From c070e216e5cee4f26fae76a11b419ec888fb2fcc Mon Sep 17 00:00:00 2001
From: Harold <8866033@gmail.com>
Date: Sun, 19 Apr 2026 20:39:48 +0800
Subject: [PATCH] fix: support nested cupfox player data

---
 py/tests/test_茶杯狐.py | 18 ++++++++++++++++
 py/茶杯狐.py            | 48 +++++++++++++++++++++++++++++++++++++++--
 2 files changed, 64 insertions(+), 2 deletions(-)

diff --git a/py/tests/test_茶杯狐.py b/py/tests/test_茶杯狐.py
index 3b5d483..39e341b 100644
--- a/py/tests/test_茶杯狐.py
+++ b/py/tests/test_茶杯狐.py
@@ -123,6 +123,24 @@ class TestCupfoxSpider(unittest.TestCase):
         self.assertEqual(data["url"], "vid-1")
         self.assertEqual(data["from"], "lineA")
 
+    def test_extract_player_data_supports_nested_object_literal(self):
+        html = """
+        <script>
+        var player_aaaa={
+          "flag":"play",
+          "encrypt":0,
+          "vod_data":{"vod_name":"八千里路云和月","vod_actor":"王阳,万茜"},
+          "url":"DJP5QlL6j66wOTl+NrtIwkJCenh0ZVBlTHJzTWJWd0Y1eTNBd1crTmt5NmpjK05XOTFxb25iSTJ4d2RSUms4YllrVVhSU2xyK0JGcFFZZThibmVqU1BWUFlEMnVWTUpQVmR1Vm9FZHJJWlFaaHNaR3pmTHYwbjVyUE11dkdwbzVOMTJpRFpBK3cvYUl5RnNuQXNaelFZZ0plWVhWYitFenJOMHhUZz09",
+          "from":"rb",
+          "server":"no"
+        };
+        </script>
+        """
+        data = self.spider._extract_player_data(html)
+        self.assertEqual(data["encrypt"], 0)
+        self.assertEqual(data["vod_data"]["vod_name"], "八千里路云和月")
+        self.assertEqual(data["from"], "rb")
+
     def test_decode2_recovers_shifted_text(self):
         encoded = "QXdCQ2tE"
         self.assertEqual(self.spider._decode2(encoded), "P0")
diff --git a/py/茶杯狐.py b/py/茶杯狐.py
index af8d5b4..be976b2 100644
--- a/py/茶杯狐.py
+++ b/py/茶杯狐.py
@@ -160,14 +160,58 @@ class Spider(BaseSpider):
         return second["text"]
 
     def _extract_player_data(self, html_text):
-        matched = re.search(r"player_aaaa\s*=\s*(\{[\s\S]*?\})\s*;?", str(html_text or ""))
+        text = str(html_text or "")
+        matched = re.search(r"player_aaaa\s*=", text, re.I)
         if not matched:
             return None
+        object_start = text.find("{", matched.end())
+        if object_start < 0:
+            return None
+        raw_object = self._extract_balanced_object(text, object_start)
+        if not raw_object:
+            return None
         try:
-            return json.loads(matched.group(1))
+            return json.loads(raw_object)
         except Exception:
             return None
 
+    def _extract_balanced_object(self, text, start_index):
+        raw = str(text or "")
+        if start_index < 0 or start_index >= len(raw) or raw[start_index] != "{":
+            return ""
+
+        depth = 0
+        in_string = False
+        quote_char = ""
+        escaped = False
+
+        for index in range(start_index, len(raw)):
+            char = raw[index]
+            if in_string:
+                if escaped:
+                    escaped = False
+                elif char == "\\":
+                    escaped = True
+                elif char == quote_char:
+                    in_string = False
+                continue
+
+            if char in ('"', "'"):
+                in_string = True
+                quote_char = char
+                continue
+
+            if char == "{":
+                depth += 1
+                continue
+
+            if char == "}":
+                depth -= 1
+                if depth == 0:
+                    return raw[start_index : index + 1]
+
+        return ""
+
     def _decode2(self, encoded):
         if not encoded:
             return ""