Re: A couple of locate-user-emacs-file questions

[Stefan Monnier]

> is mildly incompatible, in the sense that some external tool *could*
> potentially depend on the session file being in $HOME; but that seems
> unlikely. Does anyone see a problem with it?

I don't.

> And speaking of `locate-user-emacs-file', there are a couple of cases
> (in cmuscheme.el and shell.el, both related to passing a filename to
> comint only if the file does really exist) that could benefit from it,
> if there was a way to tell it to check for OLDNAME and NEWNAME, but
> return nil (and not create `user-emacs-directory') if neither file
> exists. Something like this (diff -b to ignore irrelevant indentation
> changes):

I think it should still not return nil but the "preferred name".

> On one hand, seems like a bit overengineered; on the other hand, a
> function called *locate*-user-emacs-file should be able to tell
> whether it did, in fact, locate it without causing side effects...

I think the problem is that the function should have come with
a `and-create-directory' argument instead, so the directory is only
created when explicitly requested.


        Stefan