Re: image-load-path/load-path function

[Bill Wohler <wohler@newt.com>]

Reiner Steib <reinersteib+gmane@imap.cc> wrote:

> On Fri, Mar 03 2006, Bill Wohler wrote:
> 
> > Here is the same thing as my last patch, [...]
> 
> (BTW, if it's already in Emacs CVS, you don't need to include it
> here.)

;-).

> > I'm definitely now heavily leaning towards the name
> > image-load-path-for-library. Does it work for you too? The "relative"
> > variants didn't work for me since you could also pull the directories
> > off of load-path or image-load-path.
> 
> Okay with me.  Although I'd slightly prefer to emphasize "relative" a
> little bit.  But maybe adding it to the doc string is sufficient:
> 
> "Return a suitable search path for images relative to LIBRARY.
> [...]"

OK.

> > In an Emacs 22 environment, given the name, you'd expect a modified
> > image-load-path, but you get a modified load-path. If the optional
> > argument path isn't present, I'd suggest returning a modified
> > image-load-path if it exists; else a modified load-path. What do you
> > think?
> 
> What if a user shares configuration files between Emacs 21 and 22 and
> sets `image-load-path' unconditionally (without a boundp test; I
> sometimes don't add test such if I assume that it will be just ignored
> in older Emacsen)?  The you'll get a bogus `load-path' like...

Good thinking.

> I think #2 and #3 both are a little problematic.

I'm convinced. I'll avoid them and stick with #1.

> But maybe we can suggest an example which avoids calling
> `image-load-path-for-library' twice:
> 
> (let* ((load-path (image-load-path-for-library "mh-e" "mh-logo.xpm"))
>        (image-load-path (cons (car load-path) image-load-path)))
>   ;; do something.  Test:
>   image-load-path)

That assumes we guarantee that the image-directory is always first. That
appears to be the case now. I think I'd be a little leery about
committing to this in case we find out later that we needed to preserve
the existing order in the path if the image directory was already in the
path. But if we want to commit to this, then I'll add it to the
documentation and use this example, otherwise...

> > Here is an example that uses a common idiom to provide
> > compatibility with versions of Emacs that lack the variable
> > `image-load-path':
> >
> >   (let ((load-path
> >          (image-load-path-for-library \"mh-e\" \"mh-logo.xpm\" 'load-path))
> 
> I'd omit 'load-path as it is optional.

...I'll do that. Let me know which you prefer. Maybe someone else has a
gut feel about this too. My gut tells me to avoid committing more to the
API than we have to.

Thanks for the great feedback.

-- 
Bill Wohler <wohler@newt.com>  http://www.newt.com/wohler/  GnuPG ID:610BD9AD
Maintainer of comp.mail.mh FAQ and MH-E. Vote Libertarian!
If you're passed on the right, you're in the wrong lane.


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642