Hi, The superscripts in mpl don't seem to be placed at the correct height for small fonts. The y-tics on the attached plot shows this. The effect is similar in svg and pdf backends. I'm using 0.91.2 because the latest svn version gives me a KeyError for ufunc isfinite. - Paul
Paul Kienzle wrote: > Hi, > > The superscripts in mpl don't seem to be placed at the correct height for > small fonts. The y-tics on the attached plot shows this. The effect is > similar in svg and pdf backends. > Thanks. When looking up the x-height, the code was not taking the font size into account. This has been fixed on the branch and trunk. > I'm using 0.91.2 because the latest svn version gives me a KeyError for > ufunc isfinite. > Can you provide the full traceback here so we can diagnose and fix this problem? The only reference to isfinite seems to be in the PDF backend, which is working for me, and is identical in 0.91.2 and trunk in that respect. There is probably a more complex Numpy version interaction going on. Cheers, Mike > - Paul > > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save 100ドル. > Use priority code J8TL2D2. > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > ------------------------------------------------------------------------ > > _______________________________________________ > Matplotlib-devel mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel > -- Michael Droettboom Science Software Branch Operations and Engineering Division Space Telescope Science Institute Operated by AURA for NASA
On Thu, Apr 24, 2008 at 08:38:02AM -0400, Michael Droettboom wrote: > Paul Kienzle wrote: > > Hi, > > > > The superscripts in mpl don't seem to be placed at the correct height for > > small fonts. The y-tics on the attached plot shows this. The effect is > > similar in svg and pdf backends. > > > Thanks. When looking up the x-height, the code was not taking the font > size into account. This has been fixed on the branch and trunk. Thanks. It looks much better in png, but pdf has the subscript raised much higher, and svg isn't rendering fonts at all (see attached). > > I'm using 0.91.2 because the latest svn version gives me a KeyError for > > ufunc isfinite. > > > Can you provide the full traceback here so we can diagnose and fix this > problem? The only reference to isfinite seems to be in the PDF backend, > which is working for me, and is identical in 0.91.2 and trunk in that > respect. There is probably a more complex Numpy version interaction > going on. I recompiled with Numpy 1.1 and the problem went away. I'm going to ignore the issue for now. - Paul
Paul Kienzle wrote: > On Thu, Apr 24, 2008 at 08:38:02AM -0400, Michael Droettboom wrote: > >> Paul Kienzle wrote: >> >>> Hi, >>> >>> The superscripts in mpl don't seem to be placed at the correct height for >>> small fonts. The y-tics on the attached plot shows this. The effect is >>> similar in svg and pdf backends. >>> >>> >> Thanks. When looking up the x-height, the code was not taking the font >> size into account. This has been fixed on the branch and trunk. >> > > Thanks. It looks much better in png, but pdf has the subscript raised much > higher, That was a DPI-related bug, that is hopefully now fixed in SVN. > and svg isn't rendering fonts at all (see attached). > Thanks. The SVG was broken in SVN March 26 trying to workaround an Inkscape copy/paste bug. I think it is fixed now so it works in Mozilla and Inkscape simultaneously. But please let me know if you see anything fishy -- there seems to be a great deal of variability between SVG viewers. Are you seeing this bug with 0.91.2 release, or just the 0.91.2 SVN branch? I'd be surprised by the former (and I can't reproduce it there). Cheers, and thanks for your patience, Mike -- Michael Droettboom Science Software Branch Operations and Engineering Division Space Telescope Science Institute Operated by AURA for NASA
On Mon, Apr 28, 2008 at 09:51:54AM -0400, Michael Droettboom wrote: > Paul Kienzle wrote: > > > > Thanks. It looks much better in png, but pdf has the subscript raised much > > higher, > That was a DPI-related bug, that is hopefully now fixed in SVN. > > and svg isn't rendering fonts at all (see attached). Yes. > Thanks. The SVG was broken in SVN March 26 trying to workaround an > Inkscape copy/paste bug. I think it is fixed now so it works in Mozilla > and Inkscape simultaneously. But please let me know if you see anything > fishy -- there seems to be a great deal of variability between SVG viewers. Works in Safari 3.1.1, Inkscape 0.45 and FF 2.0.0.14 > Are you seeing this bug with 0.91.2 release, or just the 0.91.2 SVN > branch? I'd be surprised by the former (and I can't reproduce it there). 0.91.2 SVG does indeed work (with the old superscript position of course). - Paul
Paul Kienzle wrote: > On Mon, Apr 28, 2008 at 09:51:54AM -0400, Michael Droettboom wrote: > >> Paul Kienzle wrote: >> >>> Thanks. It looks much better in png, but pdf has the subscript raised much >>> higher, >>> >> That was a DPI-related bug, that is hopefully now fixed in SVN. >> >>> and svg isn't rendering fonts at all (see attached). >>> > > Yes. > > >> Thanks. The SVG was broken in SVN March 26 trying to workaround an >> Inkscape copy/paste bug. I think it is fixed now so it works in Mozilla >> and Inkscape simultaneously. But please let me know if you see anything >> fishy -- there seems to be a great deal of variability between SVG viewers. >> > > Works in Safari 3.1.1, Inkscape 0.45 and FF 2.0.0.14 > > >> Are you seeing this bug with 0.91.2 release, or just the 0.91.2 SVN >> branch? I'd be surprised by the former (and I can't reproduce it there). >> > > 0.91.2 SVG does indeed work (with the old superscript position of course). > I'm surprised by this. Can you do a full rebuild and try again? It is working for me now. Cheers, Mike -- Michael Droettboom Science Software Branch Operations and Engineering Division Space Telescope Science Institute Operated by AURA for NASA
Michael Droettboom wrote: > Paul Kienzle wrote: > >> On Mon, Apr 28, 2008 at 09:51:54AM -0400, Michael Droettboom wrote: >> >> >>> Paul Kienzle wrote: >>> >>> >>>> Thanks. It looks much better in png, but pdf has the subscript raised much >>>> higher, >>>> >>>> >>> That was a DPI-related bug, that is hopefully now fixed in SVN. >>> >>> >>>> and svg isn't rendering fonts at all (see attached). >>>> >>>> >> Yes. >> >> >> >>> Thanks. The SVG was broken in SVN March 26 trying to workaround an >>> Inkscape copy/paste bug. I think it is fixed now so it works in Mozilla >>> and Inkscape simultaneously. But please let me know if you see anything >>> fishy -- there seems to be a great deal of variability between SVG viewers. >>> >>> >> Works in Safari 3.1.1, Inkscape 0.45 and FF 2.0.0.14 >> >> >> >>> Are you seeing this bug with 0.91.2 release, or just the 0.91.2 SVN >>> branch? I'd be surprised by the former (and I can't reproduce it there). >>> >>> >> 0.91.2 SVG does indeed work (with the old superscript position of course). >> >> Apologies -- I was just confused. The 0.91.2 *release* should (as you say) have working text but with the wrong superscript position. The 0.91.x *maintenance branch* should be working on both counts. Cheers, Mike -- Michael Droettboom Science Software Branch Operations and Engineering Division Space Telescope Science Institute Operated by AURA for NASA