You've already forked incutec-KiCad-Library
mirror of
https://github.com/incutec-hw/KiCad-Library.git
synced 2026-08-13 14:36:35 +00:00
72042a6ea1
Answers the sourcing question directly: which parts, especially passives, are already used on a board, so they are sourced, footprinted and probably in stock. Generated from the board schematics, so nothing needs a library symbol to appear here. 214 parts across 8 boards. This is the alternative to adding a library symbol for every capacitor value. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
75 lines
3.1 KiB
Python
75 lines
3.1 KiB
Python
#!/usr/bin/env python3
|
|
"""Build PARTS-USED.md: every component used across the board repos.
|
|
|
|
The point is sourcing, not symbols. If a part appears here we have used it on a
|
|
real board, so it is already sourced, footprinted and probably in stock. Check
|
|
here before drawing a new part. Nothing needs a library symbol to be listed.
|
|
|
|
Run from the KiCad-Library checkout: python3 tools/build-parts-index.py
|
|
"""
|
|
import os, re, glob, collections, subprocess
|
|
|
|
BOARDS = os.path.abspath(os.path.join(os.path.dirname(__file__), '..', '..'))
|
|
SKIP = {'KiCad-Library'}
|
|
PROP = re.compile(r'\(property "([^"]+)" "([^"]*)"')
|
|
|
|
|
|
def symbol_blocks(text):
|
|
"""Yield each symbol instance's property dict from a .kicad_sch."""
|
|
for m in re.finditer(r'\(symbol\b', text):
|
|
start = m.start()
|
|
depth, i = 0, start
|
|
while i < len(text):
|
|
if text[i] == '(':
|
|
depth += 1
|
|
elif text[i] == ')':
|
|
depth -= 1
|
|
if depth == 0:
|
|
break
|
|
i += 1
|
|
block = text[start:i]
|
|
if len(block) < 40000:
|
|
yield dict(PROP.findall(block))
|
|
|
|
|
|
def main():
|
|
parts = collections.defaultdict(lambda: {'boards': set(), 'value': '', 'fp': '', 'mpn': ''})
|
|
for repo in sorted(os.listdir(BOARDS)):
|
|
if repo in SKIP or not os.path.isdir(os.path.join(BOARDS, repo, '.git')):
|
|
continue
|
|
for sch in glob.glob(os.path.join(BOARDS, repo, '**', '*.kicad_sch'), recursive=True):
|
|
if '/libs/' in sch or '/archive/' in sch:
|
|
continue
|
|
try:
|
|
text = open(sch, encoding='utf-8', errors='ignore').read()
|
|
except OSError:
|
|
continue
|
|
for props in symbol_blocks(text):
|
|
lcsc = (props.get('LCSC') or '').strip()
|
|
if not re.fullmatch(r'C\d+', lcsc):
|
|
continue
|
|
e = parts[lcsc]
|
|
e['boards'].add(repo)
|
|
e['value'] = e['value'] or props.get('Value', '')
|
|
e['fp'] = e['fp'] or props.get('Footprint', '')
|
|
e['mpn'] = e['mpn'] or props.get('MPN', '') or props.get('Manufacturer Part Number', '')
|
|
|
|
rows = sorted(parts.items(), key=lambda kv: (-len(kv[1]['boards']), kv[0]))
|
|
out = ["# Parts used", "",
|
|
"Every component with an LCSC code used on an OpenDrone board, generated",
|
|
"from the board schematics. A part listed here is already sourced and",
|
|
"footprinted, so reuse is cheaper than drawing something new.",
|
|
"",
|
|
"Regenerate: `python3 tools/build-parts-index.py`", "",
|
|
f"{len(rows)} distinct parts across {len({b for _, v in rows for b in v['boards']})} boards.", "",
|
|
"| LCSC | Value | MPN | Footprint | Boards |", "|---|---|---|---|---|"]
|
|
for lcsc, e in rows:
|
|
fp = e['fp'].split(':')[-1]
|
|
out.append(f"| {lcsc} | {e['value']} | {e['mpn']} | {fp} | {', '.join(sorted(e['boards']))} |")
|
|
open(os.path.join(os.path.dirname(__file__), '..', 'PARTS-USED.md'), 'w').write('\n'.join(out) + '\n')
|
|
print(f"{len(rows)} parts written to PARTS-USED.md")
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|