diff options
Diffstat (limited to 'modules/CIAO/connectors/dds4ccm/docs/details.txt')
| -rw-r--r-- | modules/CIAO/connectors/dds4ccm/docs/details.txt | 33 |
1 files changed, 33 insertions, 0 deletions
diff --git a/modules/CIAO/connectors/dds4ccm/docs/details.txt b/modules/CIAO/connectors/dds4ccm/docs/details.txt new file mode 100644 index 00000000000..8d871ba686b --- /dev/null +++ b/modules/CIAO/connectors/dds4ccm/docs/details.txt @@ -0,0 +1,33 @@ +$Id$ + +For more information regarding DDS4CCM also check the CCM tutoral, see +$CIAO_ROOT/docs/OMG-CCM-Tutorial.ppt + +Currently the exact syntax of IDL3+ is not clear. The batch 4 document has +been outdated and we are waiting for a new batch 5. Before this will be +published a decision on IDL3+ has to be made. The latest proposal is to just +use templated modules, no templated interfaces. The usage of $ as +concatanation symbol has been removed from the spec. + +A lis of all open issues can be found at: +http://www.omg.org/issues/dds4ccm-ftf.open.html + +The list of issues that have been resolved are in the DDS for CCM FTF +27Aug09 document in the same directory as this file. + +One issue that has to be reraised is the fact that DDS SampleInfo is not +available at the CCM level, we need to have access to that. + +The DDS4CCM spec still has to be reviewed whether the given ports deliver the +correct abstraction. We want to prevent that we get lightweight DDS for +lightweight CCM. + +The CIAO tooling will generate the full connector. The goal is to have just +template instantiations and simple pass through code inside the connector, all +real code should be in a supporting core library. That makes it possible to +change the connector without changing the code generators. + +A problem to resolve is the fact that DDS defines an implicit sequence. We +need to be able to have the same sequence in IDL to be used for DDS and CCM. +This is not possible at this moment because we have 2 different code +generators each generating the same type. |