Permalink
Commits on Feb 18, 2012
  1. Fixes header buffer names

    alobbs committed Feb 18, 2012
Commits on Feb 14, 2012
  1. Removes reply_100_continue()

    alobbs committed Feb 14, 2012
Commits on Feb 13, 2012
  1. protocol::add_response

    alobbs committed Feb 13, 2012
Commits on Feb 10, 2012
  1. No rx/tx in request any longer

    alobbs committed Feb 10, 2012
Commits on Jan 12, 2012
  1. WIP: new conn 2

    alobbs committed Jan 12, 2012
  2. WIP: nw conn

    alobbs committed Jan 12, 2012
Commits on Jan 10, 2012
  1. Renames a whole lot of 'conn' variables to 'req'. The target is to

    clean the namespace a little bit, so the actual work of implementing
    the new classes can be started safely.
    alobbs committed Jan 10, 2012
Commits on Jan 8, 2012
  1. Initial renaming of cherokee_connection_t to cherokee_request_t. The

    internal symbols exported by the old connection.c file have not been
    converted to the new cherokee_request_* namespace though.
    alobbs committed Jan 8, 2012
  2. Namespace clean up: Renames cherokee_request_t to

    cherokee_client_request_t.
    alobbs committed Jan 8, 2012
Commits on Jan 5, 2012
  1. Little clean up

    alobbs committed Jan 5, 2012
  2. cherokee_connection_build_header() can return ret_eagain with and

    without requiring any I/O polling.
    alobbs committed Jan 5, 2012
Commits on Jan 4, 2012
  1. Gets rid of more ancient code from the pre-new-events age. Basically,

    we do not need to keep track of connections's I/O mode any longer.
    alobbs committed Jan 4, 2012
Commits on Jan 3, 2012
  1. Fixes the kqueue() backend.

    alobbs committed Jan 3, 2012
Commits on Jan 2, 2012
  1. Timout error responses should be send only if the client sent a

    request previously. Otherwise, the connection should be just closed.
    alobbs committed Jan 2, 2012
Commits on Dec 30, 2011
  1. Gets rid of a couple of properties of the cherokee_thread_t class. It

    was not necessary to keep track of the length of the connection lists.
    alobbs committed Dec 30, 2011
  2. The new events management mechanism shouldn't need the "Pending

    connections number" thread property either. This could introduce
    regressions in the management of pipelined requests.
    alobbs committed Dec 30, 2011
  3. This branch should make the "pending read number" unnecessary.

    Hopefully this does not introduce any regression in the SSL support.
    alobbs committed Dec 30, 2011
Commits on Dec 27, 2011
Commits on Dec 25, 2011
Commits on Dec 23, 2011
Commits on Dec 21, 2011
  1. Two minor fixes

    alobbs committed Dec 21, 2011
Commits on Dec 19, 2011
  1. Fix for the multi-thread mode

    alobbs committed Dec 19, 2011
Commits on Dec 18, 2011
  1. Fixes the Win32 backend

    alobbs committed Dec 18, 2011