-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
On Thursday, July 29 at 11:32 AM, quoth Paul E Condon:
>I'm still having problems. I'd like to read some documentation that
>expands on what these configuration lines do.
Without a better idea of what you're after, I'd say the Mutt manual is
the place to look up the definition of charset-hook.
>I can't find any mention of unknown-8bits or x-user-defined in the
>Mutt E-Mail Client manual.
And you won't; that's just experience talking. I've seen emails that
label their own charset as "unknown-8bits" or "x-user-defined". For
example, one of them was generated by pine when the user was trying to
respond to a utf-8 message; the message had quoted the utf-8
characters, but pine could only identify them as "not latin-1", and so
fell back to "unknown-8bits". I forget where I saw x-user-defined...
the point is, they're bogus character-set labels instituted by
software that's gotten confused or that doesn't have support for a
modern charset or both.
> The charset-hook command defines an alias for a character set. This
> is useful to properly display messages which are tagged with a
> character set name not known to Mutt.
>
> The iconv-hook command defines a system-specific name for a
> character set. This is helpful when your systems character
> conversion library insists on using strange, system-specific names
> for character sets.
>___________________________
> In this, iconv-hook is described as a method of handling a
> 'character set name' that is not known to Mutt. Is there a place
> where I can find a list of the character set names that are known to
> the copy of Mutt on my machine? Where? How? Or (gently, please) why
> is this a silly question?
You're right, the distinction is quite fuzzy. The way I would think of
it is this way:
charset-hook is used for mapping a weird/unknown/wrong charset
name onto the "correct" charset name.
iconv-hook is used for mapping a "correct" charset name onto a
system-specific one (i.e. if your system is broken).
The example for iconv-hook given in the mutt man page is for
transforming "iso-8859-1" (i.e. the "correct" name) to a
system-specific "8859-1". Thus, they can stack: the charset-hook can
transform latin1 (and many others) into iso-8859-1, and iconv-hook
munges iso-8859-1 into whatever your system prefers.
In practice, however, I believe you can use them both to achieve the
same effect. In other words, both can transform latin1 into 8859-1 (or
whatever) by themselves; the only reason that both exist is to allow
for a certain amount of stacking---in other words, the charset-hook
hooks (for "correcting" or "canonicalizing" incoming charset names)
can be distributed among many systems, while the iconv-hooks (for
translating "correct" or "canonicalized" names into system-specific
non-canonical names) should only be true for a single system (and
probably one that has a broken iconv implementation).
Does that make sense?
~Kyle
- --
Always forgive your enemies; nothing annoys them so much.
-- Oscar Wilde
-----BEGIN PGP SIGNATURE-----
Comment: Thank you for using encryption!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=30fE
-----END PGP SIGNATURE-----
