GithubBackup/MoviePilot
1
0
Fork 0
mirror of https://github.com/jxxghp/MoviePilot.git synced 2026-09-05 23:47:41 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(site): adapt GPW Gazelle seeding

Browse Source
This commit is contained in:
jxxghp
2026-08-26 07:05:53 +08:00
parent 33a367ebf9
commit a706474310
2 changed files with 173 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/modules/indexer/parser/gazelle.py
+33 -9
View File
@@ -140,17 +140,41 @@ class GazelleSiteUserInfo(SiteParserBase):
page_seeding = 0 page_seeding = 0
page_seeding_size = 0 page_seeding_size = 0
page_seeding_info = [] page_seeding_info = []
seeding_sizes = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{size_col}]') semantic_rows = html.xpath(
seeding_seeders = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{seeders_col}]/text()') '//table[contains(@id, "torrent")]//tr['
if seeding_sizes and seeding_seeders: 'contains(concat(" ", normalize-space(@class), " "), " TableTorrent-rowTitle ")]'
page_seeding = len(seeding_sizes) )
if semantic_rows:
for i in range(0, len(seeding_sizes)): # GPW 等现代 Gazelle 页面以语义 class 标记列，且标题单元格带 colspan，
size = size_tools.parse_size(seeding_sizes[i].xpath("string(.)").strip()) # 固定列序号无法反映视觉列位置，必须逐行按 class 配对读取。
seeders = int(seeding_seeders[i]) for row in semantic_rows:
size_cells = row.xpath(
'./td[contains(concat(" ", normalize-space(@class), " "), '
'" TableTorrent-cellStatSize ")]'
)
seeder_cells = row.xpath(
'./td[contains(concat(" ", normalize-space(@class), " "), '
'" TableTorrent-cellStatSeeders ")]'
)
if not size_cells or not seeder_cells:
continue
size = size_tools.parse_size(size_cells[0].xpath("string(.)").strip())
seeders = text_tools.parse_int(seeder_cells[0].xpath("string(.)").strip())
page_seeding += 1
page_seeding_size += size page_seeding_size += size
page_seeding_info.append([seeders, size]) page_seeding_info.append([seeders, size])
else:
seeding_sizes = html.xpath(f'//table[contains(@id, "torrent")]//tr[position()>1]/td[{size_col}]')
seeding_seeders = html.xpath(f'//table[contains(@id, "torrent")]//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 = size_tools.parse_size(seeding_sizes[i].xpath("string(.)").strip())
seeders = text_tools.parse_int(seeding_seeders[i])
page_seeding_size += size
page_seeding_info.append([seeders, size])
if multi_page: if multi_page:
self.seeding += page_seeding self.seeding += page_seeding
tests/test_greatposterwall_indexer.py
+140
View File
@@ -0,0 +1,140 @@
from app.modules.indexer.parser.gazelle import GazelleSiteUserInfo
from app.modules.indexer.spider import SiteSpider
GPW_HTML = """
<html><body>
<table class="TableTorrent Table" id="torrent_table"><tbody>
<tr class="TableTorrent-rowTitle Table-row torrent_checked">
<td class="Table-cell is-name" colspan="3">
<div class="TableTorrent-title">
<span class="TableTorrent-titleActions">
<a href="/torrents.php?action=download&amp;id=115362">DL</a>
</span>
<a href="/torrents.php?id=8064&amp;torrentid=115362">
<span class="TorrentTitle">
<span class="TorrentTitle-item resolution">1080p</span>
<span class="TorrentTitle-item is-releaseGroup">Example-GROUP</span>
</span>
</a>
</div>
</td>
<td class="Table-cell TableTorrent-cellStat TableTorrent-cellStatTime">
<span title="Aug 24 2026, 22:51">today</span>
</td>
<td class="Table-cell TableTorrent-cellStat TableTorrent-cellStatSize">1.50 GiB</td>
