df16c98366
공공기관2 작업 중. _temp 몽타주(재생성가능)는 제외. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
58 lines
2.3 KiB
Python
58 lines
2.3 KiB
Python
import sys, io, json
|
|
sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8')
|
|
import openpyxl, urllib.request, ssl, re
|
|
from concurrent.futures import ThreadPoolExecutor
|
|
|
|
P = r'D:\01.프로젝트\DB수집\작업파일\광역_사이트맵\충청남도\10.아산시\충청남도_아산시.xlsx'
|
|
wb = openpyxl.load_workbook(P); ws = wb.active
|
|
ctx = ssl.create_default_context(); ctx.check_hostname=False; ctx.verify_mode=ssl.CERT_NONE
|
|
hdr={'User-Agent':'Mozilla/5.0'}
|
|
def fetch(url):
|
|
req=urllib.request.Request(url, headers=hdr)
|
|
with urllib.request.urlopen(req, context=ctx, timeout=25) as r:
|
|
return r.read().decode('utf-8','replace')
|
|
|
|
targets=[]
|
|
for r in range(37, ws.max_row+1):
|
|
k=ws[f'K{r}'].value; l=ws[f'L{r}'].value; o=ws[f'O{r}'].value
|
|
if not k or 'asan.go.kr' not in str(k) or l=='사이트': continue
|
|
targets.append((r,str(k),l,o))
|
|
|
|
def work(t):
|
|
r,k,l,o=t
|
|
try: html=fetch(k)
|
|
except Exception as e: return (r,k,l,o,'ERR',None,None)
|
|
mt=re.search(r'img_opentype_(\d)\.gif', html)
|
|
otype=mt.group(1) if mt else None
|
|
# broad: any <ul>/<div> nav containing >=3 anchors with &pg= and text labels (not pure numbers)
|
|
tabs=None
|
|
for um in re.finditer(r'<(ul|div)([^>]*)>(.*?)</\1>', html, re.S):
|
|
attrs=um.group(2); body=um.group(3)
|
|
anchors=re.findall(r'<a[^>]*href="([^"]*&pg=\d+[^"]*)"[^>]*>(.*?)</a>', body, re.S)
|
|
labels=[(h, re.sub(r'<[^>]+>','',t).strip()) for h,t in anchors]
|
|
named=[x for x in labels if x[1] and not x[1].isdigit()]
|
|
if len(named)>=3:
|
|
cls=re.search(r'class="([^"]*)"', attrs)
|
|
tabs={'cls':cls.group(1) if cls else '', 'items':named}
|
|
break
|
|
return (r,k,l,o,None,otype,tabs)
|
|
|
|
res=[]
|
|
with ThreadPoolExecutor(max_workers=10) as ex:
|
|
for out in ex.map(work, targets): res.append(out)
|
|
|
|
print('=== in-page pg= tab navs (broad) ===')
|
|
seen=set()
|
|
for r,k,l,o,err,otype,tabs in sorted(res):
|
|
if tabs:
|
|
# base no
|
|
base=re.search(r'no=(\d+)', k)
|
|
print(f'r{r}: cls={tabs["cls"]!r} n={len(tabs["items"])} L={l} no={base.group(1) if base else "?"} K={k}')
|
|
|
|
print('\n=== any opentype marks (should be only the ones already typed posts) ===')
|
|
for r,k,l,o,err,otype,tabs in sorted(res):
|
|
if otype:
|
|
print(f'r{r}: TYPE={otype} O={o!r} L={l} K={k}')
|
|
|
|
print('\nerrors:', [r for r,k,l,o,err,otype,tabs in res if err])
|