rcParams["text.usetex"]=True
t = text(0.5, 0.5, "field", size=50)
from matplotlib.patheffects import withStroke
t.set_path_effects([withStroke(foreground="red", linewidth=5)])

and "fi" is not displayed.

The patch tries to convert the glyph id reported by the dviread module to the proper glyph id for the font to be used using the encoding information.

By the way, I'm hardly an expert on this subject and I hope some one more knowledgeable than me review the patch and see if I get this right.

-JJ

textpath tries the adobe standard encoding as a default for fonts

efd1c4b

Copy link

Member

mdboom commentedMay 11, 2011

I think@jkseppan is probably the best person to review this.

Copy link

Member

Sorry for the slow reaction. The patch looks fine to me, although I am no expert in Freetype. Since it seems that a lot of the font-related code must be written by trial and error in the absence of good documentation, I would like there to be test cases for each successive improvement. How difficult would it be to write a test case from the "field" example? I think we already ship some TeX fonts, so we could use one of them and get pretty repeatable output.

But I'm fine with merging this as-is.

efiring added a commit that referenced this pull request

Jul 7, 2011

Merge pull request#112from leejjoon/fix_textpath

11d2c52

textpath tries the adobe standard encoding as a default for fonts

efiring merged commit11d2c52 intomatplotlib:master

Jul 7, 2011

Copy link

Member

efiring commentedJul 7, 2011

The test idea is good, but I think it is best not to let it hold up the merge of the present change. If anyone is inspired to add a test, it can be done at any time.