独播库.py

2026-04-18 12:55:09 +08:00
parent b72587d86b
commit f94acdccf7
4 changed files with 205 additions and 27 deletions
@@ -4,8 +4,12 @@ from importlib.machinery import SourceFileLoader
 from pathlib import Path
 from unittest.mock import patch

-from Cryptodome.Cipher import AES
-from Cryptodome.Util.Padding import pad
+try:
+    from Crypto.Cipher import AES
+    from Crypto.Util.Padding import pad
+except ModuleNotFoundError:
+    from Cryptodome.Cipher import AES
+    from Cryptodome.Util.Padding import pad


 ROOT = Path(__file__).resolve().parents[1]
@@ -32,7 +32,7 @@ class TestDBKUSpider(unittest.TestCase):
            cards,
            [
                {
-                    "vod_id": "https://www.dbku.tv/voddetail/123.html",
+                    "vod_id": "123",
                    "vod_name": "示例影片",
                    "vod_pic": "https://img.example/dbku.jpg",
                    "vod_remarks": "更新至10集",
@@ -93,7 +93,7 @@ class TestDBKUSpider(unittest.TestCase):
        </div>
        """
        result = self.spider.searchContent("繁花", False, "1")
-        self.assertEqual(result["list"][0]["vod_id"], "https://www.dbku.tv/voddetail/321.html")
+        self.assertEqual(result["list"][0]["vod_id"], "321")
        self.assertEqual(result["list"][0]["vod_name"], "搜索影片")

    def test_parse_detail_page_extracts_meta_and_episodes(self):
@@ -112,12 +112,66 @@ class TestDBKUSpider(unittest.TestCase):
        <a href="/vodplay/100-1-1.html">第1集</a>
        <a href="/vodplay/100-1-2.html">第2集</a>
        """
-        result = self.spider._parse_detail_page(html, "https://www.dbku.tv/voddetail/100.html")
+        result = self.spider._parse_detail_page(html, "100")
        vod = result["list"][0]
+        self.assertEqual(vod["vod_id"], "100")
        self.assertEqual(vod["vod_name"], "独播剧")
        self.assertEqual(vod["vod_year"], "2025")
        self.assertEqual(vod["vod_play_from"], "独播库")
-        self.assertIn("第1集$https://www.dbku.tv/vodplay/100-1-1.html", vod["vod_play_url"])
+        self.assertIn("第1集$100-1-1", vod["vod_play_url"])
+
+    def test_parse_detail_page_extracts_structured_fields_from_detail_block(self):
+        html = """
+        <div class="myui-content__thumb">
+          <img data-original="/poster-ai.jpg" />
+        </div>
+        <div class="myui-content__detail">
+          <h1 class="title">AI教我谈恋爱</h1>
+          <div id="rating" class="score" data-mid="1" data-id="143508" data-score="3">
+            <span class="branch">6</span>
+          </div>
+          <p class="data">
+            <span class="text-muted">分类：</span><a href="/vodshow/21-----------.html">短剧</a>
+            <span class="split-line"></span>
+            <span class="text-muted hidden-xs">地区：</span><a href="/vodshow/21-大陆----------.html">大陆</a>
+            <span class="split-line"></span>
+            <span class="text-muted hidden-xs">年份：</span><a href="/vodshow/21-----------2026.html">2026</a>
+          </p>
+          <p class="data hidden-sm"><span class="text-muted">更新：</span><span class="text-red">2026-04-18 09:59:00</span></p>
+          <p class="data">
+            <span class="text-muted">主演：</span>
+            <a href="/vodsearch/-%E7%AB%A0%E7%85%9C%E5%A5%87------------.html" target="_blank">章煜奇</a>&nbsp;
+            <a href="/vodsearch/-%E8%A2%81%E4%BC%8A------------.html" target="_blank">袁伊</a>&nbsp;
+            <a href="/vodsearch/-%E9%83%AD%E5%AD%90%E6%B8%9D------------.html" target="_blank">郭子渝</a>&nbsp;
+            <a href="/vodsearch/-%E5%88%98%E5%B8%8C%E5%A9%A7------------.html" target="_blank">刘希婧</a>&nbsp;
+          </p>
+          <p class="data">
+            <span class="text-muted">导演：</span>
+            <a href="/vodsearch/-----%E5%BC%A0%E4%B8%96%E5%8D%9A--------.html" target="_blank">张世博</a>&nbsp;
+            <a href="/vodsearch/-----%E6%BA%90%E8%AF%97%E5%98%89--------.html" target="_blank">源诗嘉</a>&nbsp;
+          </p>
+          <p class="data hidden-xs"><span class="text-muted">简介：</span>《AI教我谈恋爱》线上看，共26集，《AI教我谈恋爱》简介:漫画家袁七柚意外绑定...<a href="#desc">详情</a></p>
+        </div>
+        <a href="/vodplay/143508-1-1.html">第1集</a>
+        """
+        result = self.spider._parse_detail_page(html, "143508")
+        vod = result["list"][0]
+        self.assertEqual(vod["vod_id"], "143508")
+        self.assertEqual(vod["path"], "https://www.dbku.tv/voddetail/143508.html")
+        self.assertEqual(vod["vod_pic"], "https://www.dbku.tv/poster-ai.jpg")
+        self.assertEqual(vod["type_name"], "短剧")
+        self.assertEqual(vod["vod_area"], "大陆")
+        self.assertEqual(vod["vod_year"], "2026")
+        self.assertEqual(vod["vod_time"], "2026-04-18 09:59:00")
+        self.assertEqual(vod["vod_actor"], "章煜奇,袁伊,郭子渝,刘希婧")
+        self.assertEqual(vod["vod_director"], "张世博,源诗嘉")
+        self.assertEqual(vod["vod_content"], "《AI教我谈恋爱》线上看，共26集，《AI教我谈恋爱》简介:漫画家袁七柚意外绑定...")
+        self.assertEqual(vod["vod_play_url"], "第1集$143508-1-1")
+        self.assertEqual(vod["vod_tag"], "")
+        self.assertEqual(vod["vod_remarks"], "6")
+        self.assertEqual(vod["vod_lang"], "")
+        self.assertNotIn("dbid", vod)
+        self.assertNotIn("type", vod)

    @patch.object(Spider, "_request_html")
    def test_detail_content_reads_from_vod_id_url(self, mock_request_html):
@@ -125,8 +179,9 @@ class TestDBKUSpider(unittest.TestCase):
        <h1 class="title">详情影片</h1>
        <a href="/vodplay/200-1-1.html">第1集</a>
        """
-        result = self.spider.detailContent(["https://www.dbku.tv/voddetail/200.html"])
-        self.assertEqual(result["list"][0]["vod_id"], "https://www.dbku.tv/voddetail/200.html")
+        result = self.spider.detailContent(["200"])
+        self.assertEqual(mock_request_html.call_args.args[0], "https://www.dbku.tv/voddetail/200.html")
+        self.assertEqual(result["list"][0]["vod_id"], "200")
        self.assertEqual(result["list"][0]["vod_name"], "详情影片")

    def test_parse_player_data_reads_json_block(self):
@@ -162,18 +217,36 @@ class TestDBKUSpider(unittest.TestCase):
        var player_data = {"url":"https://video.example/final.m3u8","encrypt":"0"};
        </script>
        """
-        result = self.spider.playerContent("独播库", "https://www.dbku.tv/vodplay/100-1-1.html", {})
+        result = self.spider.playerContent("独播库", "100-1-1", {})
        self.assertEqual(result["parse"], 0)
        self.assertEqual(result["url"], "https://video.example/final.m3u8")
        self.assertEqual(result["header"]["Referer"], "https://www.dbku.tv/vodplay/100-1-1.html")

