Files
dxdesigner-part-replacement/tools/extract_cbb_native_replay.py

102 lines
6.3 KiB
Python

"""Build an explicit UI replay from two equal, successful real CAD inventories."""
import argparse
from collections import defaultdict
import csv
from dataclasses import asdict
import json
from pathlib import Path
import re
import sys
sys.path.insert(0, str(Path(__file__).resolve().parents[1] / 'src'))
from openpyxl import load_workbook
from dx_productivity.trial_task_center.bom_plan import read_bom_records, write_bom_template
from dx_productivity.trial_task_center.inventory import read_component_inventory
from dx_productivity.trial_task_center.quick_run import _read_discovered_sheets, _decode_vbs_field
def extract(native_runs, converted_dir, output):
output.mkdir(parents=True, exist_ok=True)
candidates = defaultdict(list)
for run in native_runs:
for job in (run / 'data/tasks').glob('*/quick-run-history/sheet-discovery-*'):
if not all((job / name).exists() for name in (
'components.tsv', 'worker-output.txt', 'private-desktop.json', 'sheets.tsv')):
continue
session = json.loads((job / 'private-desktop.json').read_text(encoding='utf-8'))
if (session.get('state') != 'released' or not session.get('clean')
or (job / 'worker-output.txt').read_text(encoding='utf-8').strip()):
continue
with (job / 'command.tsv').open(encoding='utf-8-sig', newline='') as stream:
project = Path(next(csv.DictReader(stream, delimiter='\t'))['project_path'])
parts = frozenset(_decode_vbs_field(row) for row in
(job / 'inventory-request.tsv').read_text(encoding='utf-8-sig').splitlines()[1:] if row)
sheets = _read_discovered_sheets(job / 'sheets.tsv', project)
inventory = read_component_inventory(job / 'components.tsv', project, sheets)
candidates[(str(project), parts)].append((job, sheets, inventory))
verified, evidence = {}, {}
for (project, parts), reads in candidates.items():
reads.sort(key=lambda item: (item[0] / 'components.tsv').stat().st_mtime_ns)
if len(reads) >= 2:
first, second = reads[-2:]
if set(first[1]) == set(second[1]) and frozenset(first[2]) == frozenset(second[2]):
verified[project] = second[2]
evidence[project] = {'requested_parts': sorted(parts), 'equal_successful_reads':
[str(first[0].resolve()), str(second[0].resolve())]}
source = converted_dir / 'CBB_转换模板_已匹配工程.xlsx'
workbook = load_workbook(source)
comments = {i: workbook.active.cell(i, 4).comment.text if workbook.active.cell(i, 4).comment else ''
for i in range(2, workbook.active.max_row + 1)}
workbook.close()
accepted, audit = [], []
for number, record in enumerate(read_bom_records(source), 2):
owners = [path for path in verified if Path(path).parent.name == Path(record.project_path).parent.name]
owner = owners[0] if len(owners) == 1 else None
inventory = verified[owner] if owner else ()
matches = {ref: sum(row.refdes.casefold() == ref.casefold() and
row.old_part_number.casefold() == record.old.casefold() for row in inventory)
for ref in record.refs}
covered = owner is not None and record.old in evidence[owner]['requested_parts']
ok = covered and all(count == 1 for count in matches.values())
source_row = re.search(r'来源:结果!(\d+)', comments[number])
audit.append({'template_row': number, 'original_row': int(source_row[1]) if source_row else '',
'project': Path(record.project_path).parent.name, 'old': record.old, 'new': record.new,
'enabled': record.enabled, 'native_inventory_verified': covered,
'matched': ok, 'refdes_matches': matches})
if ok:
accepted.append({'project_name': Path(record.project_path).parent.name,
'project_path': record.project_path, 'old_part_number': record.old,
'new_part_number': record.new, 'refdes': ','.join(record.refs),
'enabled': record.enabled, 'sync_nc': record.sync_nc, 'custom_value': record.custom})
file = write_bom_template(output / 'CBB_真实位号匹配行.xlsx', accepted)
with (output / 'CBB_真实位号核对.csv').open('w', encoding='utf-8-sig', newline='') as stream:
writer = csv.writer(stream)
writer.writerow(('模板行', '原始结果表行', '工程', '旧料号', '新料号', '启用', '已验证原生采集', '整行匹配', '各位号匹配数'))
for row in audit:
writer.writerow((*[row[key] for key in ('template_row', 'original_row', 'project', 'old', 'new',
'enabled', 'native_inventory_verified', 'matched')],
json.dumps(row['refdes_matches'], ensure_ascii=False)))
def save(name, data):
(output / name).write_text(json.dumps(data, ensure_ascii=False, indent=2, default=str), encoding='utf-8')
save('inventory.json', {path: [asdict(row) for row in rows] for path, rows in verified.items()})
save('evidence.json', evidence)
save('matching-problems.json', [row for row in audit if not row['matched']])
save('conversion-report.json', {'mapped_file': str(file.resolve()),
'full_file': str((converted_dir / 'CBB_转换模板_全量待核对.xlsx').resolve())})
summary = {'candidate_rows': len(audit), 'matched_rows': len(accepted),
'matched_projects': len({row['project_path'] for row in accepted}),
'enabled_rules': sum(row['enabled'] for row in accepted),
'verified_native_projects': len(verified),
'native_project_components': {Path(path).parent.name: len(rows) for path, rows in verified.items()}}
save('summary.json', summary)
return summary
if __name__ == '__main__':
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--native-run', type=Path, action='append', required=True)
parser.add_argument('--converted-dir', type=Path, required=True)
parser.add_argument('--output', type=Path, required=True)
args = parser.parse_args()
print(json.dumps(extract(args.native_run, args.converted_dir, args.output), ensure_ascii=False, indent=2))