http://nagoya.apache.org/bugzilla/show_bug.cgi?id=1030 *** shadow/1030 Mon Mar 19 14:05:51 2001 --- shadow/1030.tmp.6126 Mon Apr 23 11:38:30 2001 *************** *** 4,10 **** | Bug #: 1030 Product: Regexp | | Status: NEW Version: unspecified | | Resolution: Platform: PC | ! | Severity: Normal OS/Version: | | Priority: Medium Component: Other | +----------------------------------------------------------------------------+ | Assigned To: [EMAIL PROTECTED] | --- 4,10 ---- | Bug #: 1030 Product: Regexp | | Status: NEW Version: unspecified | | Resolution: Platform: PC | ! | Severity: Normal OS/Version: All | | Priority: Medium Component: Other | +----------------------------------------------------------------------------+ | Assigned To: [EMAIL PROTECTED] | *************** *** 22,25 **** I tested the same string using the ORO and GNU libraries - both agreed that this string should not match this pattern. ! I ran all examples under Windows 2000 using JDK 1.3. --- 22,40 ---- I tested the same string using the ORO and GNU libraries - both agreed that this string should not match this pattern. ! I ran all examples under Windows 2000 using JDK 1.3. ! ! ------- Additional Comments From [EMAIL PROTECTED] 2001-04-23 11:38 ------- ! I have found some associated problems with with the {n,m} notation. It doesn't ! appear to work correctly with 0 as a lower bound, and the upper bound allows one ! higher than it should. ! ! For the pattern: a(b{0,1})a ! ! Test String Matches Should Match ! aa false true ! aba true true ! abba true false ! abbba false false ! ! I got these results with Sun's Linux 1.3 SDK, and Win NT 1.3 SDK.
