matplotlib-users Mailing List for matplotlib

Thought I'd summarize my experiences. They may result from a messed up =20=
installation, but could be useful for others -- Bradley Minch, using =20
10.2.8, seems to have had my initial problem, #I below.
Excuse the overlong email here.
Many thanks to Darren Dale for helping me out.
Setup:
matplotlib 0.80, ipython 0.6.13, readline.so [from =20
www.pycs.net/bbum/2004/1/21/readline.so.gz]
 zlib, libpng, tk_inter8.4, freetype 2.1.9 [produces =20
libfreetype.6.3.7.dylib], wx-2.6-mac-unicode,
 standard Panther Apple Python 2.3, updated to MacPython,
 scipy 0.3.2 [from =20
http://cbis.anu.edu.au/misc/SciPy_complete-0.3.2.mpkg.zip]
Numeric 23.1, OS X 10.3.9
Running with TkAgg backend set in .matplotlibrc
1. Original problem:
First time I did
from pylab import * (or ipython -pylab)
got error while font_manager was trying to build the ttffont.cache...
 =C2=A0 File =20
"/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-=20=
packages/matplotlib/font_manager.py", line 347, in ttfFontProperty
 =C2=A0=C2=A0=C2=A0 size =3D str(float(font.get_fontsize()))
 AttributeError: get_fontsize
Now, when building the ttf cache, font_manager searches for TrueType =20
fonts.
1. in the matplotlib data directory =20
/System/Library/Frameworks/Python.framework/Versions/2.3/share/=20
matplotlib
2. In the normal Mac places
/Library/Fonts/, /System/Library/Fonts,/Network/Library/Fonts, =20
~/Library/Fonts
Tracked this problem down to the font VT102Font that I had in =20
~/Library/Fonts.
Moved VT102Font to (new directory) ~/SuspectFonts off the font path.
Problem 1 solved! from pylab import * now happily built a =20
~/.ttffont.cache for me.
2. Next problem -- absurdly large, corrupted, PS files.
Ran this script that produces simple plots, with axes labelled (by =20
default) with sans-serif numbers.
dt =3D 0.01
t =3D arange(0,10,dt)
subplot(211)
s =3D sin(t)
plot(t,s)
subplot(212)
c =3D cos(t)
plot(t,c)
savefig('/Users/agn/sincos.png')
 -- produced a nice file ~28KB in size
savefig('/Users/agn/sincos.ps')
 -- produced a file of 4.6MB.
 Trying to view this file using Apple preview, or converting it to pdf =20=
with ps2pdf (ps2pdf13, v1.3, 21Feb2002) or Apple pstopdf gave an error. =20=
E.g. ps2pdf sincos.ps gave:
Error: /invalidfont in -dict-
Operand stack:
 LucidaGrande --dict:10/14(L)-- Font LucidaGrande =20
--dict:10/14(L)-- LucidaGrande
Darren Dale sorted me out here. The default sans-serif font used to =20
label the axes is the first font in the san-serif line (about line 113) =20=
in the .matplotlibrc startup file. The default .matplotlibrc gives =20
LucidaGrande as that default font. LucidaGrande.dfont is an Apple =20
standard font living in /System/Library/Fonts.
He suggested that I put Bitstream Vera Sans as the first font on the =20=
san-serif line in .matplotlibrc, so that this font, which comes with =20
matplotlib, and lives in =20
/System/Library/Frameworks/Python.framework/Versions/2.3/share/=20
matplotlib is used as the default sans-serif font instead.
Problem 2 solved. savefig('/Users/agn/sincos.ps') now gives a 176 KB =20
ps, that reduces to a 36KB pdf. Perfectly OK now.
Suspected same problem would arise for other families, so put Bitstream =20=
Vera Serif as the default serif font, & Bitstream Vera Sans Mono as =20
the default monospace font.
3. John Hunter suggested that problem 2 might be solved if I were to =20
modify line 180 of .matplotlibrc to set
 ps.useafm : True
This should give smaller Ps files at the cost of breaking mathtext =20
(which incidentally *does* (almost completely) work for me with =20
ps.useafm : False).
Unfortunately, setting ps.useafm true doesn't work for me. Now if =20
ps.useafm is True, font_manager only makes a cache -- .afmfont.cache =20=
-- when you first try to save to PS. I have at present the problem that =20=
font_manager is falling over when trying to make this cache. One of the =20=
Chinese-type opentype (.otf) fonts --- =
/System/Library/Fonts/=E3=83=92=E3=83=A9=E3=82=AD=E3=82=99=E3=83=8E=E6=98=8E=
=E6=9C=9D =20
Pro W3.otf -- or in hex
/System/Library/Fonts/=20
\xe3\x83\x92\xe3\x83\xa9\xe3\x82\xad\xe3\x82\x99\xe3\x83\x8e\xe6\x98\x8e=20=
\xe6\x9c\x9d Pro W3.otf
causes an infinite loop with 'Key error converting in AFM' in afm.py, =20=
when trying to parse the header.
Quite a few fonts are in 'seen', --
{'pbkd8a.afm': 1, 'pagd8a.afm': 1, 'pbkl8a.afm': 1, 'ptmri8a.afm': 1, =20=
'pncb8a.afm': 1, 'phvbo8a.afm': 1, 'phvb8an.afm': 1, 'phvro8an.afm': 1, =20=
'ptmb8a.afm': 1, 'pcrro8a.afm': 1, 'pcrb8a.afm': 1, 'pplr8a.afm': 1, =20
'pzdr.afm': 1, 'cmsy10.afm': 1, 'phvlo8a.afm': 1, 'cmr10.afm': 1, =20
'phvb8a.afm': 1, 'phvl8a.afm': 1, 'pncri8a.afm': 1, 'pagdo8a.afm': 1, =20=
'putri8a.afm': 1, 'pzcmi8a.afm': 1, 'ptmbi8a.afm': 1, 'phvro8a.afm': 1, =20=
'putr8a.afm': 1, 'phvbo8an.afm': 1, 'pcrr8a.afm': 1, 'putbi8a.afm': 1, =20=
'pcrbo8a.afm': 1, 'cmtt10.afm': 1, 'pplb8a.afm': 1, 'pbkdi8a.afm': 1, =20=
'pagk8a.afm': 1, 'pncr8a.afm': 1, 'putb8a.afm': 1, 'psyr.afm': 1, =20
'pagko8a.afm': 1, 'pplbi8a.afm': 1, 'pbkli8a.afm': 1, 'pncbi8a.afm': 1, =20=
'cmmi10.afm': 1, 'ptmr8a.afm': 1, 'pplri8a.afm': 1, 'cmex10.afm': 1, =20
'phvr8a.afm': 1, =20
'\xe3\x83\x92\xe3\x83\xa9\xe3\x82\xad\xe3\x82\x99\xe3\x83\x8e\xe6\x98\x8=20=
e\xe6\x9c\x9d Pro W3.otf': 1, 'phvr8an.afm': 1}
so have presumably been successfully parsed, except for this .otf font.
Haven't sorted this problem out yet. I suppose I could try removing the =20=
East Asian fonts from the system folder, or changing font_manager so it =20=
does not read them in.
4. Conclusion: For me, the only fonts that work are the BitStream Vera =20=
Fonts supplied with matplotlib. Not ideal, because they do do not =20
include italic, cursive, or fantasy fonts (although they do include an =20=
*oblique* sans serif font). But this is sufficient to make matplotlib =20=
work for most practical purposes, especially as mathtext does work =20
(though ps2pdf can't convert the ps fonts to pdf; Apple pstopdf (used =20=
by Preview) must be used)
Again, many thanks to Darren Dale and also to John Hunter and Fernando =20=
Perez for their replies.
George Nurser.

