You are viewing a plain text version of this content. The canonical link for it is here.

Posted to notifications@ant.apache.org by bu...@apache.org on 2008/09/01 10:38:01 UTC

DO NOT REPLY [Bug 45665] DirectoryScanner Performance

https://issues.apache.org/bugzilla/show_bug.cgi?id=45665





--- Comment #2 from Gilles Scokart <gs...@gmail.com>  2008-09-01 01:38:01 PST ---
Yes, the problem described is fixed.  The patterns are now parsed only once per
scanning.

I have also seen some further possible enhancements, but I'm not sure they will
bring a noticeable enhancement :
1. When multiple include/exclude patterns are used, the filename is also
tokenized multiple times (once per pattern)
2. The tokenizePathAsArray implementation loop twice on the String (pattern or
filename) using charAt method.  Maybe using indexOf, and/or using a Tokenizer
class instead of an array could provide faster results.

But the major thing is now fixed.  If the 2 previous point give noticebale
enhancement, then I will re-open this issue.

I gain +/- 2% of my total build time which is at 80% an ivy build (not using
the directory Scanner).  A build that do a lot of directory scanning could gain
up to 10% of processing time.


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.