F
Fernando Perez
Torsten said:Hallöchen!
Fernando Perez said:[...]
Well, it's true that the latex-type (called mathtext) support in
matplotlib is not really up to par with true latex (kerning is off
in places, mixed text/math doesn't work well, etc). I've been
willing to live with it so far, but an alternative option is to
use the PS backend and then play psfrag tricks.
The problem is that mostly, you'll have a *lot* to substitute.
I've yet to experiment with it, but it might (with some additional
handywork) give final results identical to those of the pslatex
backend in gnuplot.
What do you mean with this? Do you want to mimic TeX's quality as a
typesetter, or do you think the goal should be output in real LaTeX
format (like pslatex does)? The latter would be more useful in my
opinion, and much easier, too.
Easier... psfrag tricks can be done right now, while full latex output requires
writing a new matplotlib backend. It would certainly be a _great_ project,
but not one I'm about to undertake, while it's reasonable for me to use psfrag
to fix a few labels here and there to use proper latex. So while I agree with
you that in the long run a latex backend would be ideal, as a stop-gap
solution I can live with psfrag.
[...] But there are a number of things it simply can't offer due
to its design as a standalone program, which matplotlib (being a
library/widget collection) can do much better. [...] I finally
made the switch and I'm extremely happy.
I'm not a fanatic Gnuplot user either. I use it for 11 years, and I
like exactly two things about it: its simplicity and the pslatex
backend. I think for my thesis I'll still use it, because its
integration in a batch process that builds my thesis is much easier
than to write Python programs.
well, unless your batch process _is_ in python Mine was, so my
make_plots.py was a single script, which ironically (for this discussion) was
all gnuplot.py-based, since this was a few years ago.
Cheers,
f