>>>>> "Ryan" == Ryan Krauss <rya...@co...> writes:
 Ryan> Setting the pad in the .matplotlibrc file was an excellent
 Ryan> solution for me. And being able to set all my font
 Ryan> preferences there is a great feature that I don't think
 Ryan> Matlab has. One of the first things I needed to do in Matlab
 Ryan> was write a script that formatted my figures the way I
 Ryan> liked. I assumed I needed to do the same thing in mpl, but
 Ryan> not so. Good Stuff!
As a long time matlab user, I grew tired of all the arcane commands
you have to execute to customize the figure. So I tried to make the
most common stuff configurable in the rc file. One feature I use a lot
is directory level customization. mpl looks in the current directory
first for the rc file, then HOME and then in the site locations.
Typically, I want different defaults for the web, for publication and
for the screen, and use directory specific rc file for these. See
also the "rc" command, for changing the defaults on a per script
basis...
JDH

Setting the pad in the .matplotlibrc file was an excellent solution for 
me. And being able to set all my font preferences there is a great 
feature that I don't think Matlab has. One of the first things I needed 
to do in Matlab was write a script that formatted my figures the way I 
liked. I assumed I needed to do the same thing in mpl, but not so. Good 
Stuff!
Ryan
John Hunter wrote:
>>>>>>"Ryan" == Ryan Krauss <rya...@co...> writes:
>>>>>> 
>>>>>>
>
> Ryan> I have a question about tick formatting. I have a semilogx
> Ryan> plot and if I resize the xticks using locs , labels = xticks
> Ryan> () set(labels , size=ticksize) I like the size and the font,
> Ryan> but the exponents (10^0) run into the axis.
>
>The reasons for are complicated and have to do with an optimization to
>make exponential ticking faster, and I won't go into them right now.
>Suffice it to say that it is a bug, but there may be a workaround
>
>Do any of the suggestions here help?
>
> http://matplotlib.sourceforge.net/faq.html#TEXTOVERLAP
>
>Note you can also control the "pad" in points of the offset of the
>ticks from the xaxis
>
> ticks = ax.xaxis.get_major_ticks()
> for tick in ticks:
> tick.set_pad(6)
>
>or if you prefer
>
> set(ax.xaxis.get_major_ticks(), pad=6)
>
>The default pad is controlled by an rc parameter
>
> tick.major.pad : 4 # distance to major tick label in points
> tick.minor.pad : 4 # distance to the minor tick label in points
>
>See http://matplotlib.sf.net/.matplotlibrc
>
>JDH
>
> 
>