# -*- coding: utf-8 -*-
import openpyxl, re
from copy import copy
from openpyxl.styles import PatternFill

FN='국민건강보험공단.xlsx'
BLUE=PatternFill(fill_type='solid', fgColor='FFBDD7EE')
H='https://www.nhis.or.kr'
def pg(label, path): return (label, H+path, '페이지', 1, '어문', '미부착')
def bd(label, path, m, n='어문'): return (label, H+path, '게시판', m, n, '미부착')

# (parent_url_substr, level 'G'|'F', tabs[first=self,...])
GROUPS=[
 ('wbhaca04500m01','G',[pg('일반건강검진','/nhis/healthin/wbhaca04500m01.do'),pg('암검진','/nhis/healthin/wbhaca04600m01.do'),
    pg('의료급여생애 전환기검진','/nhis/healthin/wbhaca04700m01.do'),pg('영유아건강검진','/nhis/healthin/wbhaca04800m01.do'),
    pg('학생건강검진','/nhis/healthin/wbhaca04900m01.do'),pg('학교 밖 청소년 건강검진','/nhis/healthin/wbhaca05200m01.do')]),
 ('retrieveDiseVltGnlSymp','G',[pg('일반증상','/nhis/healthin/retrieveDiseVltGnlSymp.do'),pg('어린이','/nhis/healthin/retrieveDiseVltChild.do'),
    pg('남성','/nhis/healthin/retrieveDiseVltMan.do'),pg('여성','/nhis/healthin/retrieveDiseVltWoman.do')]),
 ('wbhace05600m01','G',[bd('건강검진','/nhis/healthin/wbhace05600m01.do',4,'어문,이미지,영상'),
    bd('만성질환','/nhis/healthin/wbhace05800m01.do',1,'어문,이미지,영상'),
    bd('운동 및 건강강좌','/nhis/healthin/wbhace06800m01.do',8,'어문,이미지,영상')]),
 ('wbhaec14100m01','F',[pg('강당','/nhis/together/wbhaec14100m01.do'),pg('회의실','/nhis/together/wbhaec14200m01.do'),
    pg('체육시설','/nhis/together/wbhaec14300m01.do'),pg('주차장','/nhis/together/wbhaec14400m01.do'),
    pg('문화시설','/nhis/together/wbhaec14500m01.do')]),
 ('wbhabe08100m01','F',[pg('자세히 알아보기','/nhis/policy/wbhabe08100m01.do'),pg('대상여부 확인','/nhis/policy/wbhabe08200m01.do'),
    pg('적법요건','/nhis/policy/wbhabe08200m03.do')]),
 ('wbhaea02200m01','G',[bd('본부','/nhis/together/wbhaea02200m01.do',6648),bd('지역본부·지사','/nhis/together/wbhaea02210m01.do',437)]),
]

wb=openpyxl.load_workbook(FN); ws=wb.active
MAXC=27
DATA_END=159  # last possible data row to clear up to (use max_row)

# effective D,E,F via merges (F also merged in this sheet? F merges exist per ranges)
eff={4:{},5:{},6:{}}
for col in (4,5,6):
    for r in range(3,161): eff[col][r]=ws.cell(r,col).value
    for mr in ws.merged_cells.ranges:
        if mr.min_col<=col<=mr.max_col and mr.min_row>=3:
            top=ws.cell(mr.min_row,col).value
            for r in range(mr.min_row,mr.max_row+1): eff[col][r]=top

data_rows=[r for r in range(3,161) if ws.cell(r,2).value is not None]
recs=[]
for r in data_rows:
    rec={'src':r,'new':False,'mark':set()}
    rec['D']=eff[4][r]; rec['E']=eff[5][r]; rec['F']=eff[6][r]
    for c in range(7,MAXC+1): rec[c]=ws.cell(r,c).value
    rec['url']=ws.cell(r,11).value
    hl=ws.cell(r,11).hyperlink; rec['hlink']=hl.target if hl else rec['url']
    recs.append(rec)

def midx(sub):
    for i,rc in enumerate(recs):
        if rc['url'] and sub in str(rc['url']): return i
    return None

