feat(meta): 新增 WEB 平台来源识别并支持更多音视频格式。

app/core/meta/metavideo.py

@@ -10,6 +10,7 @@ from app.core.meta.releasegroup import ReleaseGroupsMatcher
 from app.schemas.types import MediaType
 from app.utils.string import StringUtils
 from app.utils.tokens import Tokens
+from app.core.meta.streamingplatform import StreamingPlatforms
 
 
 class MetaVideo(MetaBase):
@@ -31,7 +32,7 @@ class MetaVideo(MetaBase):
     _part_re = r"(^PART[0-9ABI]{0,2}$|^CD[0-9]{0,2}$|^DVD[0-9]{0,2}$|^DISK[0-9]{0,2}$|^DISC[0-9]{0,2}$)"
     _roman_numerals = r"^(?=[MDCLXVI])M*(C[MD]|D?C{0,3})(X[CL]|L?X{0,3})(I[XV]|V?I{0,3})$"
     _source_re = r"^BLURAY$|^HDTV$|^UHDTV$|^HDDVD$|^WEBRIP$|^DVDRIP$|^BDRIP$|^BLU$|^WEB$|^BD$|^HDRip$|^REMUX$|^UHD$"
-    _effect_re = r"^SDR$|^HDR\d*$|^DOLBY$|^DOVI$|^DV$|^3D$|^REPACK$|^HLG$|^HDR10(\+|Plus)$"
+    _effect_re = r"^SDR$|^HDR\d*$|^DOLBY$|^DOVI$|^DV$|^3D$|^REPACK$|^HLG$|^HDR10(\+|Plus)$|^EDR$|^HQ$"
     _resources_type_re = r"%s|%s" % (_source_re, _effect_re)
     _name_no_begin_re = r"^[\[【].+?[\]】]"
     _name_no_chinese_re = r".*版|.*字幕"
@@ -51,7 +52,7 @@ class MetaVideo(MetaBase):
     _resources_pix_re = r"^[SBUHD]*(\d{3,4}[PI]+)|\d{3,4}X(\d{3,4})"
     _resources_pix_re2 = r"(^[248]+K)"
     _video_encode_re = r"^(H26[45])$|^(x26[45])$|^AVC$|^HEVC$|^VC\d?$|^MPEG\d?$|^Xvid$|^DivX$|^AV1$|^HDR\d*$|^AVS(\+|[23])$"
-    _audio_encode_re = r"^DTS\d?$|^DTSHD$|^DTSHDMA$|^Atmos$|^TrueHD\d?$|^AC3$|^\dAudios?$|^DDP\d?$|^DD\+\d?$|^DD\d?$|^LPCM\d?$|^AAC\d?$|^FLAC\d?$|^HD\d?$|^MA\d?$|^HR\d?$|^Opus\d?$|^Vorbis\d?$"
+    _audio_encode_re = r"^DTS\d?$|^DTSHD$|^DTSHDMA$|^Atmos$|^TrueHD\d?$|^AC3$|^\dAudios?$|^DDP\d?$|^DD\+\d?$|^DD\d?$|^LPCM\d?$|^AAC\d?$|^FLAC\d?$|^HD\d?$|^MA\d?$|^HR\d?$|^Opus\d?$|^Vorbis\d?$|^AV[3S]A$"
 
     def __init__(self, title: str, subtitle: str = None, isfile: bool = False):
         """
@@ -66,6 +67,8 @@ class MetaVideo(MetaBase):
         original_title = title
         self._source = ""
         self._effect = []
+        self.web_source = None
+        self._index = 0
         # 判断是否纯数字命名
         if isfile \
                 and title.isdigit() \
@@ -93,9 +96,12 @@ class MetaVideo(MetaBase):
         # 拆分tokens
         tokens = Tokens(title)
         self.tokens = tokens
+        # 实例化StreamingPlatforms对象
+        streaming_platforms = StreamingPlatforms()
         # 解析名称、年份、季、集、资源类型、分辨率等
         token = tokens.get_next()
         while token:
+            self._index += 1  # 更新当前处理的token索引
             # Part
             self.__init_part(token)
             # 标题
@@ -116,6 +122,9 @@ class MetaVideo(MetaBase):
             # 资源类型
             if self._continue_flag:
                 self.__init_resource_type(token)
+            # 流媒体平台
+            if self._continue_flag:
+                self.__init_web_source(token, streaming_platforms)
             # 视频编码
             if self._continue_flag:
                 self.__init_video_encode(token)
@@ -131,6 +140,9 @@ class MetaVideo(MetaBase):
             self.resource_effect = " ".join(self._effect)
         if self._source:
             self.resource_type = self._source.strip()
+        # 添加流媒体平台
+        if self.web_source:
+            self.resource_type = f"{self.web_source} {self.resource_type}"
         # 提取原盘DIY
         if self.resource_type and "BluRay" in self.resource_type:
             if (self.subtitle and re.findall(r'D[Ii]Y', self.subtitle)) \
@@ -574,6 +586,57 @@ class MetaVideo(MetaBase):
                 self._effect.append(effect)
             self._last_token = effect.upper()
 
