fix(indexer): parse PterClub seeding stats

app/modules/indexer/parser/mtorrent.py

@@ -1,15 +1,14 @@
 # -*- coding: utf-8 -*-
 import json
-from typing import Any, Optional, Tuple, cast
+from typing import Optional, Tuple
 from urllib.parse import urljoin
 
+from app.runtime.log import logger
 from app.modules.indexer.parser import SiteParserBase, SiteSchema
 from app.domain import site as site_rules
 
 
 class MTorrentSiteUserInfo(SiteParserBase):
-    """使用 M-Team API 解析用户信息和当前做种数据。"""
-
     schema = SiteSchema.MTorrent
     request_mode = "apikey"
 
@@ -114,7 +113,7 @@ class MTorrentSiteUserInfo(SiteParserBase):
 
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: Optional[bool] = False) -> Optional[str]:
         """
-        累加当前页的做种数和体积，并根据列表分页元数据继续读取。
+        解析用户做种信息
         """
         if not html_text:
             return None
@@ -145,30 +144,24 @@ class MTorrentSiteUserInfo(SiteParserBase):
         self.seeding_size += page_seeding_size
         self.seeding_info.extend(page_seeding_info)
 
-        # 用户信息解析成功后已初始化分页参数，后续页数从做种列表响应中读取。
-        torrent_params = cast(dict[str, Any], self._torrent_seeding_params)
-        page_number = int(torrent_params["pageNumber"])
-        page_size = int(torrent_params["pageSize"])
+        # 查询总做种数
+        seeder_count = 0
         try:
-            total_pages = int(seeding_data.get("totalPages") or 0)
-        except (TypeError, ValueError):
-            total_pages = 0
-        try:
-            total_torrents = int(seeding_data.get("total") or 0)
-        except (TypeError, ValueError):
-            total_torrents = 0
-
-        if total_pages:
-            has_next_page = page_number < total_pages
-        elif total_torrents:
-            has_next_page = page_number * page_size < total_torrents
-        else:
-            # 旧版响应未提供总量时，满页后再请求一次确认是否结束。
-            has_next_page = len(torrents) >= page_size
-        if not has_next_page:
+            result = self._get_page_content(
+                url=urljoin(self._base_url, "api/tracker/myPeerStatus"),
+                params={"uid": self.userid},
+            )
+            if result:
+                seeder_info = json.loads(result)
+                seeder_count = int(seeder_info.get("data", {}).get("seeder") or 0)
+        except Exception as e:
+            logger.error(f"获取做种数失败: {str(e)}")
+        if not seeder_count:
             return None
-
-        torrent_params["pageNumber"] = page_number + 1
+        if self.seeding >= seeder_count:
+            return None
+        # 还有下一页
+        self._torrent_seeding_params["pageNumber"] += 1
         return ""
 
     def _parse_message_unread_links(self, html_text: str, msg_links: list) -> Optional[str]:

app/modules/indexer/parser/nexus_php.py

@@ -220,6 +220,33 @@ class NexusPhpSiteUserInfo(SiteParserBase):
                 return True, gold * 100 * 100 + silver * 100 + copper
         return False, 0.0
 
