Regression: files copied into superevents have mangled filenames
While working on the acceptance tests for v2.0.1 #481 (closed), I found that the MDC superevent I was working on had some mangled filenames for the files that were copied after an update alert was queued: (entry 225 here https://gracedb-playground.ligo.org/superevents/MS221026u/view/)
This seems to have happened after the file was copied from a G-event.
Looking at another MDC superevent (one that wasn't touched by acceptance tests), I found that the subthreshold.bayestar.*
files ended up being named subthreshold.subthreshold.bayestar.*
since they were also copied from G-events.