+    def __init_web_source(self, token: str, streaming_platforms: StreamingPlatforms):
+        """
+        识别流媒体平台
+        """
+        if not self.name:
+            return
+
+        platform_name = None
+        query_range = 1
+
+        prev_token = None
+        prev_idx = self._index - 2
+        if prev_idx >= 0 and prev_idx < len(self.tokens._tokens):
+            prev_token = self.tokens._tokens[prev_idx]
+
+        next_token = self.tokens.peek()
+
+        if streaming_platforms.is_streaming_platform(token):
+            platform_name = streaming_platforms.get_streaming_platform_name(token)
+        else:
+            for adjacent_token, is_next in [(prev_token, False), (next_token, True)]:
+                if not adjacent_token or platform_name:
+                    continue
+
+                for separator in [" ", "-"]:
+                    if is_next:
+                        combined_token = f"{token}{separator}{adjacent_token}"
+                    else:
+                        combined_token = f"{adjacent_token}{separator}{token}"
+
+                    if streaming_platforms.is_streaming_platform(combined_token):
+                        platform_name = streaming_platforms.get_streaming_platform_name(combined_token)
+                        query_range = 2
+                        if is_next:
+                            self.tokens.get_next()
+                        break
+
+        if not platform_name:
+            return
+
+        web_tokens = ["WEB", "DL", "WEBDL", "WEBRIP"]
+        match_start_idx = self._index - query_range
+        match_end_idx = self._index - 1
+        start_index = max(0, match_start_idx - query_range)
+        end_index = min(len(self.tokens._tokens), match_end_idx + 1 + query_range)
+        tokens_to_check = self.tokens._tokens[start_index:end_index]
+
+        if any(tok and tok.upper() in web_tokens for tok in tokens_to_check):
+            self.web_source = platform_name
+            self._continue_flag = False
+
     def __init_video_encode(self, token: str):
         """
         识别视频编码

