leim/quail and mule doble names of pkg

leim/quail and mule doble names of pkg

[Uwe Brauer]

Hello

I  realized that in Xemacs there are files in the lisp packages
whose names coincide. People on xemacs beta think that this
is a bad idea. 

Now I looked into GNU emacs 23.3 and the situation is the
same.


For example 

hebrew.el 

Is in 


leim/quail/hebrew.el
for the input method

But also in 
lisp/language/hebrew.el

for the coding


Is this wise?

Uwe Brauer 

Re: leim/quail and mule doble names of pkg

[Andreas Schwab]

Uwe Brauer <oub@mat.ucm.es> writes:

> leim/quail/hebrew.el
> for the input method
>
> But also in 
> lisp/language/hebrew.el

These files are unrelated.  The leim/quail directory is never part of
load-path.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: leim/quail and mule doble names of pkg

[Stefan Monnier]

> For example
> hebrew.el
> Is in
> leim/quail/hebrew.el for the input method
> But also in
> lisp/language/hebrew.el for the coding
> Is this wise?

I'm not sure if "wise" is the right question, but yes, there are some
problems in this area.  With the CEDET integration we have many more
such cases.  IIRC XEmacs has other cases with its packages's
"autoload" files.

The way those "same name files" are distinguished depends on the
particular case, and the potential problems are mostly avoided in
ad-hoc ways.

For CEDET, we basically consider the file's name to be
(file-relative-name buffer-file-name load-path), which does make those
file names unique again.  The same has been used for term/*.el files in
the past.  But this is not supported consistently (e.g. eval-after-load
does not understand this convention, IIRC).

For other files, we just try to treat those files specially (they don't
`provide' their name and aren't loaded via load-path), which comes with
its own set of problems and limitations.

I'd be happy to see someone tackle this issue and try to come up with
a set of solutions which we can then try to support consistently.


        Stefan