fix(robot-server): do not save calibration offsets if an error occurs

[Laura-Danielle]

Overview

A fix for the behavior where pipette offsets get saved even if an error occurs during the calibration flow.

Test Plan

Follow steps in this ticket to reproduce.

Changelog

• Save the offset inside the robot server function rather than the pipette function
• Ensure that we're raising on edge detection failed errors as well

Review requests

Please test on a robot, let me know if anything seems off.

Risk assessment

Low. Should be fixing a small bug.

[codecov]

Codecov Report

Merging #13168 (0077b45) into internal-release_0.14.0 (57b7c9b) will increase coverage by 0.06%.
Report is 10 commits behind head on internal-release_0.14.0.
The diff coverage is n/a.

@@                     Coverage Diff                     @@
##           internal-release_0.14.0   #13168      +/-   ##
===========================================================
+ Coverage                    72.53%   72.59%   +0.06%     
===========================================================
  Files                         2359     2372      +13     
  Lines                        64994    65416     +422     
  Branches                      7213     7209       -4     
===========================================================
+ Hits                         47143    47490     +347     
- Misses                       16105    16181      +76     
+ Partials                      1746     1745       -1     

Flag	Coverage Δ
g-code-testing	96.44% <ø> (ø)
notify-server	89.13% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

see 15 files with indirect coverage changes

[sfoster1]

I like the idea but it's sort of a weird middle ground between a refactor and a new feature.

I think the simplest way to do this is to change ot3_calibration.calibrate_pipette to put the save_instrument_offset_call after the finally? (although we definitely want to restore the old one). But if we want a whole new different function like this, then we should have calibrate_pipette call the new function instead of just copy-pasting it; and also we should probably remove calibrate_pipette since I don't think anything else would use it now.

Let's do one of those two (either alter calibrate_pipette, or implement it via find_pipette_offset and/or delete it).

[ahiuchingau]

Do you think we'll need to fix the gripper offset saving mechanism as well? I don't see why this behavior would only be happening to the pipettes and not the gripper as well? What am i missing?

[Laura-Danielle]

Do you think we'll need to fix the gripper offset saving mechanism as well? I don't see why this behavior would only be happening to the pipettes and not the gripper as well? What am i missing?

@ahiuchingau the gripper already does the mechanism above. Previously, the pipette was saving calibration inside the try/except of the ot3_calibration.py function.

[sfoster1]

Thank you, looks good to me!