Unexpected behavior for channel map system

[iguinn]

When grabbing a channel map for a specific time (in this case P06 R002) using the following command:

chmap = meta.channelmap(run_dict.cal.start_key)

it is giving me the phy parameters (which differ in their PSD availability from the cal parameters).

The default should be all. It's actually not totally obvious to me what all should mean, especially in a case where cal and phy differ. In this case, all should match cal as far as I can tell by digging through some of the metadata files, so I'm confused as to why I got phy.

On top of this, it's quite confusing to get phy parameters when the start_key corresponds to cal data. And I doubt I'm the first or last person to be confused by this. I wonder if we want to change the default behavior to either select the system default using the timestamp, or have no default and require people to use the system argument to avoid this?

[gipert]

@ggmarshall thoughts?

[ggmarshall]

Hi, so cal is a special dataype which is why it has its own system which needs to be specified in the call, while phy is just the default running (i.e. runs through the basic processing) so is treated as all the same way lac or any other datatype that does not require special treatment is run.

If we were to change this I would remove the default, I don't like the idea of making the behaviour dependent on some run file that would need to be continually updated and synced through all productions/users etc.

[ggmarshall]

I guess the other option would be some kind of log message/ warning if unspecified