Re: Compile issues

[Stefan Monnier]

>> Please send it to me and I'll install it.
> Perhaps there was some reason it wasn't installed?

Of course, but it's most likely "lack of time" or some such.

>> Are these problems specific to the new compile.el or
>> where they somehow absent in the old compile.el ?
> The i18n issues are old, but it seems the right time to address i18n so
> that you don't have to make potentially-incompatible changes in future.

I'll trust you on that one.  But to tell you the truth, I have my hands full
with getting the new compile.el to works as well as the old one and the
localization seems like an orthogonal problem.  Patches are welcome, but
I won't write them myself.

> I realize that the current set of patterns doesn't look
> amenable to a limited parameterization like it used to.

I don't understand what you mean here.

> Issues with the new classification of the errors and redundancy are
> new, but less important.

Yes, that's a added problem indeed.  Hopefully it is "safe" in that
a non-recognized case is treated as `error' rather than `warning' or `info'.

> C-c C-k compiles the buffer.  This isn't relevant --
> `compilation-minor-mode' just doesn't work.  I assume the reason is
> that font-locking just isn't set up:

> (save-window-excursion
>   (save-excursion
>     (global-font-lock-mode 1)
>     (comint-run "bash")
>     (compilation-minor-mode)
>     font-lock-keywords))
>   => (t)

Ah, indeed, that's a ...hmmm... problem.  Thank you for tracking it down.
I'll look for a fix ASAP.

> I think that's an old version.  Mine came from the current OTP
> distribution, probably via Debian.  (There's stuff in it which will
> make you wince...)

There's preciously little elisp code that doesn't make me wince.  ;-)


        Stefan