codebee 0.1.103 → 0.1.105

package/app/core/publish/chaprows.py

@@ -1,138 +1,138 @@
-# -*- coding: utf-8 -*-
-"""章节管理页的行收集与分桶（校准的数据面，manager 只做编排）。
-
-从 manager.py 拆出（manager 超 1200 行行数基线，2026-10-08）：收集器只吃
-page + 平台 CONFIG，不依赖任何内部模块——放 L0 层无依赖。
-状态关键词口径与台账一致：已发布 / 待审核·审核中·排队 / 未通过·驳回。
-"""
-from __future__ import annotations
-
-import re
-import time
-
-_STATUS_PUB = "已发布"
-_STATUS_REVIEW = ("待审核", "审核中", "排队")
-_STATUS_BAD = ("未通过", "驳回")
-
-_CH_NO_RE = re.compile(r"第\s*(\d+)\s*章")
-
-
-def bucket_chapter_rows(rows):
-    """章节管理页的行文本 → {total, published, review, rejected, published_nos}。
-
-    行里含状态关键词才计入对应桶；total=识别出的章节数据行数。
-    published_nos=已发布行里解析出的章号（第N章，阿拉伯数字）——本地台账
-    只记得 CodeBee 自己发的章，用户手工补交的只有平台知道，校准把它带
-    回来供幂等/待发清单对账。"""
-    out = {"total": 0, "published": 0, "review": 0, "rejected": 0,
-           "published_nos": []}
-    nos = set()
-    for t in rows or []:
-        t = str(t)
-        if not t:
-            continue
-        if "章节名称" in t:                     # 表头兜底（正常到不了这）
-            continue
-        out["total"] += 1
-        if any(k in t for k in _STATUS_BAD):
-            out["rejected"] += 1
-        elif _STATUS_PUB in t:
-            out["published"] += 1
-            m = _CH_NO_RE.search(t)
-            if m:
-                nos.add(int(m.group(1)))
-        elif any(k in t for k in _STATUS_REVIEW):
-            out["review"] += 1
-    out["published_nos"] = sorted(nos)
-    return out
-
-
-def collect_manage_rows(page, mod):
-    """读章节管理页的全部行：跨页翻页收齐（番茄表每页 15 行，只数当前页
-    会把 30 章已发报成 15——2026-10-08 校准少计实案）。平台没配翻页 JS
-    （七猫）或翻页无进展时按单页收。行文本去重保序，重复页不会翻倍。"""
-    js_rows = mod.CONFIG.get("count_rows_js")
-    next_js = mod.CONFIG.get("pager_next_js")
-    settle = float(mod.CONFIG.get("count_page_settle") or 1.2)
-    rows, seen = [], set()
-    for _ in range(int(mod.CONFIG.get("count_max_pages") or 20)):
-        fresh = 0
-        for t in (page.call(js_rows) or []):
-            t = str(t)
-            if t and t not in seen:
-                seen.add(t)
-                rows.append(t)
-                fresh += 1
-        if not next_js or fresh == 0:
-            break
-        r = page.call(next_js)
-        if not isinstance(r, dict) or r.get("done"):
-            break
-        time.sleep(settle)
-    return rows
-
-
-def collect_all_volumes(page, mod):
-    """跨分卷收行：章节管理页带分卷筛选（番茄「第一卷：…」下拉，无「全部」
-    项），列表只显示选中卷的章节——卷二发布后只数卷一会继续少计。弹层在
-    click 后下一帧才渲染，open→读/点之间必须隔 sleep（同步 JS 一口气查必
-    落空，2026-10-08 真机实案）。收完恢复进入时的卷；一卷都没选成时按原
-    筛选兜底收一次，不让分卷自动化失败拖死整次校准。
-
-    返回 (rows, vols)：vols=读到的平台分卷显示名清单（读不到为 None），
-    供 books.json 缓存平台卷表。"""
-    vj = mod.CONFIG.get("volume_js") or {}
-    if not all(vj.get(k) for k in ("open", "options", "pick")):
-        return collect_manage_rows(page, mod), None
-    settle = float(mod.CONFIG.get("count_page_settle") or 1.2)
-
-    def open_popup():
-        try:
-            page.call(vj["open"])
-        except Exception:
-            pass
-        time.sleep(0.8)
-
-    orig = ""
-    try:
-        orig = str(page.call(vj.get("current")) or "").strip()
-    except Exception:
-        pass
-    open_popup()
-    try:
-        vols = page.call(vj["options"]) or []
-    except Exception:
-        vols = []
-    if not isinstance(vols, list) or len(vols) < 2:
-        return collect_manage_rows(page, mod), None   # 单卷：当前筛选就是全量
-    rows, seen, picked_any, misses = [], set(), False, 0
-    for v in vols:
-        v = str(v or "").strip()
-        if not v:
-            continue
-        open_popup()
-        try:
-            r = page.call(vj["pick"], v)
-        except Exception:
-            r = None
-        if not isinstance(r, dict) or not r.get("ok"):
-            misses += 1
-            if misses >= 3:
-                break                   # 弹层压根点不动：别把选项清单烧完
-            continue
-        misses = 0
-        picked_any = True
-        time.sleep(settle)
-        for t in collect_manage_rows(page, mod):
-            if t not in seen:
-                seen.add(t)
-                rows.append(t)
-    if orig:                                    # 还原进入时的卷筛选
-        open_popup()
-        try:
-            page.call(vj["pick"], orig)
-        except Exception:
-            pass
-    if not picked_any:
-        return collect_manage_rows(page, mod), list(vols)
