zorba-coders team mailing list archive

[Chris Hillery <866932@xxxxxxxxxxxxxxxxxx>]

** Description changed:

- The following test have started to fail because of the famous string
- pool:
+ There have been a number of bug reports over the years where Zorba
+ crashes on exit with a message like the following:
  
- test/unit/debugger/test_debugger_serialization
- test/unit/external_function
- test/unit/no_folding
- test/unit/plan_serializer
+ i = 94 String http://www.zorba-xquery.com/options/versioning is still in the pool
+ Zorba Internal Fatal Error in home/ceej/zo/src/src/store/naive/string_pool.cpp:42:
+ "count == 0": condition failed: 1 strings remain in the string pool
  
- Zorba Internal Fatal Error in C:\Program Files\ICW\home\zorbatst\tester\nightlies\debug\src\store\naive\string_pool.cpp:41
- X strings remain in the string pool
+ As far as I know, we have never completely figured out what causes these
+ errors nor how to correct them.
  
- More details on:
- http://zorbatest.lambda.nu:8080/cdash/viewTest.php?onlyfailed&buildid=18225
+ Gabriel initially reported this bug 866932 listing several test cases
+ which exhibited this symptom, but those were subsequently cleared up by
+ some other change without fixing the underlying problem.
+ 
+ I have just now found at least one reliable way to reproduce this result
+ (see comment #3), so I have co-opted this bug report.

-- 
You received this bug notification because you are a member of Zorba
Coders, which is the registrant for Zorba.
https://bugs.launchpad.net/bugs/866932

Title:
  "X strings remain in the string pool" failing tests

Status in Zorba - The XQuery Processor:
  New

Bug description:
  There have been a number of bug reports over the years where Zorba
  crashes on exit with a message like the following:

  i = 94 String http://www.zorba-xquery.com/options/versioning is still in the pool
  Zorba Internal Fatal Error in home/ceej/zo/src/src/store/naive/string_pool.cpp:42:
  "count == 0": condition failed: 1 strings remain in the string pool

  As far as I know, we have never completely figured out what causes
  these errors nor how to correct them.

  Gabriel initially reported this bug 866932 listing several test cases
  which exhibited this symptom, but those were subsequently cleared up
  by some other change without fixing the underlying problem.

  I have just now found at least one reliable way to reproduce this
  result (see comment #3), so I have co-opted this bug report.

To manage notifications about this bug go to:
https://bugs.launchpad.net/zorba/+bug/866932/+subscriptions