This test doesn't really make sense without the use of an inbox so it serves no useful function at the moment, therefore I'm taking it out. Signed-off-by: Frank Quinn <fquinn.ni@...> ---

MAMA expects iterators to rewind when begin is called, and remain pointing to the first entry after the *first* time that next is called. This change aligns with that requirement. Signed-off-by: Frank

The unit tests assumed that the order of entry into a message field will be identical to the order of extraction while iterating. This proved not to be the case with avis and possibly other

This unit test was testing for the wrong thing. Really all it should have been testing was that the methods work OK when a NULL name is provided, but a fid is non-NULL. Signed-off-by: Frank Quinn

There was already support for opaque data types but they could not be serialized or deserialized. This patch should remedy this. Signed-off-by: Frank Quinn <fquinn.ni@...> ---

The previous tests pretty much assumed it wasn't implemented in the bridge. This fix adds some basic testing around this functionality for self-describing messages. Signed-off-by: Frank Quinn

This was done with a simple assignment. Will need revised if MAMA ever decide to change their datetime format. Signed-off-by: Frank Quinn <fquinn.ni@...> ---

Like qpid, this is simply a call to the internal bridge function to mute the subscription's topic. Signed-off-by: Frank Quinn <fquinn.ni@...> --- mama/c_cpp/src/c/bridge/avis/sub.c | 2 +- 1 file

Several unit tests expect MAMA bridges to return standard error codes in various scenarios. This set of updates should bring avis into line with these expectations. Signed-off-by: Frank Quinn

In future, we may want to make this code central. In the meantime, at least making the code identical with qpid should make it easier to maintain. Signed-off-by: Frank Quinn <fquinn.ni@...> ---

In future, we may want to make this code central. In the meantime, at least making the code identical with qpid should make it easier to maintain. Signed-off-by: Frank Quinn <fquinn.ni@...> ---

This is with the exception of the timer implementation which has several bugs which have already been fixed in the qpid queue implementation so the qpid timer should instead simply be converted into

Hi Folks, I am planning on dropping support for Qpid Proton 0.5 and 0.6 in the next OpenMAMA release and instead sticking to versions 0.7+. Those guys move quickly and make small changes to interfaces

Hi Nestor, If you modify each new subscriber to use a different incoming URL and reply URL, you can actually do this today, though as you can imagine it becomes a bit of a brute to configure very

Hi guys, I'd like to know if there was or there will be some change to the Qpid Proton Bridge such that I can have a publish/subscribe connection in which only one advanced publisher can have multiple

Pushed this change earlier on but sending around for transparency. Avis tests still fail but at least they finish running (and a lot of the failures are expected due to lack of support in the avis

Hi Folks, It is with great pleasure that we can announce the release of OpenMAMA 2.3.3. The latest release contains several minor bugfixes since 2.3.2 and the following new features: * Support added

Also deprecated support for EOL distros FC19 and FC20. Signed-off-by: Frank Quinn <fquinn.ni@...> --- release_scripts/openmama-rpm.sh | 15 ++++++++------- release_scripts/openmama.spec | 4 ++--

Added i686 and a few other similar ancestors for good measure. Signed-off-by: Frank Quinn <fquinn.ni@...> --- site_scons/community/command_line.py | 2 +- 1 file changed, 1 insertion(+), 1

From: Frank Quinn <fquinn.ni@...> --- mama/c_cpp/src/c/bridge/qpid/qpiddefs.h | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git