[kriboe90]

Fixes #173.

start_time is on the real clock: get_start_time subtracts exif_time_infront_real_time from the first image's EXIF time. Experiment_Time[s], however, was measured from the uncorrected EXIF time, so it was off by the offset:

• For a single camera, the first image did not sit at t = 0.
• For several cameras, the stacked analysis kept them apart by the difference of their offsets.

Change

• _calc_experiment_and_real_time now measures the experiment time from the corrected image time.
• New test file test_experiment_time_clock_offset.py with two tests:
  • when start_time is derived from the first image, that image is at 0;
  • two cameras with offsets of +30 s and −5 s and the same start_time get the same experiment times.

Both tests fail without the fix. All unit tests pass with it. The acceptance tests were not run locally.

Note

image_infos*.csv files written before this change, and any temperature reduction tables aligned to them, are shifted by the offset and need regenerating. The demo's times move by 1 s, because its offset is −1.

🤖 Generated with Claude Code