Scripts to convert KiCad files for JLCPCB assembly service

convert_bom.py

# This converts a BOM file in CSV format, produced by KiCad Pcbnew,
# by selecting File / Fabrication Outputs / BOM File
# into the format expected by the JLCPCB manufacture service.
# At least in KiCad 5.1.4 the CSV output seems a bit broken in terms of column names,
# which is reflected in this code. If it changes in later KiCad versions, this script
# needs to change, too.

# Use this by running the command line: python convert_bom.py inputfile.csv outputfile.csv
# It will overwrite your output file without asking so be careful.

import pandas as pd
from sys import argv, exit

# Select these columns from the input file
FILTER = ["Id", "Quantity", "Designator"]

# Rename the columns based on these rules
JLCPCB_NAMES = {
    "Designator" : "Footprint",
    "Quantity" : "Comment",
    "Id" : "Designator"
}

# Read data, filter, rename and write output
data = pd.read_csv(argv[1], delimiter=";")
filtered = data[FILTER]
result = filtered.rename(JLCPCB_NAMES, axis="columns", errors="raise")
result.to_csv(argv[2], index=False)

convert_pos.py

# This converts a component placement file in CSV format, produced by KiCad Pcbnew,
# by selecting File / Fabrication Outputs / Footprint position (.pos) file
# into the format expected by the JLCPCB manufacture service.

# In KiCad, use the following options:
#   Format = CSV
#   Units = Millimeters
#   Files = Single file for board
#   Include footprints with SMD pads even if ... = unchecked

# Use this by running the command line: python convert_pos.py inputfile.csv outputfile.csv
# It will overwrite your output file without asking so be careful.

import pandas as pd
from sys import argv, exit

# Select these columns from the input file
FILTER = ["Ref", "PosX", "PosY", "Side", "Rot"]

# Rename the columns based on these rules
JLCPCB_NAMES = {
    "Ref" : "Designator",
    "PosX" : "Mid X",
    "PosY" : "Mid Y",
    "Side" : "Layer",
    "Rot" : "Rotation"
}

# Read data, filter, rename and write output
data = pd.read_csv(argv[1])
filtered = data[FILTER]
result = filtered.rename(JLCPCB_NAMES, axis="columns", errors="raise")
result.to_csv(argv[2], index=False)