Re: small code smell

On Mon, Sep 01, 2003 at 12:04:46PM +0200, Bram Moolenaar wrote:
> 
> Tim Hemel wrote:
> 
> > Here's one that can perhaps be made a lot simpler, by borrowing someone else's
> > code:
> > 
> > shorten_name in Util.py
> > 
> > A similar function is described in the Python Cookbook:
> > 
> > http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/208993
> > 
> > periperi(116):~/programs> python relpath.py
> > from /a/b/c/d to /a/b/c1/d1  ->  ../../c1/d1
> > from /foo/bdir/foo.o to /foo/test  ->  ../../test
> > from /foo/test to /foo/bdir/foo.o  ->  ../bdir/foo.o
> > 
> > The functions do not work with . and .. in the filenames, but that can be
> > solved by always normalizing the pathnames.
> 
> This doesn't look simpler at all.  Looking at the code I can guess it's
> a lot slower, because commonpath() calls itself recursively for every
> character.  This smells like a classroom example...

It looked a lot easier to understand, as is usually the case with classroom
examples.

> Path handling needs to be fast, because it's done very often.  And it
> also needs to take care of forward- vs back-slashes.
> 
> Reminds me to do some profiling one of these days...


Tim
-- 
Zen Microsystems: we're the om in .commmmmmmmm...


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf