fix key error modules_mappings

Description

It seems that I merged the latest fix for the summary reports with a key error bug.

This is a hotfix for it on top of 3.2.3 tag.

How Has This Been Tested?

Locally:

xfel-calibrate agipd DARK \
--cal-db-interface tcp://max-exfl016:8015#8044 \
--concurrency-par karabo_da \
--karabo-da AGIPD00 AGIPD01 AGIPD02 AGIPD03 AGIPD04 \
AGIPD05 AGIPD06 AGIPD07 AGIPD08 AGIPD09 AGIPD10 \
AGIPD11 AGIPD12 AGIPD13 AGIPD14 AGIPD15 \
--receiver-id {}CH0 --karabo-id-control SPB_IRU_AGIPD1M1 \
 --karabo-da-control AGIPD1MCTRL00 \
--h5path-ctrl /CONTROL/{}/MDL/FPGA_COMP \
--mem-cells 0 --thresholds-offset-hard 0 0 \
--thresholds-offset-hard-hg 3500 6000 \
--thresholds-offset-hard-mg 3500 9000 \
--thresholds-offset-hard-lg 3500 9000 \
--in-folder /gpfs/exfel/exp/SPB/202130/p900188/raw \
--out-folder /gpfs/exfel/data/scratch/ahmedk/test/fixed_gain/SPB_summary_fix2 \
--karabo-id SPB_DET_AGIPD1M-1 \
--operation-mode ADAPTIVE_GAIN \
--run-high 302 --run-med 303 --run-low 304

AGIPDCharacterizeDarkImages.pdf

on xcal@max-exfl016: http://max-exfl016:8008//gpfs/exfel/d/cal/caldb_store/xfel/reports/SPB/SPB_DET_AGIPD1M-1/dark/dark_002932_r46_r47_r48_210412_122200.pdf

Relevant Documents (optional)

Types of changes

• Bug fix (non-breaking change which fixes an issue)

Checklist:

• My code follows the code style of this project.

Reviewers

@calibration @kluyvert

changed the description

I like the fix, LGTM.

Any ideas on how we could avoid this type of problems in the future?

Make sure to fill the checklist of tests after deployment as we always do. I think we already have what we need to avoid this. This is a mistake from my side. That I haven't double-checked the production reports to fill the test checklist for the latest deployed tag 3.2.3

That's okay, I'm not looking for blame.
The testing is definitely a good step, but I wonder if we could catch this at the MR level.

More tests

Yeah, integration tests are the only way I can see to catch this automatically.

Where does the YAML come from? The code I can see in Characterize_AGIPD_Gain_Darks_NBC.ipynb looks it saves it with a key "module_mapping" (underscore, no s).

I've also just noticed a potential error on this line (just after this change):

if qm not in metadata["modules_mapping"].keys() and metadata["modules_mapping"][qm]:

It looks like if the first test is True, qm is not a key, so the second test will always throw KeyError. It works only where all modules are found, unless I'm missing something.

changed the description

added 1 commit

• 4d7731f7 - fix conditions of unknown detector module

Compare with previous version

added 1 commit

• 4d7731f7 - fix conditions of unknown detector module

Compare with previous version

added 1 commit

• 8e747ce6 - fix conditions of unknown detector module

Compare with previous version

added 1 commit

• 1ed9ee04 - apply MR suggestion

Compare with previous version

changed the description

If there are no further comments @kluyvert I will merge and deploy 3.2.4 on xcal@max-exfl016

LGTM

Thank you for the review!

merged

mentioned in commit 7b94c17a