yeyinly/tvboxzt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat: finish 两个BT spider

Browse Source
This commit is contained in:
Harold
2026-04-24 18:43:05 +08:00
parent 8f5451670c
commit 9c232bb68b
2 changed files with 35 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
py/tests/test_两个BT.py
+23
View File
@@ -191,6 +191,29 @@ class TestLiangGeBTSpider(unittest.TestCase):
self.assertEqual(result["url"], "https://media.example/final.mp4")
self.assertEqual(result["header"]["Referer"], "https://www.bttwoo.com/v_play/play-100.html")
@patch.object(Spider, "_request_html")
def test_player_content_extracts_media_from_iframe(self, mock_request_html):
mock_request_html.side_effect = [
'<iframe src="/embed/demo.html"></iframe>',
'<script>var source = "https://media.example/iframe.m3u8";</script>',
]
play_id = self.spider._encode_play_id("play-200", "900", "第2集")
result = self.spider.playerContent("两个BT", play_id, {})
self.assertEqual(mock_request_html.call_args_list[0].args[0], "https://www.bttwoo.com/v_play/play-200.html")
self.assertEqual(mock_request_html.call_args_list[1].args[0], "https://www.bttwoo.com/embed/demo.html")
self.assertEqual(result["parse"], 0)
self.assertEqual(result["url"], "https://media.example/iframe.m3u8")
@patch.object(Spider, "_request_html")
def test_player_content_falls_back_to_parse_when_no_media_found(self, mock_request_html):
mock_request_html.return_value = "<html><body>empty</body></html>"
play_id = self.spider._encode_play_id("play-300", "900", "第3集")
result = self.spider.playerContent("两个BT", play_id, {})
self.assertEqual(result["parse"], 1)
self.assertEqual(result["jx"], 1)
self.assertEqual(result["url"], "https://www.bttwoo.com/v_play/play-300.html")
self.assertEqual(result["header"]["Referer"], "https://www.bttwoo.com/v_play/play-300.html")
if __name__ == "__main__":
unittest.main()
py/两个BT.py
+12
View File
@@ -90,6 +90,14 @@ class Spider(BaseSpider):
media_url = self._extract_media_url(html)
if media_url:
return self._build_player_result(media_url, play_page_url)
iframe_url = self._extract_iframe_url(html)
if iframe_url:
iframe_html = self._request_html(iframe_url, referer=play_page_url)
iframe_media_url = self._extract_media_url(iframe_html)
if iframe_media_url:
return self._build_player_result(iframe_media_url, iframe_url)
return self._build_parse_result(play_page_url, play_page_url)
def _request_html(self, url, referer=None):
@@ -292,6 +300,10 @@ class Spider(BaseSpider):
},
}
def _extract_iframe_url(self, html):
matched = re.search(r'<iframe[^>]+src=["\']([^"\']+)["\']', str(html or ""), re.I)
return self._abs_url(matched.group(1)) if matched else ""
def _abs_url(self, value):
raw = str(value or "").strip()
if not raw: