Bugfix for recent regression of using value "correction" which is out of scope.
This was introduced in commit 5ecf6c9b138d0d192f615a7cce9f6a3f0cf5b490
"correction" will simply contain the last correction that was in the database.
Instead, last_correction is the correct value for the last correction with a
matching pattern.
Refactor how bottom layer rotations are calculated, for better clarity. (no functional change)
This splits up bottom layer calculations in two halves:
1) The Kicad-specific half, of corrections needing to be applied with a negative sign on the bottom layer
2) The JLCPCB-specific half, of how to further tweak the rotation numbers on the bottom for it to be correctly handled by JLCPCB.
Update how bottom layer rotations are calculated, to match current JLCPCB behaviour (as first noticed on 2022 Jul 31).
matthewlai
commented
2 years ago
This PR contains 3 commits:
Bugfix for recent regression of using value "correction" which is out of scope.
This was introduced in commit 5ecf6c9b138d0d192f615a7cce9f6a3f0cf5b490 "correction" will simply contain the last correction that was in the database. Instead, last_correction is the correct value for the last correction with a matching pattern.
Refactor how bottom layer rotations are calculated, for better clarity. (no functional change)
This splits up bottom layer calculations in two halves: 1) The Kicad-specific half, of corrections needing to be applied with a negative sign on the bottom layer 2) The JLCPCB-specific half, of how to further tweak the rotation numbers on the bottom for it to be correctly handled by JLCPCB.
Update how bottom layer rotations are calculated, to match current JLCPCB behaviour (as first noticed on 2022 Jul 31).