Skip to content

armgit / peach

Public Imported from GitHub
0

Peach: Local-first video metadata scraper and media manager.

refactor(dmm): 片单翻页拆出逐页与逐行解析

longmeidaof9ea2e0fc42c
diff --git a/src/peach/sources/dmm.py b/src/peach/sources/dmm.py
index d06fdbebf..829da1e9e 100644
--- a/src/peach/sources/dmm.py
+++ b/src/peach/sources/dmm.py
@@ -166,6 +166,38 @@ def release_date(value) -> str:
return moment.astimezone(JST).date().isoformat()
+def listed_page(data: dict) -> tuple[list[dict], int, bool]:
+ """片单一页的(有 cid 的行,接口自报总部数,还有没有下一页)。形状对不上归 `parse_error`。"""
+ result = (data.get("legacySearchPPV") or {}).get("result")
+ contents = result.get("contents") if isinstance(result, dict) else None
+ if not isinstance(contents, list):
+ raise SourceFailure(FailureReason.PARSE_ERROR, "DMM 返回的不是片单 JSON")
+ info = result.get("pageInfo") if isinstance(result.get("pageInfo"), dict) else {}
+ total = info["totalCount"] if isinstance(info.get("totalCount"), int) else 0
+ rows = [row for row in contents if isinstance(row, dict) and row.get("id")]
+ return rows, total, bool(info.get("hasNext"))
+
+
+def _cast(row: dict) -> list[dict]:
+ return [item for item in row.get("actresses") or [] if isinstance(item, dict)]
+
+
+def listed_work(row: dict, maker_code: str) -> ListedWork:
+ """片单里的一行。发售日用 `deliveryStartAt`,预购的是预定配信日。"""
+ return ListedWork(
+ cid=str(row["id"]), title=str(row.get("title") or "").strip(), maker_code=maker_code,
+ release_date=release_date(row.get("deliveryStartAt")),
+ pre_order=row.get("releaseStatus") == PRE_ORDER,
+ performers=tuple(str(item["name"]) for item in _cast(row) if item.get("name")),
+ studio=str((row.get("maker") or {}).get("name") or ""))
+
+
+def actress_name(rows, actress_id: str) -> str:
+ """这个编号在出演表里挂的名字:订阅源列表靠它分清同一个人的几个编号。"""
+ return next((str(item.get("name") or "") for row in rows for item in _cast(row)
+ if str(item.get("id") or "") == actress_id), "")
+
+
class DmmSource(SiteSource):
DEFAULT = DMM
@@ -249,28 +281,14 @@ class DmmSource(SiteSource):
data = self.graphql(ACTRESS_WORKS_QUERY, {
"limit": ACTRESS_PAGE, "offset": offset, "sort": "RELEASE_DATE",
"filter": {"actressIds": {"ids": [{"id": str(actress_id)}]}}}, session=session)
- result = (data.get("legacySearchPPV") or {}).get("result")
- contents = result.get("contents") if isinstance(result, dict) else None
- if not isinstance(contents, list):
- raise SourceFailure(FailureReason.PARSE_ERROR, "DMM 返回的不是片单 JSON")
- info = result.get("pageInfo") if isinstance(result.get("pageInfo"), dict) else {}
- total = info["totalCount"] if isinstance(info.get("totalCount"), int) else 0
- rows = [row for row in contents if isinstance(row, dict) and row.get("id")][:limit - offset]
+ rows, total, has_next = listed_page(data)
+ rows = rows[:limit - offset]
fresh = [str(row["id"]) for row in rows if str(row["id"]) not in known]
codes = self.maker_codes(fresh, session=session)
- for row in rows:
- cid = str(row["id"])
- cast = [item for item in row.get("actresses") or [] if isinstance(item, dict)]
- name = name or next((str(item.get("name") or "") for item in cast
- if str(item.get("id") or "") == str(actress_id)), "")
- works.append(ListedWork(
- cid=cid, title=str(row.get("title") or "").strip(), maker_code=codes.get(cid, ""),
- release_date=release_date(row.get("deliveryStartAt")),
- pre_order=row.get("releaseStatus") == PRE_ORDER,
- performers=tuple(str(item["name"]) for item in cast if item.get("name")),
- studio=str((row.get("maker") or {}).get("name") or "")))
+ works.extend(listed_work(row, codes.get(str(row["id"]), "")) for row in rows)
+ name = name or actress_name(rows, str(actress_id))
offset += len(rows)
- if not info.get("hasNext") or not rows or len(fresh) < len(rows):
+ if not has_next or not rows or len(fresh) < len(rows):
break
return ActressWorks(name=name, total=total, works=tuple(works))