xiaji/diary-news
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(scripts): backfill_body 也回填翻译过但 body 短的文章(让 worker 重译)

Browse Source
This commit is contained in:
Mavis
2026-06-08 16:09:22 +08:00
parent 4cc5d41e39
commit 82a92032bb
1 changed files with 2 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
scripts/backfill_body.py
View File

@@ -52,15 +52,10 @@ async def backfill(
stats = {"scanned": 0, "matched": 0, "updated": 0, "skipped": 0, "failed": 0} stats = {"scanned": 0, "matched": 0, "updated": 0, "skipped": 0, "failed": 0}
async with AsyncSessionLocal() as session: async with AsyncSessionLocal() as session:
# 1) 找短 body 文章 # 1) 找短 body 文章(不管翻译状态 — 已翻译的也要回填,重置让 worker 用完整 body 重译)
stmt = ( stmt = select(Article)
select(Article)
.where((Article.body_text.is_(None)) | (Article.body_text == "") | (Article.translation_status != "ok"))
# 简化:只看 body 短的(无论翻译状态)
)
if source_slug: if source_slug:
stmt = stmt.join(Source, Source.id == Article.source_id).where(Source.slug == source_slug) stmt = stmt.join(Source, Source.id == Article.source_id).where(Source.slug == source_slug)
# 全部扫一次(在内存里过滤 short body,避免 SQL 长度函数复杂)
rows = (await session.execute(stmt.order_by(Article.id.desc()))).scalars() rows = (await session.execute(stmt.order_by(Article.id.desc()))).scalars()
all_articles = list(rows) all_articles = list(rows)
# 过滤 short body # 过滤 short body