[status] belated weekly report

From: Ryan Lee <ryanlee_at_w3.org>
Date: Tue, 02 May 2006 11:52:53 -0600

Last week:

  - Found the deadlock problem - it seems the startTransaction() method
    initiaties a transaction but doesn't turn off auto-commit. Since
    updates on a repository will also initiate a transaction,
    startTransaction() appears to introduce the possibility of deadlock.
    The beginning of an atomic transaction should start with turning off
    auto-commit and end with a commit() (or rollback()), turning auto-
    commit back on afterwards

    (Corrections to my understanding of Sesame transactions are quite
    welcome)

  - Helped merge Stefano and David's performance tuning branch back into
    Longwell trunk

  - Addressed some of the minor requested enhancements to the library
    demo

  - Working on ISWC06 paper

This week:

  - Continue working on ISWC06 paper

  - Continue enhancing library demo

  - Dig up repository calls that are generating the messages:
    [WARNING] Releasing active lock due to object destruction

    (To avoid, explicitly instantiate a CloseableIterator and close()
    it when finished; most repository read methods will return a
    CloseableIterator).

-- 
Ryan Lee                 ryanlee_at_w3.org
W3C Research Engineer    +1.617.253.5327
http://simile.mit.edu/
Received on Tue May 02 2006 - 17:52:32 EDT

This archive was generated by hypermail 2.3.0 : Thu Aug 09 2012 - 16:39:18 EDT