Question about XTCE

[ginoscott]

Hi F' team,

I've found some older references to F' supporting XTCE, but haven't been able to see anything more recently than 2018-ish. I have an interest in the interoperability of open source tools and am wondering if there is support for it and if there isn't, if there are plans to and if there are no plans, what were some of the concerns that made XTCE an unattractive option.

Thank you!

[timcanham]

We looked into XTCE in an admittedly superficial way to determine what the mapping is from our dictionaries to XTCE, but we didn't try to incorporate it into our tools. At some point we need to do a deeper dive to characterize it more, but on the surface we found:

1. It doesn't seem to support the F Prime notion of events, i.e. notifications of activities on board. It seems to be more telemetry-oriented.
2. It has limited support for complex types (i.e. structs and structs of structs, arrays, etc) in telemetry and commanding.

F Prime now represents its dictionaries as JSON documents which gives us the expressiveness we need. We had decided that if a project wanted to deliver an XTCE dictionary to a ground system, they could write a converter that would take the F Prime JSON dictionary and convert it to an XTCE equivalent. A project could limit the specification of their commands and telemetry to that which fits in the XTCE specification and perhaps use it, but we never really had a chance to propose a way to add events to XTCE.

If someone on this forum wanted to do a deeper analysis to counter any wrong conclusions and propose a way to incorporate XTCE, it would certainly be welcome! For us, it's a bandwidth limitation.