From fc41cf9544d296410e8d6ef3999fba2c2a1e39b1 Mon Sep 17 00:00:00 2001 From: andimik Date: Fri, 25 Oct 2019 21:27:33 +0200 Subject: Rename TODO to TODO.md --- TODO.md | 74 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) create mode 100644 TODO.md (limited to 'TODO.md') diff --git a/TODO.md b/TODO.md new file mode 100644 index 0000000..12a4a40 --- /dev/null +++ b/TODO.md @@ -0,0 +1,74 @@ +This TODO file lists ideas and features for future developments. They are +more or less ordered according to their benefit, but that is subjective +to some degree. + +Unless written, no activity has been started on the topics. + + +EDI/STI-D Input +--------------- +Add support for EDI input, so that third-party encoders can be interfaced +to the multiplexer. Relevant spec: ETSI TS 102 693 + +It would be beneficial to write and EDI decoding library that can be re-used in +ODR-DabMod to add EDI input there too. + +Stuff in http://git.mpb.li/git/odr-edilib/ could be helpful. + +There is experimental support for STI(PI, X)/RTP using UDP, which is used by +AVT encoders. It does not properly reorder incoming UDP packets nor align +STI frames with ETI frames. + +Also, it doesn't print out nice messages (like ZMQ's "prebuffering complete"). + + +Explicit Service Linking +------------------------ +It is impossible to activate/deactive linkage sets. Commit 5c3c6d7 added +some code to transmit a FIG0/6 CEI, but this was subsequently reverted +because it was not tested enough. + + +Inputs for packet data +---------------------- +It is currently unclear what input formats and sources work for packet data, +and which ones would make sense to add. + +Also, there is no documentation on the possibilites of packet data. + + +Multicast support for the UDP input +----------------------------------- +The current UDP input implementation cannot join a multicast group. + + +Improvements for inputs +----------------------- +Add statistics to UDP input, in a similar way that ZeroMQ offers statistics. +This would mean we have to move the packet buffer from the operating system +into our own buffer, so that we can actually get the statistics. + +Also see the STI-D over EDI topic above. + +For error handling, there are parts of the code that use return values, parts +use exceptions. While it works, we could clean it up a bit. + + +Fix DMB input +------------- +The code that does interleaving and reed-solomon encoding for DMB is not used +anymore, and is untested. The relevant parts are src/dabInputDmb* and src/Dmb.cpp + + +Communicate Leap Seconds +------------------------ +Actually, we're supposed to say in FIG0/10 when there is a UTC leap second +upcoming, but since that's not trivial to find out because the POSIX time +concept is totally unaware of that, this is not done. We need to know for EDI +TIST, and the ClockTAI class can get the information from the Internet, but it +is not used in FIG0/10. + + +Implement FIG0/20 Service List +------------------------------ +See ETSI TS 103 176 -- cgit v1.2.3