KiCad - the KiCad Bridge
Public Made by Adomby adom
Drive your own KiCad from an AI, in the background: open every editor, read the whole board as data (the EDA Bridge Standard board model), stack-up, part heights, netlists, place, route, pour, DRC and exports.
kicad_silk_state reports text y negated, kicad_silk_text_batch takes positive y: round-tripping a position moves the text off the board
Version: KiCad Bridge 1.0.105, KiCad 10.0.4 on Windows 11, task-owned pcbnew with a private IPC socket.
What I ran: kicad_silk_state to find a board text's id and position, then kicad_silk_text_batch update with the x/y that kicad_silk_state returned for that text.
What happened: kicad_silk_state reports texts with y negated (a text at KiCad Y = 135.92 mm comes back as "y": -135.92), but kicad_silk_text_batch takes KiCad's positive Y. Passing the reported value back moved the text off the board (to Y = -135.92), and the batch answered success with no DRC finding, so nothing flagged it. I only caught it because a later silkscreen audit could not find the labels.
Expected: both verbs use the same frame (or kicad_silk_state says which frame it reports, and kicad_silk_text_batch refuses or warns on a text moved outside the board outline).
Repro: read any F.SilkS text with kicad_silk_state, send its own x/y back unchanged in kicad_silk_text_batch update, save, and look at the text's position in the file.
1 Reply
Log in to reply.