At 06:05 AM 11/19/01 -0800, Jean-Paul Miéville wrote:
>I don't understand the difference between these two functions: find and
>finddepth in module File::Find. I don't understand the explanation in
>perldoc. My experience shows that finddepth is about 50% faster than find.
>Why ?
A depth-first search means that all subdirectories are visited before their
parent directories. There are advantages to this ordering for certain
tasks. Given this hierarchy:
$ ls -lR top
top:
total 4
drwxrwxr-x 3 peter peter 4096 Nov 19 10:50 second
-rw-rw-r-- 1 peter peter 0 Nov 19 10:50 topfile
top/second:
total 4
-rw-rw-r-- 1 peter peter 0 Nov 19 10:50 secondfile
drwxrwxr-x 2 peter peter 4096 Nov 19 10:50 third
top/second/third:
total 0
-rw-rw-r-- 1 peter peter 0 Nov 19 10:50 thirdfile
Then here is the difference between the orderings: notice that while the
subdirectories are processed before their parents with finddepth, the
ordinary files are processed before the subdirectories. It's never been a
problem for me but somehow it seems wrong *shrug*:
$ perl -MFile::Find -le 'find(sub {print $File::Find::name}, "top")'
top
top/topfile
top/second
top/second/secondfile
top/second/third
top/second/third/thirdfile
$ perl -MFile::Find -le 'finddepth(sub {print $File::Find::name}, "top")'
top/topfile
top/second/secondfile
top/second/third/thirdfile
top/second/third
top/second
top
I don't honestly know why finddepth should be faster.
--
Peter Scott
Pacific Systems Design Technologies
http://www.perldebugged.com
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