-    return rows, [str(v) for v in vols if str(v or "").strip()]
+# -*- coding: utf-8 -*-
+"""章节管理页的行收集与分桶（校准的数据面，manager 只做编排）。
+
+从 manager.py 拆出（manager 超 1200 行行数基线，2026-10-08）：收集器只吃
+page + 平台 CONFIG，不依赖任何内部模块——放 L0 层无依赖。
+状态关键词口径与台账一致：已发布 / 待审核·审核中·排队 / 未通过·驳回。
+"""
+from __future__ import annotations
+
+import re
+import time
+
+_STATUS_PUB = "已发布"
+_STATUS_REVIEW = ("待审核", "审核中", "排队")
+_STATUS_BAD = ("未通过", "驳回")
+
+_CH_NO_RE = re.compile(r"第\s*(\d+)\s*章")
+
+
+def bucket_chapter_rows(rows):
+    """章节管理页的行文本 → {total, published, review, rejected, published_nos}。
+
+    行里含状态关键词才计入对应桶；total=识别出的章节数据行数。
+    published_nos=已发布行里解析出的章号（第N章，阿拉伯数字）——本地台账
+    只记得 CodeBee 自己发的章，用户手工补交的只有平台知道，校准把它带
+    回来供幂等/待发清单对账。"""
+    out = {"total": 0, "published": 0, "review": 0, "rejected": 0,
+           "published_nos": []}
+    nos = set()
+    for t in rows or []:
+        t = str(t)
+        if not t:
+            continue
+        if "章节名称" in t:                     # 表头兜底（正常到不了这）
+            continue
+        out["total"] += 1
+        if any(k in t for k in _STATUS_BAD):
+            out["rejected"] += 1
+        elif _STATUS_PUB in t:
+            out["published"] += 1
+            m = _CH_NO_RE.search(t)
+            if m:
+                nos.add(int(m.group(1)))
+        elif any(k in t for k in _STATUS_REVIEW):
+            out["review"] += 1
+    out["published_nos"] = sorted(nos)
+    return out
+
+
+def collect_manage_rows(page, mod):
+    """读章节管理页的全部行：跨页翻页收齐（番茄表每页 15 行，只数当前页
+    会把 30 章已发报成 15——2026-10-08 校准少计实案）。平台没配翻页 JS
+    （七猫）或翻页无进展时按单页收。行文本去重保序，重复页不会翻倍。"""
+    js_rows = mod.CONFIG.get("count_rows_js")
+    next_js = mod.CONFIG.get("pager_next_js")
+    settle = float(mod.CONFIG.get("count_page_settle") or 1.2)
+    rows, seen = [], set()
+    for _ in range(int(mod.CONFIG.get("count_max_pages") or 20)):
+        fresh = 0
+        for t in (page.call(js_rows) or []):
+            t = str(t)
+            if t and t not in seen:
+                seen.add(t)
+                rows.append(t)
+                fresh += 1
+        if not next_js or fresh == 0:
+            break
+        r = page.call(next_js)
+        if not isinstance(r, dict) or r.get("done"):
+            break
+        time.sleep(settle)
+    return rows
+
+
+def collect_all_volumes(page, mod):
+    """跨分卷收行：章节管理页带分卷筛选（番茄「第一卷：…」下拉，无「全部」
+    项），列表只显示选中卷的章节——卷二发布后只数卷一会继续少计。弹层在
+    click 后下一帧才渲染，open→读/点之间必须隔 sleep（同步 JS 一口气查必
+    落空，2026-10-08 真机实案）。收完恢复进入时的卷；一卷都没选成时按原
+    筛选兜底收一次，不让分卷自动化失败拖死整次校准。
+
+    返回 (rows, vols)：vols=读到的平台分卷显示名清单（读不到为 None），
+    供 books.json 缓存平台卷表。"""
+    vj = mod.CONFIG.get("volume_js") or {}
+    if not all(vj.get(k) for k in ("open", "options", "pick")):
+        return collect_manage_rows(page, mod), None
+    settle = float(mod.CONFIG.get("count_page_settle") or 1.2)
+
+    def open_popup():
+        try:
+            page.call(vj["open"])
+        except Exception:
+            pass
+        time.sleep(0.8)
+
+    orig = ""
+    try:
+        orig = str(page.call(vj.get("current")) or "").strip()
+    except Exception:
+        pass
+    open_popup()
+    try:
+        vols = page.call(vj["options"]) or []
+    except Exception:
+        vols = []
+    if not isinstance(vols, list) or len(vols) < 2:
+        return collect_manage_rows(page, mod), None   # 单卷：当前筛选就是全量
+    rows, seen, picked_any, misses = [], set(), False, 0
+    for v in vols:
+        v = str(v or "").strip()
+        if not v:
+            continue
+        open_popup()
+        try:
+            r = page.call(vj["pick"], v)
+        except Exception:
+            r = None
+        if not isinstance(r, dict) or not r.get("ok"):
+            misses += 1
+            if misses >= 3:
+                break                   # 弹层压根点不动：别把选项清单烧完
+            continue
+        misses = 0
+        picked_any = True
+        time.sleep(settle)
+        for t in collect_manage_rows(page, mod):
+            if t not in seen:
+                seen.add(t)
+                rows.append(t)
+    if orig:                                    # 还原进入时的卷筛选
+        open_popup()
+        try:
+            page.call(vj["pick"], orig)
+        except Exception:
+            pass
+    if not picked_any:
+        return collect_manage_rows(page, mod), list(vols)
+    return rows, [str(v) for v in vols if str(v or "").strip()]