df16c98366
공공기관2 작업 중. _temp 몽타주(재생성가능)는 제외. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
49 lines
1.6 KiB
Python
49 lines
1.6 KiB
Python
# -*- coding: utf-8 -*-
|
|
"""브라우저로 추출한 GNB 트리(_gnb\{num}.json)를 phase1.write_excel로 엑셀 기록.
|
|
사용: python _gnb_write.py {num} [{num} ...]
|
|
"""
|
|
import sys, io, os, json, importlib.util
|
|
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
|
|
|
|
SCRIPTS = r"D:\01.프로젝트\DB수집\_스크립트"
|
|
GNB_DIR = r"D:\01.프로젝트\DB수집\작업파일\공공기관2\_gnb"
|
|
|
|
# phase1 모듈 로드 (write_excel 재사용)
|
|
spec = importlib.util.spec_from_file_location("p1", os.path.join(SCRIPTS, "_공공기관2_phase1.py"))
|
|
p1 = importlib.util.module_from_spec(spec)
|
|
spec.loader.exec_module(p1)
|
|
|
|
COL = {1: 'D', 2: 'E', 3: 'F', 4: 'G', 5: 'H', 6: 'I', 7: 'J'}
|
|
|
|
def to_raw_rows(gnb):
|
|
stack = {}
|
|
raw = []
|
|
for it in gnb:
|
|
d = int(it['depth'])
|
|
if d > 7:
|
|
d = 7
|
|
stack[d] = (it.get('label') or '').strip()
|
|
for k in list(stack):
|
|
if k > d:
|
|
del stack[k]
|
|
row = {}
|
|
for i in range(1, d + 1):
|
|
row[COL[i]] = stack.get(i, '')
|
|
row['href'] = it.get('href') or ''
|
|
raw.append(row)
|
|
return raw
|
|
|
|
def main():
|
|
nums = sys.argv[1:]
|
|
for ns in nums:
|
|
path = os.path.join(GNB_DIR, f"{ns}.json")
|
|
if not os.path.exists(path):
|
|
print(f"[{ns}] json 없음 — 스킵"); continue
|
|
data = json.load(open(path, encoding='utf-8'))
|
|
raw = to_raw_rows(data['gnb'])
|
|
n = p1.write_excel(data['name'], int(data['num']), data['base'], raw)
|
|
print(f"[{data['num']}.{data['name']}] write_excel → {n}행")
|
|
|
|
if __name__ == '__main__':
|
|
main()
|