<td class="Table-cell TableTorrent-cellStat TableTorrent-cellStatSnatches">12</td>
<td class="Table-cell TableTorrent-cellStat TableTorrent-cellStatSeeders u-colorRatio00">1,234</td>
<td class="Table-cell TableTorrent-cellStat TableTorrent-cellStatLeechers">2</td>
</tr>
</tbody></table>
</body></html>
"""
def _gpw_indexer() -> dict:
"""构造与站点资源一致的 GPW 搜索配置。"""
detail_selector = 'a[href*="torrents.php?id="][href*="torrentid="]'
return {
"id": "greatposterwall",
"name": "海豹",
"domain": "https://greatposterwall.com/",
"search": {
"paths": [
{"path": "torrents.php?searchstr={keyword}", "method": "get"}
]
},
"torrents": {
"list": {
"selector": "table#torrent_table > tbody > tr.TableTorrent-rowTitle"
},
"fields": {
"id": {
"selector": detail_selector,
"attribute": "href",
"filters": [
{"name": "re_search", "args": [r"torrentid=(\d+)", 1]}
],
},
"title": {"selector": f"{detail_selector} > span.TorrentTitle"},
"details": {"selector": detail_selector, "attribute": "href"},
"download": {
"selector": 'a[href*="torrents.php?action=download"]',
"attribute": "href",
},
"size": {"selector": "td.TableTorrent-cellStatSize"},
"seeders": {"selector": "td.TableTorrent-cellStatSeeders"},
"leechers": {"selector": "td.TableTorrent-cellStatLeechers"},
"grabs": {"selector": "td.TableTorrent-cellStatSnatches"},
"date_elapsed": {
"selector": "td.TableTorrent-cellStatTime > span",
"attribute": "title",
"optional": True,
},
"date": {
"text": "{% if fields['date_elapsed'] %}{{ fields['date_elapsed'] }}{% else %}now{% endif %}",
"filters": [{"name": "dateparse", "args": "%b %d %Y, %H:%M"}],
},
},
},
}
def _build_parser() -> GazelleSiteUserInfo:
"""构造 Gazelle 用户数据解析器测试实例。"""
return GazelleSiteUserInfo(
site_name="海豹",
url="https://greatposterwall.com/",
site_cookie="",
apikey=None,
token=None,
)
def test_greatposterwall_search_parses_tabletorrent_row(monkeypatch):
"""GPW 搜索结果应从 TableTorrent 行提取完整种子信息。"""
monkeypatch.setattr(
"app.modules.indexer.spider.rust_accel.parse_indexer_torrents",
lambda **_: None,
)
results = SiteSpider(_gpw_indexer(), keyword="Example").parse(GPW_HTML)
assert len(results) == 1
assert results[0]["title"] == "1080p Example-GROUP"
assert results[0]["page_url"].endswith("torrents.php?id=8064&torrentid=115362")
assert results[0]["enclosure"].endswith("torrents.php?action=download&id=115362")
assert results[0]["size"] == 1610612736
assert results[0]["seeders"] == 1234
assert results[0]["peers"] == 2
assert results[0]["grabs"] == 12
assert results[0]["pubdate"] == "2026-08-24 22:51:00"
def test_greatposterwall_seeding_uses_semantic_stat_cells():
"""GPW 标题列使用 colspan 时，做种统计仍应按语义 class 正确配对。"""
parser = _build_parser()
next_page = parser._parse_user_torrent_seeding_info(GPW_HTML)
assert next_page is None
assert parser.seeding == 1
assert parser.seeding_size == 1610612736
assert parser.seeding_info == [[1234, 1610612736]]
def test_gazelle_seeding_keeps_legacy_column_fallback():
"""没有语义 class 的传统 Gazelle 表格仍应按既有列号统计做种。"""
parser = _build_parser()
html_text = """
<table id="torrent_table">
<tr><td>Name</td><td>Type</td><td>Time</td><td>Size</td><td>Snatches</td><td>Seeders</td><td>Leechers</td></tr>
<tr><td>Example</td><td>Movie</td><td>today</td><td>2 GiB</td><td>3</td><td>45</td><td>1</td></tr>
</table>
"""
parser._parse_user_torrent_seeding_info(html_text)
assert parser.seeding == 1
assert parser.seeding_size == 2147483648
assert parser.seeding_info == [[45, 2147483648]]