Hey Marcus,
Quite apart from the GPSD-specific stuff, it would be useful to have a kind of "annotated
formatted data" file sink for low-rate
data (like scientific data, etc) which supports CSV outputs, and has some way of
ordering any tags that come in with the
data and formatting them appropriately, but still be reasonably "generic" (in that the
block doesn't really *know* anything
about the tags, and can be told how to format them). Maybe *in addition* you could
pass the block a series of methods to
call at some configurable cadence.
HDF5, maybe? I think Haystack had something in that direction…
This somewhat overlaps with what SigMF tries to do, but this is formatted data, inherently
low-rate, with formatted
annotations.
I nearly always end up doing this in a custom way for my applications, but having a
generic block would be cool, and not
hard to do. Could be entirely in Python, since we're talking "low rate" (like perhaps
a few per second at most).
Do agree, but really don't like CSV, too underspecified a format, too many ways that comes
back to bite you (aside from a thousand SDR users writing emails that their PC can't keep
up with writing a few MS/s of CSV…)
How important is plain-textness in your applications?
Best,
Marcus
