kangshifu-board/scripts/package-industry-recovery.py

"""Freeze historical recovery additions on sealed r9; never touch a deployed service."""
import argparse
from contextlib import closing
from datetime import datetime, timezone
import hashlib
import json
from pathlib import Path
import shutil
import sqlite3
import zipfile

ROOT = Path(__file__).resolve().parents[1]
BASE = ROOT / 'artifacts/ksf-industry-monitor-20261107-r9/package'
WORK = ROOT / 'work/industry-mainline-continuation-20261107-r1'
RELEASE = 'ksf-industry-monitor-20261107-r10'
SOURCES = ['src/pages/industry-pipeline-view.js', 'server/voc-library/monitoring_ledger.py',
           'server/voc-library/monitoring_dispatch.py', 'server/voc-library/test_monitoring_ledger.py',
           'scripts/manage-industry-dispatch.py', 'scripts/prepare-industry-handoff.py',
           'scripts/audit-industry-archived-details.py', 'scripts/reuse-industry-archived-details.py',
           'scripts/verify-industry-monitor.py', 'scripts/verify-industry-recovery-release.py',
           'scripts/verify-industry-recovered-data.py', 'scripts/package-industry-recovery.py']
DOCUMENTS = ['agent-queries.md', 'deployment.md', 'request-guard.md', '主线待决策清单.md',
             'r10本地实现与未完成条件.md']


def sha(raw): return hashlib.sha256(raw).hexdigest()


def save(path, value):
    path.parent.mkdir(parents=True, exist_ok=True)
    path.write_text(json.dumps(value, ensure_ascii=False, indent=2) + '\n', 'utf-8')


def build(out):
    if out.exists(): raise ValueError('RELEASE_DIRECTORY_MUST_BE_NEW')
    for line in (BASE/'FILES.sha256').read_text('utf-8').splitlines():
        digest, name = line.split('  ', 1)
        if sha((BASE/name).read_bytes()) != digest: raise ValueError('FROZEN_R9_CHANGED')
    captured = {name:(ROOT/name).read_bytes() for name in SOURCES}
    shutil.copytree(BASE, out)
    for name, raw in captured.items():
        target = out/('payload/'+name if name.startswith('src/') else name)
        target.parent.mkdir(parents=True, exist_ok=True); target.write_bytes(raw)
    if any((ROOT/name).read_bytes() != raw for name, raw in captured.items()):
        raise ValueError('SOURCE_CHANGED_DURING_FREEZE')
    save(out/'evidence/r10-source-capture.json', {'release':RELEASE,
         'sourceFilesSha256':{name:sha(raw) for name, raw in captured.items()},
         'viewIndustrySource':'sealed r9; concurrent workspace innovation edits excluded'})
    store = ROOT/'work/industry-monitoring-20261107-r1/monitoring-registry'
    with closing(sqlite3.connect((store/'monitoring.sqlite').resolve().as_uri()+'?mode=ro', uri=True)) as source:
        with closing(sqlite3.connect(out/'data/monitoring-registry/monitoring.sqlite')) as destination:
            source.backup(destination)
            destination.execute("UPDATE pipeline_workers SET state='ARCHIVED_COPY'")
            destination.execute("UPDATE dispatch_audits SET state='ARCHIVED_COPY'"); destination.commit()
    for file in (store/'assets').iterdir():
        if sha(file.read_bytes()) != file.stem: raise ValueError('AUXILIARY_ASSET_CHANGED')
        target = out/'data/monitoring-registry/assets'/file.name
        if not target.exists(): shutil.copyfile(file, target)
    # Replace only the bounded scope; preserve all sealed source evidence and receipts.
    for file in (WORK/'handoff').iterdir():
        if file.is_file(): shutil.copyfile(file, out/'handoff'/file.name)
    target = out/'evidence/r10-recovery'; target.mkdir()
    for name in ['ledger-import.json', 'detail-reuse-receipt.json', 'detail-reuse-replay.json', 'connector-capabilities.json']:
        shutil.copyfile(WORK/name, target/name)
    inventory = json.loads((WORK/'archived-details/inventory.json').read_bytes())
    save(target/'archive-inventory-summary.json', {k:v for k,v in inventory.items()
         if k not in ('details','unboundDetailCandidates','rankingEndpointCandidates')})
    save(target/'public-access-status.json', {'rankingUrlSource':'verified historical JD detail rankInfo h5JumpUrl',
         'ordinaryFreshBrowserTitle':'京东登录注册','loginRequired':True,'loginAttempted':False,
         'challengeHandling':False,'completeTop100Or400':0,'newOfficialMetadataVerified':0,
         'paidCalls':0,'privateBrowserDiagnosticsIncluded':False})
    release = json.loads((out/'RELEASE.json').read_bytes())
    release.update(release=RELEASE,createdAt=datetime.now(timezone.utc).isoformat(),
                   builtFromFrozenRelease='ksf-industry-monitor-20261107-r9',
                   settledCloseoutIdentifiedRows=154,parameterBoundSettledRows=139,
                   unboundSettledParameterRows=15,serverLedgerRefetched=False,
                   detailObservations=52,productsWithDetailText=13,primaryDetailsWithResponse=41,
                   primaryDetailsWithText=12,nativePython39Verified=False,
                   overallComplete=False,productionDeployed=False,paidCalls=0)
    release['backendFiles'].append('server/voc-library/monitoring_ledger.py')
    save(out/'RELEASE.json',release)
    mpath = out/'deploy/industry-ui-files.json'; manifest = json.loads(mpath.read_bytes()); manifest['release']=RELEASE
    for entry in manifest['files']:
        old = sha((BASE/'payload'/entry['path']).read_bytes().replace(b'\r\n',b'\n'))
        if old not in entry['priorNormalizedSha256']: entry['priorNormalizedSha256'].append(old)
        entry['sha256']=sha((out/'payload'/entry['path']).read_bytes().replace(b'\r\n',b'\n'))
    save(mpath,manifest)
    publisher=out/'scripts/publish-industry-ui.mjs'
    publisher.write_text(publisher.read_text('utf-8').replace('ksf-industry-monitor-20261107-r9',RELEASE),'utf-8')
    for file in (out/'deploy').iterdir():
        if file.suffix=='.example':
            file.write_text(file.read_text('utf-8').replace('ksf-industry-monitor-20261107-r9',RELEASE),'utf-8')
    for name in DOCUMENTS: shutil.copyfile(ROOT/'docs/industry-monitoring-20261107'/name,out/'docs'/name)
    (out/'docs/README.md').write_text('''# 方便面行业 VOC 后台 r10 本地验收版

