without this, deleting the last reader/writer that references the topic results in a dangling pointer ... but there is another intriguing solution: erase the topic from the proxy reader/writer when the last matching local one disappears, so that the topic completely disappears. I rather like this second solution, but I am not yet sure of the consequences and the first (implemented one) is such a simple change that fixes a real problem that it is a no-brainer
Signed-off-by: Erik Boasson <eb@ilities.com>
This addresses the deadlock of #41 but leaves another issue open: sequencing of listener invocations on publication/subscription matched events: there is a risk that the "unmatch" even precedes the "match" event from the application perspective, even though it is quite unlike in practice. Various ways of addressing it exist, but it looks like sequencing at the level of the "dds" entities suffers from similar risks. So better to just avoid the deadlock for now.
Signed-off-by: Erik Boasson <eb@ilities.com>
Previously it would fall through and assert in a debug build or return an error in a release build. The behaviour in a release build was almost correct, as the flag means the entity should be completely ignored if the parameter is not understood by the implementation, but I don't believe it should result in a warning — certainly not that claims the parameter list is invalid. A specific return code is now used to indicate a parameter list that was rejected because of this flag, and that suppresses the warning.
Signed-off-by: Erik Boasson <eb@ilities.com>
this is just an intermediate step: firstly, iid-to-key lookup is a linear scan; secondly, the lookup should happen on the RHC or the WHC and not on some internal map to associates key values with instance handles and is independent of entities
Signed-off-by: Erik Boasson <eb@ilities.com>
They were made available from the DCPS layer to DDSI using some function pointers, but it makes more sense to have them as part of the DDSI core.
Signed-off-by: Erik Boasson <eb@ilities.com>
liveliness monitoring requires rtps stack initialised if initially deaf, so starting the monitoring thread must happen after rtps_init
Signed-off-by: Erik Boasson <eb@ilities.com>
- topic-erased key-only serdata for use in tkmap
- restoration of including key values in invalid samples
- special handling of keyless topics
- keyhash generation via streams
- elimination of dynamically allocated buffers in keyhash
- removal of the last vestiges of "serstate"
Signed-off-by: Erik Boasson <eb@ilities.com>
besides the old state being in dire need of cleaning up, this also paves the way for having any number of different sample representations in the system
Signed-off-by: Erik Boasson <eb@ilities.com>
* Consolidated and cleaned up some (duplicate) functions.
* Removed some unused functions that did not make sense to keep around.
Signed-off-by: Jeroen Koekkoek <jeroen@koekkoek.nl>
