Missing EN line in orientation file

[scipascal]

The SpacecraftHistory removes the last line from the orientation (.ori) file during loading. Since there is no EN line in DC3_final_530km_3_month_with_slew_15sbins_GalacticEarth_SAA.ori, the last time bin is lost, causing some background events to have timetags outside of the time axis. Seems also to cause problems with the SAA cut.

[israelmcmc]

While we could fix the file, it seems easier to check in the code if the last line is indeed "EN", and otherwise not remove it.

[GallegoSav]

@jdbuhler Seems to be related to .ori extension but did you also remove the last value when you converted the ori file into fits file ?

[jdbuhler]

I parsed the .ori file to create the SpacecraftHistory and then called write_fits() to write the FITS table. So whatever that code path does is what I did. I think the the .ori reader strips the last value, and the FITS writer puts a dummy value back in?

[GallegoSav]

@jdbuhler The fact is now there is no more EN at the end of DC3_final_530km_3_month_with_slew_15sbins_GalacticEarth_SAA.ori so it removes the last time value

[jdbuhler]

I commented on this in the PR. The FITS files I generated from DC3_final_* are clearly wrong because they dropped the last line, so I'll need to regenerate. But are we sure we want to accept .ori files without a trailing EN marker? This seems like we're inviting future unexpected behavior.

[jdbuhler]

I have uploaded fixed FITS orientation files to the develop tree on wasabi and issued PR #517 with the corresponding notebook checksum changes. I have not touched the parsing code in SpacecraftHistory.