AI 生成: AVTV - 自动收集、过滤、去重并生成 m3u 直播源订阅文件，每日自动更新，涵盖央视、卫视及地方台。

avtv/parser.py

@@ -0,0 +1,88 @@
+"""解析 m3u / txt 格式的直播源文本，提取频道条目。"""
+
+import re
+from dataclasses import dataclass, field
+
+
+@dataclass
+class Channel:
+    """单个频道条目。"""
+    name: str
+    url: str
+    logo: str = ""
+    group: str = ""
+    tvg_id: str = ""
+    extra: dict = field(default_factory=dict)
+
+
+_ATTR_RE = re.compile(r'([\w-]+)="([^"]*)"')
+
+
+def _parse_extinf(line: str) -> dict:
+    """解析 #EXTINF 行中的属性。"""
+    attrs = {}
+    for key, value in _ATTR_RE.findall(line):
+        attrs[key.lower()] = value
+    # 逗号后面的名称
+    if "," in line:
+        attrs["name"] = line.rsplit(",", 1)[1].strip()
+    return attrs
+
+
+def parse_m3u(text: str) -> list[Channel]:
+    """解析 m3u 文本。"""
+    channels: list[Channel] = []
+    lines = [ln.strip() for ln in text.splitlines()]
+    i = 0
+    while i < len(lines):
+        line = lines[i]
+        if line.startswith("#EXTINF"):
+            attrs = _parse_extinf(line)
+            # 下一行非注释的即为 URL
+            j = i + 1
+            while j < len(lines) and (not lines[j] or lines[j].startswith("#")):
+                j += 1
+            if j < len(lines):
+                url = lines[j]
+                name = attrs.get("name") or attrs.get("tvg-name") or ""
+                if name and url:
+                    channels.append(
+                        Channel(
+                            name=name,
+                            url=url,
+                            logo=attrs.get("tvg-logo", ""),
+                            group=attrs.get("group-title", ""),
+                            tvg_id=attrs.get("tvg-id", ""),
+                        )
+                    )
+                i = j + 1
+                continue
+        i += 1
+    return channels
+
+
+def parse_txt(text: str) -> list[Channel]:
+    """解析常见 txt 格式：频道名,URL 或 频道名 URL。"""
+    channels: list[Channel] = []
+    for raw in text.splitlines():
+        line = raw.strip()
+        if not line or line.startswith("#"):
+            continue
+        # 支持 , 或空格分隔
+        if "," in line:
+            name, url = line.split(",", 1)
+        elif " " in line:
+            name, url = line.split(" ", 1)
+        else:
+            continue
+        name, url = name.strip(), url.strip()
+        if name and url.startswith(("http://", "https://", "rtmp://", "rtsp://")):
+            channels.append(Channel(name=name, url=url))
+    return channels
+
+
+def parse(text: str) -> list[Channel]:
+    """自动识别格式并解析。"""
+    if "#EXTM3U" in text or "#EXTINF" in text:
+        return parse_m3u(text)
+    return parse_txt(text)