jakarta-regexp-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Patrik Beno <patrik.b...@qbizm.com>
Subject Re: Speed of A{m,n}
Date Tue, 26 Jun 2001 21:50:27 GMT
Of course, I've written a few pretty stupid lines:

On Tuesday, 26 June 2001 23:05, Patrik Beno wrote:

> I found out that compiling following regexp:

>   ^.{1,32}$

> is very slow. I've thought I have a bug in my code, later on I've
> thought regexp 1.2 is buggy and is cycling somewhere in endless loop.


Bla-bla.

Compilation of the expression is - of course - quick.

What is slow is evaluation in specific situation:
if string's length is exceeding upper limit (32 or whatever above ~20).

Try demo (org.apache.regexp.REDemo). Enter the regexp
and then string to match: after typing last character that makes
string non-matching, re.match() lasts tooooo long.

Sorry for mystification.

+---                                                              ---+
patrik.beno@{qbizm.com|inmail.sk}                         ICQ#98931498
+-                                                                  -+
Patrik Beno                                   Qbizm Technologies, Inc.
analyst / developer                           ... the art of Internet.
+--------------------------------------------------------------------+
Kralovopolska 139                                tel: +420 5 4124 2414
601 12 Brno, CZ        http://www.qbizm.com      fax: +420 5 4121 2696


Mime
View raw message