118 lines
3.7 KiB
Python
118 lines
3.7 KiB
Python
#
|
|
# Example python script to generate a BOM from a KiCad generic netlist
|
|
#
|
|
# Example: Sorted and Grouped CSV BOM
|
|
#
|
|
|
|
"""
|
|
@package
|
|
Output: CSV (comma-separated)
|
|
Grouped By: Value, Footprint, DNP, specified extra fields
|
|
Sorted By: Reference
|
|
Fields: #, Reference, Qty, Value, Footprint, DNP, specified extra fields
|
|
|
|
Outputs components grouped by Value, Footprint, and specified extra fields.
|
|
Extra fields can be passed as command line arguments at the end, one field per argument.
|
|
|
|
Command line:
|
|
python "pathToFile/bom_csv_grouped_extra.py" "%I" "%O.csv" "Extra_Field1" "Extra_Field2"
|
|
"""
|
|
|
|
# Import the KiCad python helper module and the csv formatter
|
|
import kicad_netlist_reader
|
|
import kicad_utils
|
|
import csv
|
|
import sys
|
|
|
|
# Get extra fields from the command line
|
|
extra_fields = sys.argv[3:]
|
|
|
|
comp_fields = ['Value', 'Footprint', 'DNP'] + extra_fields
|
|
header_names = ['#', 'Reference', 'Qty'] + comp_fields
|
|
|
|
def getComponentString(comp, field_name):
|
|
if field_name == "Value":
|
|
return comp.getValue()
|
|
elif field_name == "Footprint":
|
|
return comp.getFootprint()
|
|
elif field_name == "DNP":
|
|
return comp.getDNPString()
|
|
elif field_name == "Datasheet":
|
|
return comp.getDatasheet()
|
|
else:
|
|
return comp.getField( field_name )
|
|
|
|
def myEqu(self, other):
|
|
"""myEqu is a more advanced equivalence function for components which is
|
|
used by component grouping. Normal operation is to group components based
|
|
on their Value and Footprint.
|
|
|
|
In this example of a more advanced equivalency operator we also compare the
|
|
Footprint, Value, DNP and all extra fields passed from the command line. If
|
|
these fields are not used in some parts they will simply be ignored (they
|
|
will match as both will be empty strings).
|
|
|
|
"""
|
|
result = True
|
|
for field_name in comp_fields:
|
|
if getComponentString(self, field_name) != getComponentString(other, field_name):
|
|
result = False
|
|
|
|
return result
|
|
|
|
# Override the component equivalence operator - it is important to do this
|
|
# before loading the netlist, otherwise all components will have the original
|
|
# equivalency operator.
|
|
kicad_netlist_reader.comp.__eq__ = myEqu
|
|
|
|
# Generate an instance of a generic netlist, and load the netlist tree from
|
|
# the command line option. If the file doesn't exist, execution will stop
|
|
net = kicad_netlist_reader.netlist(sys.argv[1])
|
|
|
|
# Open a file to write to, if the file cannot be opened output to stdout
|
|
# instead
|
|
try:
|
|
f = kicad_utils.open_file_writeUTF8(sys.argv[2], 'w')
|
|
except IOError:
|
|
e = "Can't open output file for writing: " + sys.argv[2]
|
|
print(__file__, ":", e, sys.stderr)
|
|
f = sys.stdout
|
|
|
|
# Create a new csv writer object to use as the output formatter
|
|
out = csv.writer(f, lineterminator='\n', delimiter=',', quotechar='\"', quoting=csv.QUOTE_ALL)
|
|
|
|
# Output a CSV header
|
|
out.writerow(header_names)
|
|
|
|
# Get all of the components in groups of matching parts + values
|
|
# (see kicad_netlist_reader.py)
|
|
grouped = net.groupComponents()
|
|
|
|
# Output all of the component information
|
|
index = 1
|
|
for group in grouped:
|
|
refs = ""
|
|
|
|
# Add the reference of every component in the group and keep a reference
|
|
# to the component so that the other data can be filled in once per group
|
|
for component in group:
|
|
refs += component.getRef() + ", "
|
|
c = component
|
|
|
|
# Remove trailing comma
|
|
refs = refs[:-2]
|
|
|
|
# Fill in the component groups common data
|
|
row = []
|
|
row.append( index )
|
|
row.append( refs )
|
|
row.append( len(group) )
|
|
|
|
# Add the values of component-specific data
|
|
for field_name in comp_fields:
|
|
row.append( getComponentString( c, field_name ) )
|
|
|
|
out.writerow(row)
|
|
|
|
index += 1
|