Closed pravirkr closed 2 years ago
Thanks for the PR on this. But I'm a bit confused: is refdm
used for filterbank files? If so, what does it mean? I thought that it was only for time series.
And if it is used for filterbank files (i.e. maybe coherent DM used for each channel, or a DM used to subband a previous filterbank file), then PRESTO will possibly process the data incorrectly without an error since this PR doesn't handle it any way other than reading the parameter.
I am also not aware of any general use case of refdm
for filterbank files. However, many filterbank packages (dspsr
, sigproc
, including sigproc.py
in PRESTO) support reading this keyword (not utilized though, except for timeseries). So readfile
should at least allow parsing refdm
and not throw an error.
I am not sure how much PRESTO internals need to be changed if the refdm
also reflects the state of data in the file.
I plan to utilize refdm
in sigpyproc
to read/write a block of dedispersed data as filterbanks.
What do you suggest? Maybe we can allow this PR for refdm = 0
cases and add a warning perhaps for other values.
OK. Fair enough. I guess I was just wondering if there was a need for PRESTO to do more with that flag if it exists. And I guess the answer is no (at least for now). Given that, I see no harm in merging this. Thanks!
Adds the ability to read the
refdm
keyword in SIGPROC filterbank headers.Currently, it results in (with
readfile
)ERROR: read_filterbank_header - unknown parameter: refdm