>>>> "David" == David Masterson
I think these are all good points you're bringing up.
xemacs-beta: Who are the experts in this area?
Documentation on recover-session is indeed lacking.
Here's some evidence:
cd c:\Program Files\XEmacs\XEmacs-21.5-b2\info\
egrep -ni recover-session *.info* NUL
Compilation started at Tue Jul 31 23:38:57 2001 +0200 (W. Europe Daylight Time)
lispref.info-25:424: `recover-session' command uses these files.
xemacs-faq.info-2:360: You can use the command `M-x recover-session' after a crash
grep finished (3 matches found) at Tue Jul 31 23:39:02
>>>>> "APA" == Adrian Aichner <Adrian.Aichner(a)t-online.de>
>>>> "David" == David Masterson
>>>>> "APA" == Adrian
Aichner <Adrian.Aichner(a)t-online.de> writes:
>>>>> "David" == David Masterson <dmaster(a)synopsys.com>
David> (X)Emacs creates ".saves*" to record some information (I'm
David> sure what). The difference is that Emacs creates a name that
David> ends in "~" while XEmacs does not. The "~" in the name
David> it easier to collect and delete these files using
David> dired-flag-backup-files. Could XEmacs change to use this
APA> M-x recover-session is the proper interface to these autosave
APA> files. This will bring up a dired buffer in which you can mark,
APA> visit, delete the session files.
David> Hmmm. The documentation in that buffer could be written a
David> little better. The text mentions using 'd' to mark a file
David> for deletion, but it doesn't tell you about expunge.
APA> Any suggestions, or patches even?
David> Actually, I don't see a reason for bringing up a dired buffer for
David> recover-session anyway. Wouldn't it be enough to just ask for a
David> "Session file?" in the minibuffer and allow filename completion to
David> the rest. This special dired buffer doesn't seem to be doing anything
David> special that you need to confuse people with having yet another
David> methodology for doing file selection (minibuffer query, X popup,
David> Dired, ???).
David> Because the buffer is accessed via a special command
David> (recover-session) and because the format of the buffer is
David> slightly different, its not obvious that the buffer is a dired
David> buffer or that all the same rules apply.
APA> Doesn't it say Dired in the modeline?
David> Sure, but the differences were enough to make me think there was
David> probably something else going on here. After all, dired buffers don't
David> have "^C^C" commands for selecting a file to pass to some other
David> operation. Since the buffer went out of its way to put at the top
David> what should be done, it's easy to think that that's the only things
David> that can be done.
APA> It's probably a feature they can't inadvertently be deleted
APA> with the common idiomatic sequence of
APA> ~ runs the command dired-flag-backup-files
APA> x runs the command dired-expunge-deletions
David> Perhaps. That's certainly what I do with GNU Emacs. Since
David> I never really understood what they were and since they
David> were always showing up, I just wanted to get rid of them.
David> BTW, I have never seen any info documentation of
David> recover-session (although there is docs on recover-file).
David> Am I missing something? (This would explain the previous
APA> I get this with
APA> C-h a recover-session RET
David> Obvious (as is describe-function), but its not really an explanation
David> of recover-session in the same sense as the Auto Save section of the
David> XEmacs Info docs for recover-file.
David> * what is a session?
David> * why would I want to recover it?
David> * what will it recover when I recover it?
David> * how does it differ from (a set of) auto save files?
David> (I can guess some, but I don't really know the answers to these, so
David> don't ask me to write the info docs on it. ;-)
David> David Masterson
David> Sr. R&D Engineer
David> Synopsys, Inc.