Perl get suffix ?

i have am trying to get the suffix from a directory (.html).

I tried using fileparse but it is not consistant without a suffix list:

#!/iw-home/iw-perl/bin/iwperl
use File::Basename;

$src="/tmp/test/AA3.FINAL.TXT";
($base, $dir, $suffix) = fileparse($src);
print "Base $base dir $dir suffix $suffix\n";

yields:

Base AA3.FINAL.TXT dir /tmp/test/ suffix

Is there a trick to getting this to work or shoudl I just look for the last period ?

Andy

Find more posts tagged with

Comments

JonathonG

Is there a reason you don't want to just use a regex? It'd be more efficient (not loading in a module, etc.)

Plus, I have no idea how to fix your fileparse problem Smiley Happy

but here's the regex:

$src=~ m|^(.*/)(.*?)\.(.*?)$|;
$base = $1;
$dir = $2;
$suffix = $3;

Jonathon
Interwoven Architecture Consultant

JonathonG

Hmm...should've tested that...try replacing the other regex with this one

$src=~ m|^(.*/)(.*?)\.([^\.]+)$|

Jonathon
Interwoven Architecture Consultant

Migrateduser

I believe Adam Stollers perl benchmark paper (http://devnet.interwoven.com/site.fcgi/techlib/att/048908/1301/Perl_Efficiency.pdf) showed that using some basename functions are rather slow. I would just use a regex. Grab everyting after the last '.'

--
Jed Michnowicz
Interwoven Technical Consultant

Migrateduser

If all you want is the suffix, then this should do the trick:

s/^.*\.(.*)$/$1/;

Dave Smith
Sr. Software Engineer
Nike, Inc.
(503) 671-4238
DavidH.Smith@nike.com

jbonifaci

I would go with JG's way, but if you want to go smitty's way, change his regex to this:

(my $suffix = $src) =~ s/^.*\.([^\.]+)$/$1/;

Hope this helps,
Jeff Bonifaci

laj1


$src="/this/dir/name.tetx.ext";
$match = $src=~ m/\.([^.]+)$/o;
print "match = $match ---- $1\n";

Use the above regex. if you use the m// operator, use $1 to get the matched value inside the parens, and/or
$& to get the whole of the match (which includes the dot).
if you use the s/// operator as suggested by others, you'll have the match in your source variable,
but the source variable will have been changed, so either be sure that you can lose the contents of
the source variable, or copy it, and s/// the copy.

It is my contention, though I have no statistics to back it up, that anchoring this regex to the front and back will make more work for
the Rregex engine, and would seem to me to be inefficient, due to the fact that the text of interest here is not merely defined by its proximity to the end of string anchor, but actually abuts the anchor. Whereas the front of the string is nebulous and random as far as this particular pattern match is concerned. So anchoring to the front [using the caret ^], caring about the front of the string [ .* variously in the preceding posts], and recording the front of the string as we match it [ using (.*) variously] all waste energy.

It would not surprise me is perl were to start at the back of the string and work its match from back to front, using my regex, but once anchored to the front of the string, as in other posts, it must start at the front and read the whole string. If you have long strings, and lots of them, that's a lot of extra work.

of course, after saying all that, since this is Perl, there's more than one way to do it.

You could do this too:


$extenstion =        substr($src, rindex($src, '.') )

which finds the last dot (first dot from the back) and then returns the substring from the dot, to the end of the string. if you don't want the dot included add 1 to the rindex() call to start the substring on the first character after the dot.

Len

Len Jaffe
My Heart Is A Flower
Update your DevNet profile - let us know who you are!