app/core/meta/streamingplatform.py

@@ -0,0 +1,104 @@
+from typing import Dict, Optional, List, Tuple
+
+from app.utils.singleton import Singleton
+
+
+class StreamingPlatforms(metaclass=Singleton):
+    """
+    流媒体平台简称与全称。
+    """
+    STREAMING_PLATFORMS: List[Tuple[str, str]] = [
+        ("AMZN", "Amazon"),
+        ("NF", "Netflix"),
+        ("ATVP", "Apple TV+"),
+        ("iT", "iTunes"),
+        ("DSNP", "Disney+"),
+        ("HS", "Hotstar"),
+        ("APPS", "Disney+ MENA"),
+        ("PMTP", "Paramount+"),
+        ("HMAX", "Max"),
+        ("", "Max"),
+        ("HULU", "Hulu"),
+        ("MA", "Movies Anywhere"),
+        ("BCORE", "Bravia Core"),
+        ("MS", "Microsoft Store"),
+        ("SHO", "Showtime"),
+        ("STAN", "Stan"),
+        ("PCOK", "Peacock"),
+        ("SKST", "SkyShowtime"),
+        ("NOW", "Now TV"),
+        ("FXTL", "Foxtel Now"),
+        ("BNGE", "Binge"),
+        ("CRKL", "Crackle"),
+        ("RKTN", "Rakuten TV"),
+        ("ALL4", "All 4"),
+        ("AS", "Adult Swim"),
+        ("BRTB", "Brtb TV"),
+        ("CNLP", "Canal+"),
+        ("CRIT", "Criterion Channel"),
+        ("DSCP", "Discovery+"),
+        ("", "ESPN"),
+        ("FOOD", "Food Network"),
+        ("MUBI", "Mubi"),
+        ("PLAY", "Google Play"),
+        ("YT", "YouTube"),
+        ("", "friDay"),
+        ("", "KKTV"),
+        ("", "ofiii"),
+        ("", "LiTV"),
+        ("", "MyVideo"),
+        ("Hami", "Hami Video"),
+        ("", "meWATCH"),
+        ("CATCHPLAY", "CATCHPLAY+"),
+        ("", "LINE TV"),
+        ("VIU", "Viu"),
+        ("IQ", ""),
+        ("", "WeTV"),
+        ("ABMA", "Abema"),
+        ("ADN", ""),
+        ("AT-X", ""),
+        ("Baha", ""),
+        ("BG", "B-Global"),
+        ("CR", "Crunchyroll"),
+        ("", "DMM"),
+        ("FOD", ""),
+        ("FUNi", "Funimation"),
+        ("HIDI", "HIDIVE"),
+        ("UNXT", "U-NEXT"),
+    ]
+
+    def __init__(self):
+        """初始化流媒体平台匹配器"""
+        self._lookup_cache = {}
+        self._build_cache()
+
+    def _build_cache(self) -> None:
+        """
+        构建查询缓存。
+        """
+        self._lookup_cache.clear()
+        for short_name, full_name in self.STREAMING_PLATFORMS:
+            canonical_name = full_name or short_name
+            if not canonical_name:
+                continue
+
+            aliases = {short_name, full_name}
+            for alias in aliases:
+                if alias:
+                    self._lookup_cache[alias.upper()] = canonical_name
+
+    def get_streaming_platform_name(self, platform_code: str) -> Optional[str]:
+        """
+        根据流媒体平台简称或全称获取标准名称。
+        """
+        if platform_code is None:
+            return None
+        return self._lookup_cache.get(platform_code.upper())
+
+    def is_streaming_platform(self, name: str) -> bool:
+        """
+        判断给定的字符串是否为已知的流媒体平台代码或名称。
+        """
+        if name is None:
+            return False
+        return name.upper() in self._lookup_cache

tests/cases/meta.py

@@ -153,7 +153,7 @@ meta_cases = [{
         "part": "",
         "season": "S01",
         "episode": "E02",
-        "restype": "WEB-DL",
+        "restype": "B-Global WEB-DL",
         "pix": "1080p",
         "video_codec": "x264",
         "audio_codec": "AAC"
@@ -569,7 +569,7 @@ meta_cases = [{
         "part": "",
         "season": "S02",
         "episode": "E05",
-        "restype": "WEB-DL",
+        "restype": "Crunchyroll WEB-DL",
         "pix": "1080p",
         "video_codec": "x264",
         "audio_codec": "AAC"
@@ -649,7 +649,7 @@ meta_cases = [{
         "part": "",
         "season": "",
         "episode": "",
-        "restype": "WEBRip",
+        "restype": "Netflix WEBRip",
         "pix": "1080p",
         "video_codec": "H264",
         "audio_codec": "DDP 5.1"
@@ -681,7 +681,7 @@ meta_cases = [{
         "part": "",
         "season": "S01",
         "episode": "E16",
-        "restype": "WEB-DL",
+        "restype": "KKTV WEB-DL",
         "pix": "1080p",
         "video_codec": "x264",
         "audio_codec": "AAC"
@@ -921,7 +921,7 @@ meta_cases = [{
         "part": "",
         "season": "S06",
         "episode": "E06",
-        "restype": "WEBRip",
+        "restype": "Max WEBRip",
         "pix": "1080p",
         "video_codec": "x264",
         "audio_codec": "DD 5.1"
@@ -937,7 +937,7 @@ meta_cases = [{
         "part": "",
         "season": "S06",
         "episode": "E05",
-        "restype": "WEBRip",
+        "restype": "Max WEBRip",
         "pix": "1080p",
         "video_codec": "x264",
         "audio_codec": "DD 5.1"
@@ -969,7 +969,7 @@ meta_cases = [{
         "part": "",
         "season": "S02",
         "episode": "",
-        "restype": "WEB-DL",
+        "restype": "Netflix WEB-DL",
         "pix": "2160p",
         "video_codec": "H265",
         "audio_codec": "DDP 5.1 Atmos"