delta/apache/apr.git, branch pollcb-dev

Add a test for a timeout, but with a socket recently read from inside the pollcb.

2006-03-31T05:35:52+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/branches/pollcb-dev@390332 13f79535-47bb-0310-9956-ffa450edef68

Add a new timeout testcase, which makes sure that with no sockets inside the pollset, we should trigger a timeout.

2006-03-31T05:26:50+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/branches/pollcb-dev@390330 13f79535-47bb-0310-9956-ffa450edef68

Add a simple Callback testcase, to make sure it kind-of-works. YMMV :)

2006-03-31T05:21:32+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/branches/pollcb-dev@390328 13f79535-47bb-0310-9956-ffa450edef68

Add a first revision of the Pollset Callback API.

2006-03-31T05:07:04+00:00

Currently it only has an EPoll implementation, and the most basic test case possible.


git-svn-id: https://svn.apache.org/repos/asf/apr/apr/branches/pollcb-dev@390326 13f79535-47bb-0310-9956-ffa450edef68

Create a development branch for the new apr_pollcb API.

2006-03-31T05:05:27+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/branches/pollcb-dev@390325 13f79535-47bb-0310-9956-ffa450edef68

Part three of the read_with_timeout refactoring.

2006-03-23T23:19:32+00:00

  Loop on the WaitForSingleObject if it indicated WAIT_ABANDONED,
  which occurs when the thread/proc which created the event exits,
  and ownership of the event has been transfered.

  Always try to CancelIo if the wait has failed.

  Ignore the Wait/Cancel results and then always check the completion
  status of the original Read.  Indicate TIMEUP when appropriate.


git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@388292 13f79535-47bb-0310-9956-ffa450edef68

Fix with additional notes for older PSDK winbase.h headers.

2006-03-23T22:47:12+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@388287 13f79535-47bb-0310-9956-ffa450edef68

Minor typo fix to previous commit.

2006-03-23T22:44:16+00:00

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@388285 13f79535-47bb-0310-9956-ffa450edef68

Part 2 of the necessary read_with_timeout() fixes. Catch the

2006-03-23T21:51:11+00:00

  condition where our broken pipe occurs durring the deferred
  i/o completion phase.

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@388282 13f79535-47bb-0310-9956-ffa450edef68

Part 1 of many read_with_timeout logic fixes. Stop polluting one

2006-03-23T21:49:49+00:00

  occurance of rv with the boolean result of ReadFile() to increase
  the legibility of the success/failure of ReadFile.  This requires
  us to defer *nbytes assignment to the function's end.

  This fix catches additional cases of APR_EOF, as we had not
  tested this case from the error handling path.  So any deferred
  read of zero bytes previously returned 0 bytes APR_SUCCESS 
  rather than APR_EOF.  (This occurs when we wait to discover the 
  owner of the write end closes it without additional data)



git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@388281 13f79535-47bb-0310-9956-ffa450edef68