# snapshot styles before edit
stylesnap={}
for r in range(3,161):
    for c in range(1,MAXC+1): stylesnap[(r,c)]=copy(ws.cell(r,c)._style)

out=[]
applied=[]
i=0
# process in original order; build mapping parent substr->group
gmap={g[0]:g for g in GROUPS}
for rc in recs:
    g=None
    for sub,grp in gmap.items():
        if rc['url'] and sub in str(rc['url']):
            g=grp; break
    if not g:
        out.append(rc); continue
    sub,level,tabs=g
    first=tabs[0]
    # reuse parent for first tab
    if level=='G':
        rc[7]=first[0]
    else: # F
        rc['F']=first[0]; rc[7]=None
    rc['url']=first[1]; rc['hlink']=first[1]; rc[11]=first[1]
    rc[12]=first[2]; rc[13]=first[3]; rc[14]=first[4]; rc[15]=first[5]
    rc['mark'].update([6,7,11,12,13,14,15])
    out.append(rc)
    for t in tabs[1:]:
        ch={'src':rc['src'],'new':True,'mark':set(range(4,16))}
        ch['D']=rc['D']; ch['E']=rc['E']
        if level=='G':
            ch['F']=rc['F']; ch[7]=t[0]
        else:
            ch['F']=t[0]; ch[7]=None
        for c in range(8,MAXC+1): ch[c]=None
        ch[11]=t[1]; ch['url']=t[1]; ch['hlink']=t[1]
        ch[12]=t[2]; ch[13]=t[3]; ch[14]=t[4]; ch[15]=t[5]
        out.append(ch)
    applied.append((sub,len(tabs)))
recs=out
print('applied groups:', applied, 'final rows:', len(recs))

# WRITE
for mr in list(ws.merged_cells.ranges):
    if mr.min_row>=3: ws.unmerge_cells(str(mr))
SITE='국민건강보험공단'
n=len(recs)
for i,rc in enumerate(recs):
    r=3+i; src=rc['src']
    for c in range(1,MAXC+1): ws.cell(r,c)._style=copy(stylesnap[(src,c)])
    ws.cell(r,1).value=None
    ws.cell(r,2).value=i+1
    ws.cell(r,3).value=SITE
    ws.cell(r,4).value=rc['D']; ws.cell(r,5).value=rc['E']; ws.cell(r,6).value=rc['F']
    for c in range(7,MAXC+1): ws.cell(r,c).value=rc.get(c)
    kc=ws.cell(r,11); kc.value=rc['url']; kc.hyperlink=rc['hlink'] if rc['url'] else None
    for c in rc['mark']: ws.cell(r,c).fill=BLUE
    ws.row_dimensions[r].height=15
for r in range(3+n,161):
    for c in range(1,MAXC+1):
        cell=ws.cell(r,c); cell.value=None; cell.hyperlink=None; cell.fill=PatternFill(fill_type=None)

def remerge(col, scope):
    r=3
    while r<3+n:
        v=ws.cell(r,col).value
        if v is None or v=='': r+=1; continue
        r2=r
        while r2+1<3+n:
            if ws.cell(r2+1,col).value!=v: break
            ok=True
            for sc in scope:
                if ws.cell(r2+1,sc).value!=ws.cell(r,sc).value: ok=False;break
            if not ok: break
            r2+=1
        if r2>r:
            for rr in range(r+1,r2+1): ws.cell(rr,col).value=None
            ws.merge_cells(start_row=r,start_column=col,end_row=r2,end_column=col)
        r=r2+1
remerge(6,[4,5]); remerge(5,[4]); remerge(4,[])
wb.save(FN)
print('saved')