http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2275

*** shadow/2275 Fri Jun 22 04:00:28 2001
--- shadow/2275.tmp.26308       Fri Jun 22 20:57:35 2001
***************
*** 2,9 ****
  | Xalan doesn't fire the right templates                                     |
  +----------------------------------------------------------------------------+
  |        Bug #: 2275                        Product: XalanJ2                 |
! |       Status: ASSIGNED                    Version: 2.0.x                   |
! |   Resolution:                            Platform: PC                      |
  |     Severity: Critical                 OS/Version: Linux                   |
  |     Priority: High                      Component: Xalan                   |
  +----------------------------------------------------------------------------+
--- 2,9 ----
  | Xalan doesn't fire the right templates                                     |
  +----------------------------------------------------------------------------+
  |        Bug #: 2275                        Product: XalanJ2                 |
! |       Status: RESOLVED                    Version: 2.0.x                   |
! |   Resolution: FIXED                      Platform: PC                      |
  |     Severity: Critical                 OS/Version: Linux                   |
  |     Priority: High                      Component: Xalan                   |
  +----------------------------------------------------------------------------+
***************
*** 34,36 ****
--- 34,50 ----
  FYI: I noticed last night that you get the same bug if you use test.xsl (instead
  of testdiff.xsl), so that makes the test case marginally simpler :-)
  
+ 
+ 
+ ------- Additional Comments From [EMAIL PROTECTED]  2001-06-22 20:57 -------
+ Regression test for this bug is new position80.
+ 
+ The union pattern wasn't firing correctly.  This was caused by an 
+ optimization that occured where we tried not to clone the iterators 
+ in the UnionIterator clone function.
+ 
+ There were also some problems with getting the last node count.
+ Fixing this uncovered another problem with a previous fix to 
+ LocationPathIterator where the cachedNode list wasn't being 
+ created if it was already non-null.  This was fixed by making sure 
+ the list is reset (but not reallocated) in the WalkingIteratorSorted.

Reply via email to