Closed anetasie closed 1 month ago
So, there's a long philosophy about that should happen to the CALDBVER
keyword.
I believe the feeling is, and @kglotfelty can correct me where I'm wrong, is that we can never guarantee that we can give a sensible value here so it's safer not to. Now, I do wonder whether this is true for the"specialized" case that is chandra_repro
, but I'm willing to bet there's a reason.
What @DougBurke said. Even with all chandra_repro
values set to defaults, there will still be calibrations that are not updated (things like recomputing aspect solution, good time intervals, mission timeline, etc.). The CALDBVER
keyword is much less useful than the *FILE
keywords.
I was checking the CALDBVER for the version of the processing and was seeing the original version for processing the file. Of course I could just
grep CALDB
which would return other versions, but I'm wondering if we could update the keywords to reflect the version which was used. Now, I'm also wondering if the gain files used in the chandra_repro were the most recent ones.