[Erc-discuss] erc-encoding-coding-alist matching problems

[Johan Bockgård]

Channel names in `erc-encoding-coding-alist' should be matched
case-insensitively (or names should be canonicalized).

I was on the #FooBar channel and had a ("#FooBar" . latin-1) entry in
erc-encoding-coding-alist. On receiving text #FooBar was used as the
target name, but on sending #foobar was used (and thus didn't match).
I didn't know that the text I sent looked weird (utf-8) to other
people until they told me (since the text i received looked just
fine).

Btw, according to the doc the coding alist is supposed to use
regexps--"Alist of target regexp and coding-system pairs to use"--but
the code uses `assoc'. This has been broken for 3 years.

-- 
Johan Bockgård