Re: Order of associations in compilation-error-regexp-alist-alist

[Chong Yidong]

Eli Zaretskii <address@hidden> writes:

> This means that if an error message matches more than one regexp
> there, the first one wins.

Yep, and that's the way it's worked for ages.

> If the latter, how do we make sure only the correct regexp matches?

By cunning choice of ordering in compilation-error-regexp-alist.

See also test/automated/compile-tests.el, which is a regression test for
precisely this issue, and etc/compilation.txt for a non-ERT version.