+    def _parse_seeding_pages(self) -> None:
+        """
+        优先使用完整做种列表统计，列表为空或解析不完整时回退到用户详情页统计。
+        """
+        profile_seeding = self.seeding
+        profile_seeding_size = self.seeding_size
+        profile_seeding_info = list(self.seeding_info)
+
+        # 用户详情页可能只显示部分做种记录，先暂存，避免与完整列表相加造成重复。
+        self.seeding = 0
+        self.seeding_size = 0
+        self.seeding_info.clear()
+        try:
+            # SiteParserBase keeps this legacy hook untyped.
+            super()._parse_seeding_pages()  # type: ignore[no-untyped-call]
+        except Exception:
+            self.seeding = profile_seeding
+            self.seeding_size = profile_seeding_size
+            self.seeding_info = profile_seeding_info
+            raise
+
+        if (not self.seeding or not self.seeding_size) and (profile_seeding or profile_seeding_size):
+            self.seeding = profile_seeding
+            self.seeding_size = profile_seeding_size
+            if profile_seeding_info:
+                self.seeding_info = profile_seeding_info
+
     def _parse_user_torrent_seeding_info(self, html_text: str, multi_page: Optional[bool] = False) -> Optional[str]:
         """
         做种相关信息
@@ -259,21 +286,41 @@ class NexusPhpSiteUserInfo(SiteParserBase):
             page_seeding = 0
             page_seeding_size = 0
             page_seeding_info = []
-            # 如果 table class="torrents"，则增加table[@class="torrents"]
-            table_class = '//table[@class="torrents"]' if html.xpath('//table[@class="torrents"]') else ''
-            seeding_sizes = html.xpath(f'{table_class}//tr[position()>1]/td[{size_col}]')
-            seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]/b/a/text()')
-            if not seeding_seeders:
-                seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]//text()')
-            if seeding_sizes and seeding_seeders:
-                page_seeding = len(seeding_sizes)
-
-                for i in range(0, len(seeding_sizes)):
-                    size = self.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                    seeders = text_tools.parse_int(seeding_seeders[i])
-
+            site_domain = self._site_domain.lower().removeprefix("www.")
+            if site_domain in {"pterclub.net", "pterclub.com"}:
+                # 猫站完整做种列表按 td[2]/td[4]/td[5] 展示种子、体积和做种人数。
+                pterclub_rows = html.xpath(
+                    '//*[@id="outer"]/table/tbody/tr | //*[@id="outer"]/table/tr'
+                    '|//table[contains(concat(" ", normalize-space(@class), " "), " torrents ")]/tbody/tr'
+                    '|//table[contains(concat(" ", normalize-space(@class), " "), " torrents ")]/tr'
+                )
+                for row in pterclub_rows:
+                    columns = row.xpath('./td | ./th')
+                    if len(columns) < 5 or not columns[1].xpath('.//a'):
+                        continue
+                    size = self.num_filesize(columns[3].xpath("string(.)").strip())
+                    if not size:
+                        continue
+                    seeders = text_tools.parse_int(columns[4].xpath("string(.)").strip())
+                    page_seeding += 1
                     page_seeding_size += size
                     page_seeding_info.append([seeders, size])
+            else:
+                # 如果 table class="torrents"，则限制在种子表中，避免解析页面其他表格。
+                table_class = '//table[@class="torrents"]' if html.xpath('//table[@class="torrents"]') else ''
+                seeding_sizes = html.xpath(f'{table_class}//tr[position()>1]/td[{size_col}]')
+                seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]/b/a/text()')
+                if not seeding_seeders:
+                    seeding_seeders = html.xpath(f'{table_class}//tr[position()>1]/td[{seeders_col}]//text()')
+                if seeding_sizes and seeding_seeders:
+                    page_seeding = len(seeding_sizes)
+
+                    for i in range(0, len(seeding_sizes)):
+                        size = self.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
+                        seeders = text_tools.parse_int(seeding_seeders[i])
+
+                        page_seeding_size += size
+                        page_seeding_info.append([seeders, size])
 
             self.seeding += page_seeding
             self.seeding_size += page_seeding_size
@@ -345,19 +392,29 @@ class NexusPhpSiteUserInfo(SiteParserBase):
 
             # 做种体积 & 做种数
             # seeding 页面获取不到的话，此处再获取一次
-            seeding_sizes = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
-                                       'table[tr[1][td[4 and text()="尺寸"]]]//tr[position()>1]/td[4]')
-            seeding_seeders = html.xpath('//tr/td[text()="当前上传"]/following-sibling::td[1]//'
-                                         'table[tr[1][td[5 and text()="做种者"]]]//tr[position()>1]/td[5]//text()')
-            tmp_seeding = len(seeding_sizes)
+            seeding_tables = html.xpath(
+                '//tr/td[normalize-space(.)="当前上传" or normalize-space(.)="当前做种"]'
+                '/following-sibling::td[1]//table'
+            )
+            tmp_seeding = 0
             tmp_seeding_size = 0
             tmp_seeding_info = []
-            for i in range(0, len(seeding_sizes)):
-                size = self.num_filesize(seeding_sizes[i].xpath("string(.)").strip())
-                seeders = text_tools.parse_int(seeding_seeders[i])
+            for seeding_table in seeding_tables:
+                # 猫站使用“当前做种”作为面板行名；该表第 4、5 列分别是体积和做种人数。
+                seeding_rows = seeding_table.xpath('./tr | ./tbody/tr')
+                for seeding_row in seeding_rows:
+                    columns = seeding_row.xpath('./td | ./th')
+                    if len(columns) < 5:
+                        continue
+                    size = self.num_filesize(columns[3].xpath("string(.)").strip())
+                    # 无表头的做种表以有效体积列识别数据行，避免把表头计入数量。
+                    if not size:
+                        continue
+                    seeders = text_tools.parse_int(columns[4].xpath("string(.)").strip())
 