+    @patch.object(Spider, "log")
+    @patch.object(Spider, "_request_html")
+    def test_player_content_logs_resolved_player_fields(self, mock_request_html, mock_log):
+        mock_request_html.return_value = """
+        <script>
+        var player_data = {"url":"aHR0cHM6Ly92aWRlby5leGFtcGxlL2ZpbmFsLm0zdTg=","encrypt":"2"};
+        </script>
+        """
+        self.spider.playerContent("独播库", "100-1-1", {})
+        mock_log.assert_called()
+        payload = mock_log.call_args.args[0]
+        self.assertEqual(payload["stage"], "playerContent")
+        self.assertEqual(payload["input_url"], "100-1-1")
+        self.assertEqual(payload["request_url"], "https://www.dbku.tv/vodplay/100-1-1.html")
+        self.assertEqual(payload["encrypt"], 2)
+        self.assertEqual(payload["decoded_url"], "https://video.example/final.m3u8")
+        self.assertEqual(payload["final_url"], "https://video.example/final.m3u8")
+
    @patch.object(Spider, "_request_html")
    def test_player_content_follows_internal_jump(self, mock_request_html):
        mock_request_html.side_effect = [
            '<script>var player_data = {"url":"/vodplay/100-1-2.html","encrypt":"0"};</script>',
            '<script>var player_data = {"url":"https://video.example/jump-final.m3u8","encrypt":"0"};</script>',
        ]
-        result = self.spider.playerContent("独播库", "https://www.dbku.tv/vodplay/100-1-1.html", {})
+        result = self.spider.playerContent("独播库", "100-1-1", {})
        self.assertEqual(result["parse"], 0)
        self.assertEqual(result["url"], "https://video.example/jump-final.m3u8")

@@ -5,8 +5,12 @@ import re
 import sys
 from urllib.parse import quote, urljoin

-from Cryptodome.Cipher import AES
-from Cryptodome.Util.Padding import unpad
+try:
+    from Crypto.Cipher import AES
+    from Crypto.Util.Padding import unpad
+except ModuleNotFoundError:
+    from Cryptodome.Cipher import AES
+    from Cryptodome.Util.Padding import unpad

 from base.spider import Spider as BaseSpider

@@ -64,6 +64,42 @@ class Spider(BaseSpider):
            return self.host + raw
        return self.host + "/" + raw

+    def _extract_play_id(self, href):
+        raw = str(href or "").strip()
+        matched = re.search(r"/vodplay/(\d+-\d+-\d+)\.html", raw)
+        if matched:
+            return matched.group(1)
+        if re.fullmatch(r"\d+-\d+-\d+", raw):
+            return raw
+        return ""
+
+    def _extract_vod_id(self, href):
+        raw = str(href or "").strip()
+        matched = re.search(r"/voddetail/(\d+)\.html", raw)
+        if matched:
+            return matched.group(1)
+        if re.fullmatch(r"\d+", raw):
+            return raw
+        return ""
+
+    def _build_detail_request_url(self, id_or_url):
+        raw = str(id_or_url or "").strip()
+        if not raw:
+            return ""
+        vod_id = self._extract_vod_id(raw)
+        if vod_id:
+            return f"{self.host}/voddetail/{vod_id}.html"
+        return self._build_url(raw)
+
+    def _build_play_request_url(self, id_or_url):
+        raw = str(id_or_url or "").strip()
+        if not raw:
+            return ""
+        play_id = self._extract_play_id(raw)
+        if play_id:
+            return f"{self.host}/vodplay/{play_id}.html"
+        return self._build_url(raw)
+
    def _parse_list_cards(self, html):
        root = self.html(html)
        results = []
@@ -79,7 +115,7 @@ class Spider(BaseSpider):
            for anchor in card.xpath(".//a[@href]"):
                raw_href = (anchor.xpath("./@href") or [""])[0].strip()
                if "/voddetail/" in raw_href:
-                    href = self._build_url(raw_href)
+                    href = self._extract_vod_id(raw_href)
                    title = (
                        (anchor.xpath("./@title") or [""])[0].strip()
                        or "".join(anchor.xpath(".//text()")).strip()
@@ -173,14 +209,62 @@ class Spider(BaseSpider):
                    return clean.split("：", 1)[-1].strip()
        return ""

