ULTRA l1c fix exposure time calculation

[lacoak21]

Change Summary

Overview

When I produced the first l2 maps that included the new culling update, the ULTRA team noticed that the exposure looked too high and I realized that I had not been excluding spins that were flagged from the exposure time. I reproduced maps using this branch and they confirmed that this was the necessary fix.

File changes

• imap_processing/ultra/l1b/ultra_l1b_culling.py
  • vectorized version of energy dependent event culling
• imap_processing/ultra/l1c/ultra_l1c_pset_bins.py
  • Calculate exposure time using the correct number of valid spins in the pointing.

Testing

Add a test to check that the correct spins were used in calculating the exposure factor.

[Copilot]

Pull request overview

Fixes ULTRA L1C exposure-time scaling so flagged spins are excluded, aligning exposure factors with the quality-filtered spin selection produced in L1B “goodtimes”.

Changes:

• Update L1C exposure-time normalization to use goodtimes_dataset spin periods and energy-dependent spin-quality flags.
• Vectorize energy/spin-dependent event rejection masking in ULTRA L1B culling.
• Extend unit tests/fixtures to include energy-dependent spin-quality flags and validate the updated exposure-time behavior.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
imap_processing/ultra/l1c/ultra_l1c_pset_bins.py	Switch exposure-time scaling from universal spin table to goodtimes_dataset (quality-filtered) spins, with energy-dependent handling.
imap_processing/ultra/l1c/spacecraft_pset.py	Update callsite to pass energy bins and goodtimes_dataset into exposure-time calculation.
imap_processing/ultra/l1c/helio_pset.py	Same exposure-time callsite update for heliospheric pointing set generation.
imap_processing/ultra/l1b/ultra_l1b_culling.py	Replace per-energy-bin loop with vectorized event rejection mask based on spin/energy flags.
imap_processing/tests/ultra/unit/test_ultra_l1c_pset_bins.py	Add assertions around logging/behavior of good-spin counts per energy bin.
imap_processing/tests/ultra/unit/test_ultra_l1b.py	Update extendedspin mock data to include new energy-dependent flag fields.
imap_processing/tests/ultra/unit/conftest.py	Improve mock goodtimes dataset to include energy ranges/flags and spin-period info.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

[Copilot]

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

[Copilot]

n_spins_in_pointing can become a large negative (or NaN) if goodtimes_dataset contains the fill-value spin period used when no good spins exist (spin_period == -1e31 in calculate_goodtimes). Because np.sum(..., where=good_spins_per_ebin) will still include those fill periods for energy bins mapped to the padded "all True" rows, exposure can be scaled by an invalid spin count.

Consider masking out non-positive / non-finite spin_periods (or explicitly detecting the fill-value/no-good-spins case) when computing n_spins_in_pointing so exposure returns 0 (or another defined fill) when there are no valid spins.

[subagonsouth]

Just the one question about the comment vs. implementation.