Closed kglotfelty closed 1 month ago
Ahh. I had assumed Vinay was just using an old dataset he had lying around!
Update: OK, so the problem isn't missing *FILE keywords. The problem is that the DM is fouling up the GTIs.
r4_header_update uses the ObsInfo
class to locate the files, but after running a_p_e, the GTIs are even more fouled up and it ends up creating a generic GTI
blocks w/o a CCD_ID keyword which throws an exception in the ObsInfo init.
The "real" fix will need to happen in the DM; however, this change, to specify the file names explicitly is a good work around, so still a useful PR.
Vinay found an observation in the archive that has not been through repro3 so the *FILE keywords used to automatically locate auxillary files are missing. Ref hd tix 25280.
This PR updates the way
chandra_repro
callsr4_header_update
to specify the pbk and asol files names. This works whether the FILE keywords are present or not. (There is other logic in chandra_repro that already checks that the FILE keywords matches what it finds when globbing through the primary and secondary dirs.)