Recover metadata in DL1 files

[Bultako]

This PR recovers metadata added to DL1 files during the initial r0_to_dl1 process, that were lost when applying lstchain_dl1ab.py script.

[codecov]

Codecov Report

Merging #728 (ffe16c1) into master (7819257) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #728   +/-   ##
=======================================
  Coverage   84.50%   84.50%           
=======================================
  Files          62       62           
  Lines        5074     5074           
=======================================
  Hits         4288     4288           
  Misses        786      786           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7819257...ffe16c1. Read the comment docs.

[moralejo]

Hi @Bultako, I was just trying lstchain_dl1ab.py on real data and there seems to be a problem. The output DL1 file is produced, but then the program gives these messages:

Metadata SOURCE_FILENAMES does not exist in file ../orig/dl1_LST-1.Run02975.0080.h5
Metadata LSTCHAIN_VERSION does not exist in file ../orig/dl1_LST-1.Run02975.0080.h5
Metadata CTAPIPE_VERSION does not exist in file ../orig/dl1_LST-1.Run02975.0080.h5
Metadata CONTACT does not exist in file ../orig/dl1_LST-1.Run02975.0080.h5

and never ends... It keeps "running" (doing nothing, the output file is not growing) until one kills it. Upon killing it, the output files are readable (could not check in detail of they are fully ok).

Given the messages about metadata I guess this may have something to do with the changes in this PR. Can you please have a look?

[moralejo]

The lstchain_dl1ab script works on MC, by the way, I only found this issue with real data.

[Bultako]

Hi @moralejo
It is indeed weird to have such never-ending process when running lstchain_dl1ab.py script. The warning messages are ok when the input file does not have these metadata, but this should not prevent you from ending the script correctly. if you are running lstchain in a local env you can always comment the two lines below in lstchain_dl1ab.py script and see if we can blame on the changes of this PR.

cta-lstchain/lstchain/scripts/lstchain_dl1ab.py

Line 180 in a84c82b

metadata = read_metadata(args.input_file)

cta-lstchain/lstchain/scripts/lstchain_dl1ab.py

Line 307 in a84c82b

write_metadata(metadata, args.output_file)

In the meantime, I will try to reproduce this issue with those input files.
I will let you know.