当前版本基于封存r9，补齐154行附件账本身份对齐、139行商品页防重核对，以及历史详情响应复用。14个前端模块、16个独立后端文件。

主店165件中41件有详情响应、12件有商品文字卖点；全库52/13。历史观察不冒充新采集，宣称不冒充实证。政策8类用法和创新工作区沿用已验版本。当前细节见r10本地实现与未完成条件.md。

本包没有活动研究库、私有详情原件、完整账本ZIP、登录页诊断或凭据。历史文档/回执仅保留出处，当前执行入口和数量以本页、RELEASE与implementation-receipt-v10.json为准。

当前未部署生产、未恢复付费采集、未转发。全量当前目录/评价、完整Top100/400与跨期监听、16份官方状态及生产接通仍有真实外部缺口，overallComplete=false。封存r5至r9不变。
''','utf-8')
    print(json.dumps({'ok':True,'candidate':str(out),'staticFiles':14,'backendFiles':16,'sealed':False}))


def seal(out):
    receipt=json.loads((out/'evidence/implementation-receipt-v10.json').read_bytes())
    if not receipt['accepted'] or not receipt['packageVerificationPassed']: raise ValueError('ACCEPTANCE_REQUIRED')
    for name, digest in receipt['testedFilesSha256'].items():
        if sha((out/name).read_bytes())!=digest: raise ValueError('TESTED_FILE_CHANGED')
    archive=out.parent/(RELEASE+'.zip')
    if archive.exists(): raise ValueError('ARCHIVE_EXISTS')
    files=sorted(f for f in out.rglob('*') if f.is_file() and f!=out/'FILES.sha256')
    (out/'FILES.sha256').write_text(''.join(sha(f.read_bytes())+'  '+f.relative_to(out).as_posix()+'\n' for f in files),'utf-8')
    with zipfile.ZipFile(archive,'x',zipfile.ZIP_DEFLATED) as z:
        for f in sorted(out.rglob('*')):
            if f.is_file(): z.write(f,f.relative_to(out).as_posix())
    print(json.dumps({'ok':True,'archive':str(archive),'bytes':archive.stat().st_size,
                     'sha256':sha(archive.read_bytes()),'productionDeployed':False,'paidCalls':0}))


if __name__=='__main__':
    p=argparse.ArgumentParser(description=__doc__);p.add_argument('--out',type=Path,required=True);p.add_argument('--seal',action='store_true')
    a=p.parse_args();(seal if a.seal else build)(a.out.resolve())