[Nmh-workers] charset conversions

[Paul Fox]

i've finally gotten my workspaces set up so that i think i'm fully UTF-8
compatible in xterm, less, vile, etc.

i just got mail from someone that included an accented 'a', and it
didn't display correctly.  some notes on my experience:

first, adding the line:
    mhshow-charset-iso-8859-1: %s |
            /tmp/iconvwrap /usr/bin/iconv -f ISO-8859-1 -t UTF-8
to my mh profile fixed this, after some experimentation.  is there
a better way?

experimentation was necessary because it turns out that the only
listed enumeration of %-escapes in the mhshow manpage doesn't apply
here -- it's for mh-show-foo display strings, where %s is the content
subtype.  but when converting charsets, %s is "filled-in with the
command to display the content", which is a little ambiguous.  (i.e., what
sort of command?  is it a pipeline?  should i try and insert before,
or after, the existing command?)  and, are there any other %-escapes
relevant to charset conversion?  in particular, %f doesn't seem to be
valid when converting charsets, but it's honored in any case -- it
expands to a bunch of junk and tries to execute it.

finally, the docs for mhshow, at least, probably need some updating for
the unicode world.

to be clear, i don't consider these to be 1.5 issues, specifically, and
i'm happy to open tickets for any of these if that would be helpful.

paul
=---------------------
 paul fox, address@hidden (arlington, ma, where it's 56.7 degrees)