"""Read-only source coverage audit. Does not calculate or overwrite awards."""
from pathlib import Path
from datetime import datetime, timedelta
from collections import Counter
import json, re, warnings
import openpyxl

warnings.filterwarnings('ignore', category=UserWarning, module='openpyxl')
OUT = Path(__file__).resolve().parent
REPO = OUT.parents[1]
HROOT = Path('/tmp/mileage-hydrogen-review-0916/氢能源行驶里程')
METER = Path('/Users/shishengliang/Downloads/车辆里程查询_20260101-20260731_1024辆.xlsx')
JAN = Path('/Users/shishengliang/Code/lingniu/氢能车用氢行驶里程统计 1.17.xlsx')

def read(p, sheet=None):
    w = openpyxl.load_workbook(p, read_only=True, data_only=True)
    s = w[sheet] if sheet else w.worksheets[0]
    s.reset_dimensions()
    r = list(s.values)
    w.close()
    return r

def norm(x):
    return re.sub(r'\s+', '', str(x or '')).upper()

h = {}
bad_files = []
for p in HROOT.rglob('*.xlsx'):
    if int(p.parent.name[:-1]) > 6:
        continue
    rr = read(p)
    headers = list(rr[0])
    if '车牌号码' not in headers:
        bad_files.append(str(p))
        continue
    for row in rr[1:]:
        r = dict(zip(headers,row))
        plate = norm(r.get('车牌号码'))
        if not plate:
            continue
        d = str(r['日期范围'])[:10]
        val = r.get('氢能源行驶里程(km)')
        if val is not None and str(val).strip():
            key = (plate,d)
            assert key not in h, ('duplicate hydrogen',key)
            h[key] = float(val)
jan_status = 'not read'
try:
    rr=read(JAN)
    assert '车牌号码' in rr[0]
    for row in rr[1:]:
        r=dict(zip(rr[0],row)); plate=norm(r.get('车牌号码'))
        if plate and r.get('氢能源行驶里程(km)') is not None:
            h[(plate,str(r['日期范围'])[:10])] = float(r['氢能源行驶里程(km)'])
    jan_status='read'
except (OSError,AssertionError) as e:
    jan_status=f'{type(e).__name__}: {e}'

(OUT/'hydrogen_daily.json').write_text(json.dumps([[p,d,v] for (p,d),v in sorted(h.items())],ensure_ascii=False))
# GPS and meter acquisition superseded by the user's monthly minimum rule.
t={};meter_months=Counter()

records=[]; summary=[]
for m in range(1,7):
    rr=read(REPO/f'租赁任务考核_2026年{m}月.xlsx','业务考核视图')
    month=[]
    for rownum,row in enumerate(rr[1:],2):
        r=dict(zip(rr[0],row)); plate=norm(r.get('车牌号'))
        if not plate:
            continue
        def as_date(v):
            return v.date() if isinstance(v,datetime) else datetime.fromisoformat(str(v)).date()
        start,end=as_date(r['考核开始日期']),as_date(r['考核结束日期'])
        dates=[(start+timedelta(days=i)).isoformat() for i in range((end-start).days+1)]
        mh=[d for d in dates if (plate,d) not in h]
        mt=[d for d in dates if (plate,d) not in t]
        both=[d for d in mh if (plate,d) not in t]
        x=dict(month=m,source_row=rownum,plate=plate,sales=r.get('销售经理'),start=str(start),end=str(end),days=len(dates),hydrogen_missing_dates=mh,meter_missing_dates=mt,both_missing_dates=both,original_assessment_km=r.get('实际行驶里程(km)'),test_decision='not applicable before April' if m<4 else 'pending GPS and complete same-period meter basis')
        assert len(mh) in (0,len(dates)), 'Partial hydrogen needs a daily fallback decision'
        x['hydrogen_km'] = round(sum(h[(plate,d)] for d in dates), 2) if not mh else None
        x['candidate_non_test_km'] = x['hydrogen_km'] if not mh else x['original_assessment_km']
        x['meter_comparison_km'] = x['original_assessment_km']
        x['meter_basis'] = '用户确认历史考核源表实际里程即完整仪表区间合计'
        x['test_decision'] = '不再采用测试车辆判定；按同组月合计取小值'
        month.append(x);records.append(x)
    summary.append(dict(month=m,records=len(month),records_with_hydrogen_gaps=sum(bool(x['hydrogen_missing_dates']) for x in month),records_with_meter_gaps=sum(bool(x['meter_missing_dates']) for x in month),records_with_both_gaps=sum(bool(x['both_missing_dates']) for x in month),both_missing_record_days=sum(len(x['both_missing_dates']) for x in month)))

result=dict(jan17_status=jan_status,meter_nonblank_vehicle_days_by_month=dict(sorted(meter_months.items())),summary=summary,records=records,invalid_archive_files=bad_files)
(OUT/'source_coverage.json').write_text(json.dumps(result,ensure_ascii=False,indent=2,default=str))
print(json.dumps({'jan17_status':jan_status,'summary':summary},ensure_ascii=False,indent=2))