Commits on Oct 28, 2008
  1. Added slabReassign to the API.

    committed Oct 25, 2008
  2. Set a value before attempting to look at the slabs stats.

    The slab stats test will return nothing if there have never been any
    values added (thus any slabs created).
    committed Oct 28, 2008
Commits on Oct 25, 2008
  1. Support an argument to stats.

    committed Oct 25, 2008
Commits on Oct 19, 2008
  1. Got rid of workarounds for binary stats problems.

    In the latest code, Toru made sure all result packets contain the
    requested opaque.  No magic and no 0.
    committed Oct 19, 2008
  2. Asynchronous incr/decr with no default.

    Theoretically, this can work with default values as well, but only in
    the binary protocol.  Getting this to work with the text protocol
    would likely be quite painful.
    nbrachet committed with Oct 18, 2008
  3. Update license copyright.

    committed Oct 19, 2008
Commits on Oct 18, 2008
  1. Miscellaneous test code cleanups.

    I've got some tool that complains about various things it thinks I
    might be doing wrong.
    committed Oct 18, 2008
  2. Whalin V1 compatible transcoder.

    magictoken committed with Oct 18, 2008
  3. Updated copyright info.

    committed Oct 18, 2008
Commits on Oct 17, 2008
  1. fixes for Whalin compatible transcoder.

    corrected Boolean encoding and added Character support
    magictoken committed with Oct 17, 2008
Commits on Oct 15, 2008
  1. Binary stat support.

    committed Oct 15, 2008
  2. Remove support for hold values on deletes.

    This functionality was removed after we all agreed it was confusing as
    1157f3c5ce25918558781bd2207b6b6de702dd17 in memcached by Trond Norbye
    (merged in by Toru).  Mon Jul 28 17:55:41 2008 +0900
    That ID will probably be killed off by a rebase.
    committed Oct 15, 2008
Commits on Oct 4, 2008
Commits on Oct 3, 2008
  1. if the memcache server disconnects, try to notice.

    Robey Pointer committed with Jul 21, 2008
  2. If we attempt to read when we don't have anything to read, reconnect.

    This seems to happen in some strange server-error cases where the
    server short-circuits the connection.  This situation has shown itself
    when we've sent something too large into the server.  Precautions have
    been taken to prevent that exact thing from happening again, but this
    is a better way to handle an unexpected state.
    I'm not sure how we end up still having data to receive when we've
    already thrown away the ops, but tearing down the connection is a
    clean way to recover from a bad state.
    committed Oct 3, 2008
  3. Throw an exception if we attempt to store data we know is too big.

    This will avoid a few problems people have encountered.  I implemented
    it with a somewhat early IllegalArgumentException because it was the
    surest way to cover every case given compression and several paths in
    to sets.
    committed Oct 3, 2008
  4. Minor warning fix.

    committed Oct 3, 2008
Commits on Oct 2, 2008
Commits on Sep 21, 2008
  1. Default for incr should be a long.

    nbrachet committed with Sep 21, 2008
  2. Extract a MemcachedClient interface.

    An interface enables mocking for any users of the MemcachedClient.
    nelz9999 committed with Sep 11, 2008
Commits on Aug 29, 2008
  1. Decode async bulk gets outside of the IO thread.

    The same work still needs to happen for one-off gets, but this should
    help allow the IO thread to focus on IO while consumer threads
    distribute the more CPU-intensive work.
    kreide committed with Aug 29, 2008
Commits on Jun 6, 2008
  1. More consistent hash proving.

    nelz9999 committed with Jun 6, 2008
Commits on May 27, 2008
  1. Killed off a slightly elusive bug that was causing sporatic cancellat…

    Whenever a read returned in the binary protocol without having read
    enough bytes to fill a header packet, an NPE would be fired that would
    cause us to disconnect from the server and cancel all in-flight
    This happened occasionally in one of my tests and was rather a pain to
    track down.  It likely never affected anyone since I doubt anyone is
    actually using the binary protocol anywhere today.
    committed May 27, 2008
Commits on May 14, 2008
  1. Allow the input, op read and op write queues to be individually speci…

    By default, the read queue is 10% larger than the input queue.
    With a sufficiently large op read queue, it's possible to never
    internally overflow, but correct values are likely
    committed May 14, 2008