Rmail, Babyl, mbox, emacs 23

Rmail, Babyl, mbox, emacs 23

[Lewis Perin]

My ISP just upgraded emacs to 23.1.1.  I was alarmed when it couldn't
parse my 22-level Babyl rmail file, but I soon learned that running M-x
unrmail would fix that.

Now, though, I face the problem of what happens when I run M-x
rmail-input against an email archive (also Babyl) created under an
earlier version of emacs.  23.1.1's Rmail doesn't parse the archive any
better than it parsed the Babyl version of ~/RMAIL; it gives up with the
error message "Removing old name: no such file or directory,
/tmp/rmail17713j4x".

When I run unrmail against an archive, that doesn't work either; it
blows up with "Symbol's value as variable is void: rmail-highlight-face".

Has anyone got a suggestion for overcoming this?

/Lew
---
Lew Perin / perin@acm.org
http://www.panix.com/~perin/babelcarp.html

Re: Rmail, Babyl, mbox, emacs 23

[Glenn Morris]

Lewis Perin wrote:

> My ISP just upgraded emacs to 23.1.1.  I was alarmed when it couldn't
> parse my 22-level Babyl rmail file,

This sounds like a bug that should be reported, with as much detail as
possible (starting from emacs -Q) to help it get fixed.

Can you try the 23.1.94 pretest from

http://alpha.gnu.org/gnu/emacs/pretest/

to see if it does any better?

> but I soon learned that running M-x unrmail would fix that.

I'm a bit confused, because internally Emacs uses unrmail to convert
Babyl files to mbox when needed. So I don't see why it would work when
you call it by hand, but not when Emacs calls it.

> Now, though, I face the problem of what happens when I run M-x
> rmail-input against an email archive (also Babyl) created under an
> earlier version of emacs.  23.1.1's Rmail doesn't parse the archive any
> better than it parsed the Babyl version of ~/RMAIL; it gives up with the
> error message "Removing old name: no such file or directory,
> /tmp/rmail17713j4x".

Again, this is a bug.

> When I run unrmail against an archive, that doesn't work either; it
> blows up with "Symbol's value as variable is void: rmail-highlight-face".

This reveals another bug (sigh; now fixed), but if you start emacs as
`emacs -Q' it should not affect you.

Re: Rmail, Babyl, mbox, emacs 23

[Uday S Reddy]

Lewis Perin wrote:

> 
> Has anyone got a suggestion for overcoming this?

You could use VM, which handles mbox folders as well as Babyl folders reliably and has been doing so for 20+ years!

   http://launchpad.net/vm

Cheers,
Uday Reddy

Re: Rmail, Babyl, mbox, emacs 23

[Lewis Perin]

Glenn Morris <rgm+news@stanford.edu> writes:

> Lewis Perin wrote:
>
>> My ISP just upgraded emacs to 23.1.1.  I was alarmed when it couldn't
>> parse my 22-level Babyl rmail file,
>
> This sounds like a bug that should be reported, with as much detail as
> possible (starting from emacs -Q) to help it get fixed.

Well, that turned out to be a helpful suggestion!  (Should've thought of
that myself.)  The problem vanished when I started emacs without a
decade or so of accumluated cruft in my .emacs.  Now I have to figure
out what gets in the way of the new version of Rmail.  I wonder if it
could be this:

(custom-set-variables
  ;; [...]
 '(rmail-ignored-headers
 "^via:\\|^mail-from:\\|^origin:\\|^references:\\|^status:\\|^received:\\|^x400-originator:\\|^x400-recipients:\\|^x400-received:\\|^x400-mts-identifier:\\|^x400-content-type:\\|^\\(resent-\\|\\)message-id:\\|^summary-line:\\|^resent-date:\\|^nntp-posting-host:\\|^path:\\|^x-char.*:\\|^x-face:\\|^x-mailer:\\|^delivered-to:\\|^lines:\\|^mime-version:\\|^x-coding-system:\\|^return-path:\\|^errors-to:\\|^return-receipt-to:\\|^x-sign:\\|^x-beenthere:\\|^x-mailman-version:\\|^precedence:\\|^list-help:\\|^list-post:\\|^list-subscribe:\\|^list-id:\\|^list-unsubscribe:\\|^list-archive:\\|^content-length:\\|^x-attribution:\\|^x-disclaimer:\\|^x-trace:\\|^x-complaints-to:\\|^nntp-posting-date:\\|^user-agent:"))

To be honest, I can't recall the purpose of that code.