Removing list identifiers in rmail (was: Changes to emacs/lisp/mail/rmail.el)

Removing list identifiers in rmail (was: Changes to emacs/lisp/mail/rmail.el)

[Reiner Steib]

Alex Schroeder on emacs-diffs:

> --- emacs/lisp/mail/rmail.el:1.417	Sun Jan  8 15:18:55 2006
> +++ emacs/lisp/mail/rmail.el	Sun Jan  8 23:47:07 2006
> @@ -447,8 +447,10 @@
>    "String to prepend to Subject line when replying to a message.")
>  
>  ;; Some mailers use "Re(2):" or "Re^2:" or "Re: Re:" or "Re[2]:".
> -;; This pattern should catch all the common variants.
> -(defvar rmail-reply-regexp "\\`\\(Re\\(([0-9]+)\\|\\[[0-9]+\\]\\|\\^[0-9]+\\)?: *\\)*"
> +;; This pattern should catch all the common variants.  The pattern
> +;; also ignores mailing list identifiers sometimes added in square
> +;; brackets at the beginning of subject lines.
> +(defvar rmail-reply-regexp "\\`\\(\\[.+?\\] \\)?\\(Re\\(([0-9]+)\\|\\[[0-9]+\\]\\|\\^[0-9]+\\)?: *\\)*"
>    "Regexp to delete from Subject line before inserting `rmail-reply-prefix'.")

If this also removes tags in square brackets (like in "[PCL-CVS] foo",
"[patch] bar", ...), it is not a good idea, IMHO.

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Removing list identifiers in rmail (was: Changes to emacs/lisp/mail/rmail.el)

[Alex Schroeder]

> If this also removes tags in square brackets (like in "[PCL-CVS] foo",
> "[patch] bar", ...), it is not a good idea, IMHO.

My assumption was that if you are navigating through your mails,
however, "foo", "Re: foo", "[patch] foo", and "[patch] Re: foo" are
all the same subject.  I also felt that if your mailing list engine
adds the tag, then it will add it again when it resends your message
to people on the list, so no big deal.

My change might be a mistake, if you think that square brackets at the
beginning of a subject are sometimes not such mailing list
identifiers, and therefore removing them in a reply is wrong.

What do you think?  

Alex
-- 
http://www.emacswiki.org/alex/

Re: Removing list identifiers in rmail

[Stefan Monnier]

> I also felt that if your mailing list engine adds the tag, then it will
> add it again when it resends your message to people on the list, so no
> big deal.

IIRC there are cases where it doesn't do the right thing:

- you receive "Re: [FOOLIST] toto"
- you rewrite it as "toto"
- the reply you send says "Re: toto"
- the mailing-list rewrites it as "[FOOLIST] Re: toto"
- people who reply to it with most MUAs send "Re: [FOOLIST] Re: toto"


-- Stefan

Re: Removing list identifiers in rmail

[Alex Schroeder]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> IIRC there are cases where it doesn't do the right thing:
>
> - you receive "Re: [FOOLIST] toto"
> - you rewrite it as "toto"

Actually my change would not rewrite this form.  It would only rewrite
"[FOOLIST] Re: toto" -- but maybe it should!  :)

> - the reply you send says "Re: toto"
> - the mailing-list rewrites it as "[FOOLIST] Re: toto"
> - people who reply to it with most MUAs send "Re: [FOOLIST] Re: toto"

In this case it would seem that you're saying the mailing-list subject
munging is violating the expectations of "most MUAs".  Since I don't
know many MUAs, I have no opinion on this.

If people feel that the simple change to the regexp was a mistake, I
will revert the patch.

The alternative of introducing another regular expression or more code
just to handle jumping to the next or previous message with the "same"
subject, or collecting all the messages with the "same" subject seems
is too complicated for my taste.

Alex.
-- 
http://www.emacswiki.org/alex/

Re: Removing list identifiers in rmail (was: Changes to emacs/lisp/mail/rmail.el)

[Henrik Enberg]

> > --- emacs/lisp/mail/rmail.el:1.417	Sun Jan  8 15:18:55 2006
> > +++ emacs/lisp/mail/rmail.el	Sun Jan  8 23:47:07 2006
> > @@ -447,8 +447,10 @@
> >    "String to prepend to Subject line when replying to a message.")
> >  
> >  ;; Some mailers use "Re(2):" or "Re^2:" or "Re: Re:" or "Re[2]:".
> > -;; This pattern should catch all the common variants.
> > -(defvar rmail-reply-regexp "\\`\\(Re\\(([0-9]+)\\|\\[[0-9]+\\]\\|\\^[0-9]+\\)?: *\\)*"
> > +;; This pattern should catch all the common variants.  The pattern
> > +;; also ignores mailing list identifiers sometimes added in square
> > +;; brackets at the beginning of subject lines.
> > +(defvar rmail-reply-regexp "\\`\\(\\[.+?\\] \\)?\\(Re\\(([0-9]+)\\|\\[[0-9]+\\]\\|\\^[0-9]+\\)?: *\\)*"
> >    "Regexp to delete from Subject line before inserting `rmail-reply-prefix'.")
> 
> If this also removes tags in square brackets (like in "[PCL-CVS] foo",
> "[patch] bar", ...), it is not a good idea, IMHO.

Perhaps it should only remove text matched by members in some user
supplied list.  I vaguely remember Gnus having such a feature.