mftrace 1.2.20 chokes on a file

[sl2c]

Dear Han-Wen,

Thanks for mftrace! I am loving it. Here are my two cents: looks like mftrace 1.2.20 chokes on some files, e.g.:

sample.zip

The error:

s = (self.left[1:1 + b]).decode('ascii')
UnicodeDecodeError: 'ascii' codec can't decode byte 0xa0 in position 0: ordinal not in range(128)

The command was:

mftrace -f PFB zcmib5.mf

I’ve run this with mftrace 1.2.19, and all went smooth with that version.

Best regards!

[lodwicz]

hello! apparently, something is wrong with the tfm file itself, as delivered by metafont, for instance: but if you fix the tfm file, mftrace runs happily.

Try

tftopl zcmib5.tfm zcmib5.pl
pltotf zcmib5.pl zcmib5.tfm
mftrace zcmbi5

It worked for a hack I'm doing with some fonts.

Apparently, in my case, the metafont code failed to provide a coding scheme required by tfm.py, and thus mftrace failed to process the tfm file, trying to find an ascii string where there was none.

I bypassed this problem either by fixing the tfm file via the tf-pl-tf route, which now includes an "UNDEFINED" ascii string for the encoding searched by tfm.py; or else by providing an encoding above ascii at line 39 of tfm.py, so that mftrace may read a string where there was none (which one, I have no clue), and thus kept running. By testing a little, I found that cp1252 or cp858 are fairly safe choices, being widely used, ascii compatible, and leaving no empty slots, unlike latin1, which may leave some,