just putting everything in the same dir
Had a quick go at it on my machine.
py2exe fails when it tries to import Qt. Qt for windows isn't free, so
adding it would be a pain.
The developers do let you use Qt 2.3 (fairly old) on windows for free
- will Veusz work with this, Jeremy? This also has a catch in that it
will only compile with M
I put my Sunday hangover to good use being critical of Veusz... please
excuse this email if it sounds a bit grumpy, or if this stuff has come
up before.
Bugs:
Veusz crashes when trying to display: 5*10^5 points with asymmetric
errors. 10^5 is fine... I haven't bothered to find the exact number a
> The dialog box title is "veusz <2>" (or "veusz "). Something like
> "confirm close" would be nicer.
After a marathon coding session I've fixed this major bug, by changing
the main title bar to read "Untitled - Veusz" instead of "Veusz" when
working on something that isn't saved. Dialog box title
and major
tick not being on top of each other. Other similar combinations (i.e
0.00 - 0.04) work perfectly, which is odd.
I have a veusz save to show what I mean:
http://www-xray.ast.cam.ac.uk/~jc/greek_test.vsz
Jamie
On 4/18/05, Jeremy Sanders <[EMAIL PROTECTED]> wrote:
> On Sun, 1
Users - more trouble than they're worth?
Thanks for the rpms - very slick.
Anyway, bugs and requests. I might have mentioned these but might as
well put them on record:
Starting and immediately quitting brings up a "document is modified,
save first?" dialog.
The dialog box title is "veusz <2>" (
I'm a lousy programmer, so I'll confine myself to criticising your UI.
Which is actually pretty good, but:
There's no way in the GUI (that I can figure out) to delete widgets,
i.e. use Remove().
I think your idea of a filesystem-esque hierarchy is great, but the
syntax is pretty clunky... To('/pa
14 matches
Mail list logo