-                tmp_seeding_size += size
-                tmp_seeding_info.append([seeders, size])
+                    tmp_seeding += 1
+                    tmp_seeding_size += size
+                    tmp_seeding_info.append([seeders, size])
 
             if not self.seeding_size:
                 self.seeding_size = tmp_seeding_size

tests/fixtures/architecture/mypy-baseline.json

@@ -1702,6 +1702,7 @@
   "app/modules/indexer/parser/mtorrent.py": {
     "assignment": 11,
     "empty-body": 1,
+    "index": 2,
     "no-untyped-def": 6,
     "type-arg": 1
   },

tests/test_mtorrent_parser.py

@@ -1,8 +1,6 @@
 # -*- coding: utf-8 -*-
 import json
 
-import pytest
-
 from app.modules.indexer.parser.mtorrent import MTorrentSiteUserInfo
 
 
@@ -19,7 +17,7 @@ def _build_parser() -> MTorrentSiteUserInfo:
     )
 
 
-def test_mtorrent_seeding_skips_invalid_torrent_items():
+def test_mtorrent_seeding_skips_invalid_torrent_items(monkeypatch):
     """
     MTorrent 返回空种子对象时应跳过异常条目并只统计有效做种。
     """
@@ -32,6 +30,11 @@ def test_mtorrent_seeding_skips_invalid_torrent_items():
         "userid": parser.userid,
     }
 
+    monkeypatch.setattr(
+        parser,
+        "_get_page_content",
+        lambda **_: json.dumps({"data": {"seeder": 1}}),
+    )
     html_text = json.dumps(
         {
             "code": "0",
@@ -52,73 +55,3 @@ def test_mtorrent_seeding_skips_invalid_torrent_items():
     assert parser.seeding == 1
     assert parser.seeding_size == 1024
     assert parser.seeding_info == [[3, 1024]]
-
-
-def test_mtorrent_seeding_paginates_from_list_metadata(monkeypatch):
-    """应按做种列表的分页元数据读取全部种子，不依赖 tracker 私有接口。"""
-    parser = _build_parser()
-    parser.userid = "1"
-    parser._torrent_seeding_params = {
-        "pageNumber": 1,
-        "pageSize": 2,
-        "type": "SEEDING",
-        "userid": parser.userid,
-    }
-    monkeypatch.setattr(
-        parser,
-        "_get_page_content",
-        lambda **_: pytest.fail("解析做种列表时不应请求 tracker 私有接口"),
-    )
-
-    first_page = json.dumps(
-        {
-            "code": "0",
-            "data": {
-                "total": 3,
-                "totalPages": 2,
-                "data": [
-                    {"torrent": {"size": "1024", "source": "3"}},
-                    {"torrent": {"size": "2048", "source": "4"}},
-                ],
-            },
-        }
-    )
-    second_page = json.dumps(
-        {
-            "code": "0",
-            "data": {
-                "total": 3,
-                "totalPages": 2,
-                "data": [{"torrent": {"size": "4096", "source": "5"}}],
-            },
-        }
-    )
-
-    assert parser._parse_user_torrent_seeding_info(first_page) == ""
-    assert parser._torrent_seeding_params["pageNumber"] == 2
-    assert parser._parse_user_torrent_seeding_info(second_page) is None
-    assert parser.seeding == 3
-    assert parser.seeding_size == 7168
-    assert parser.seeding_info == [[3, 1024], [4, 2048], [5, 4096]]
-
-
-def test_mtorrent_seeding_uses_page_size_when_totals_are_missing():
-    """接口未返回总页数时，满页后继续读取并在空页结束。"""
-    parser = _build_parser()
-    parser._torrent_seeding_params = {"pageNumber": 1, "pageSize": 2}
-    full_page = json.dumps(
-        {
-            "code": "0",
-            "data": {
-                "data": [
-                    {"torrent": {"size": "1024", "source": "3"}},
-                    {"torrent": {"size": "2048", "source": "4"}},
-                ]
-            },
-        }
-    )
-    empty_page = json.dumps({"code": "0", "data": {"data": []}})
-
-    assert parser._parse_user_torrent_seeding_info(full_page) == ""
-    assert parser._torrent_seeding_params["pageNumber"] == 2
-    assert parser._parse_user_torrent_seeding_info(empty_page) is None

