[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[emacs-wiki-discuss] Re: Merge with planner-muse--mwolson
From: |
drkm |
Subject: |
[emacs-wiki-discuss] Re: Merge with planner-muse--mwolson |
Date: |
Tue, 26 Jul 2005 04:13:31 +0200 |
User-agent: |
Gnus/5.110003 (No Gnus v0.3) Emacs/22.0.50 (windows-nt) |
Michael Olson writes:
> It's supposed to print the page name (here, it's *emacs-wiki-temp*,
> which is strange ... it should prefer emacs-wiki-current-file to
> buffer-file-name). At this point, the best thing to do is to set
> debug-on-error to t and try to get a backtrace. If debug-on-error is
> t, it won't try to catch the error and display that unhelpful warning.
It was the deal, if I remember well. To not disturb the
publication by just displaying a warning, while to permit to get
a backtrace by setting 'debug-on-error'.
Personnally, I didn't like the fact that an error would be
silently (more or less silently) ignored. But I agree now that
it could be fine while publishing. A good compromise could be
the following, IMHO.
1/ Keep the ability to get a backtrace. It's fundamental.
And via 'debug-on-error' is the canonical way. Fine.
2/ Catch errors while publishing if 'debug-on-error' is nil,
and emit a warning. It's the current behavior, I think.
3/ If such an error has been catched (so at least one file
hasn't been published, right?), open a window showing at
least the list of the files not published (and the
associated errors), à la Compile Errors (enhancements such
as 'next-error' could be fine, here).
So everyone know clearly that some file was not published, in
that case.
--drkm