tests/test_nexus_php_parser.py

@@ -226,3 +226,82 @@ def test_nexus_php_uuid_seeding_next_page_uses_useruuid():
     assert query_params["useruuid"] == ["53d25816-f7b7-4372-b58d-6f445acfe090"]
     assert "userid" not in query_params
     assert query_params["type"] == ["seeding"]
+
+
+def test_nexus_php_pterclub_profile_parses_current_seeding_table():
+    """
+    猫站用户详情页的“当前做种”表应按第 4、5 列统计做种体积和做种人数。
+    """
+    parser = _build_parser()
+    parser.userid = "123"
+    html_text = """
+    <html><body>
+      <table><tr>
+        <td>当前做种</td>
+        <td><table><tbody>
+          <tr><td>详情</td><td><a>种子一</a></td><td>分类</td><td>1.5 TB</td><td><a>12</a></td></tr>
+          <tr><td>详情</td><td><a>种子二</a></td><td>分类</td><td>700 GB</td><td><a>3</a></td></tr>
+        </tbody></table></td>
+      </tr></table>
+    </body></html>
+    """
+
+    parser._parse_user_detail_info(html_text)
+
+    assert parser.seeding == 2
+    assert parser.seeding_size == 1.5 * 1024**4 + 700 * 1024**3
+    assert parser.seeding_info == [[12, int(1.5 * 1024**4)], [3, 700 * 1024**3]]
+
+
+def test_nexus_php_pterclub_full_seeding_list_uses_site_columns():
+    """
+    猫站完整做种页即使没有表头，也应读取第 4 列体积和第 5 列做种人数。
+    """
+    parser = _build_parser()
+    parser._site_domain = "pterclub.net"
+    parser.userid = "123"
+    html_text = """
+    <html><body>
+      <div id="outer"><table><tbody>
+        <tr><td>详情</td><td><a href="details.php?id=1">种子一</a></td><td>分类</td><td>1.5 TB</td><td><a>12</a></td></tr>
+      </tbody></table>
+      <p class="np-pager"><font class="gray">1</font><a href="getusertorrentlist.php?userid=123&type=seeding&page=2">下一页</a></p></div>
+    </body></html>
+    """
+
+    next_page = parser._parse_user_torrent_seeding_info(html_text)
+
+    assert parser.seeding == 1
+    assert parser.seeding_size == int(1.5 * 1024**4)
+    assert parser.seeding_info == [[12, int(1.5 * 1024**4)]]
+    assert next_page == "getusertorrentlist.php?userid=123&type=seeding&page=2"
+
+
+def test_nexus_php_seeding_list_replaces_profile_fallback_without_double_counting(monkeypatch):
+    """
+    做种列表成功时应覆盖详情页中的部分做种统计，避免重复累加。
+    """
+    parser = _build_parser()
+    parser.userid = "123"
+    parser._torrent_seeding_page = "getusertorrentlistajax.php?userid=123&type=seeding"
+    parser._parse_user_detail_info("""
+        <html><body><table><tr>
+          <td>当前做种</td>
+          <td><table><tbody>
+            <tr><td>详情</td><td>种子一</td><td>分类</td><td>1.5 TB</td><td>12</td></tr>
+          </tbody></table></td>
+        </tr></table></body></html>
+    """)
+    list_html = """
+    <html><body><table class="torrents">
+      <tr><td>标题</td><td>分类</td><td>大小</td><td>在做种</td></tr>
+      <tr><td>种子一</td><td>分类</td><td>500 GB</td><td><b><a>7</a></b></td></tr>
+    </table></body></html>
+    """
+    monkeypatch.setattr(parser, "_get_page_content", lambda **_: list_html)
+
+    parser._parse_seeding_pages()
+
+    assert parser.seeding == 1
+    assert parser.seeding_size == 500 * 1024**3
+    assert parser.seeding_info == [[7, 500 * 1024**3]]