+    def _clean_text(self, text):
+        return re.sub(r"\s+", " ", str(text or "").replace("\xa0", " ")).strip()
+
+    def _extract_detail_text(self, detail_root, html, label, joiner=""):
+        value = self._extract_detail_field(detail_root, label, joiner=joiner)
+        if value:
+            return value
+        return self._extract_text_by_prefix(html, [f"{label}："])
+
+    def _extract_detail_field(self, detail_root, label, joiner=""):
+        if detail_root is None:
+            return ""
+        nodes = detail_root.xpath(f'.//span[contains(normalize-space(.), "{label}：")]')
+        if not nodes:
+            return ""
+        label_node = nodes[0]
+        values = []
+        if self._clean_text(label_node.tail):
+            values.append(self._clean_text(label_node.tail))
+        for sibling in label_node.itersiblings():
+            classes = " ".join(sibling.xpath("./@class"))
+            if sibling.tag == "span" and "text-muted" in classes:
+                break
+            if sibling.tag == "span" and "split-line" in classes:
+                continue
+            if sibling.tag == "a" and (sibling.xpath("./@href") or [""])[0].startswith("#"):
+                break
+            text = self._clean_text("".join(sibling.xpath(".//text()")))
+            if text:
+                values.append(text)
+            tail = self._clean_text(sibling.tail)
+            if tail:
+                values.append(tail)
+
+        cleaned = []
+        for value in values:
+            if value and value not in cleaned:
+                cleaned.append(value)
+        if joiner:
+            return joiner.join(cleaned)
+        return self._clean_text("".join(cleaned))
+
    def _parse_detail_page(self, html, vod_id):
        root = self.html(html)
-        title = ((root.xpath("//*[contains(@class,'title')][1]//text()") or [""])[0]).strip()
+        detail_root = (
+            root.xpath("//*[contains(concat(' ', normalize-space(@class), ' '), ' myui-content__detail ')][1]")
+            or [root]
+        )[0]
+        title = ((detail_root.xpath(".//*[contains(@class,'title')][1]//text()") or [""])[0]).strip()
        pic = (
            (root.xpath("//*[contains(@class,'myui-content__thumb')]//img/@data-original") or [""])[0].strip()
            or (root.xpath("//*[contains(@class,'myui-content__thumb')]//img/@src") or [""])[0].strip()
        )
-        content = "".join(root.xpath("//*[contains(@class,'data')][1]//text()")).strip()
+        content = self._extract_detail_field(detail_root, "简介") or "".join(
+            root.xpath("//*[contains(@class,'data')][1]//text()")
+        ).strip()

        episodes = []
        seen = set()
@@ -189,30 +273,42 @@ class Spider(BaseSpider):
            html,
            re.I,
        ):
-            url = self._build_url(href)
+            play_id = self._extract_play_id(href)
            name = re.sub(r"<[^>]*>", "", label).strip()
-            if not url or not name or "立即播放" in name or url in seen:
+            if not play_id or not name or "立即播放" in name or play_id in seen:
                continue
-            seen.add(url)
-            episodes.append(f"{name}${url}")
+            seen.add(play_id)
+            episodes.append(f"{name}${play_id}")

        vod = {
            "vod_id": vod_id,
+            "path": self._build_detail_request_url(vod_id),
            "vod_name": title,
            "vod_pic": self._build_url(pic),
-            "vod_year": self._extract_text_by_prefix(html, ["年份："]),
-            "vod_area": self._extract_text_by_prefix(html, ["地区："]),
-            "vod_actor": self._extract_text_by_prefix(html, ["主演："]),
-            "vod_director": self._extract_text_by_prefix(html, ["导演："]),
-            "vod_content": content,
+            "vod_tag": "",
+            "vod_time": self._extract_detail_text(detail_root, html, "更新"),
+            "vod_remarks": self._clean_text(
+                ((detail_root.xpath('.//*[@id="rating"]//*[contains(@class,"branch")][1]//text()') or [""])[0]).strip()
+                or ((detail_root.xpath('.//*[@id="ratewords"][1]//text()') or [""])[0]).strip()
+            ),
            "vod_play_from": "独播库",
            "vod_play_url": "#".join(episodes),
+            "type_name": self._extract_detail_text(detail_root, html, "分类"),
+            "vod_content": content,
+            "vod_year": self._extract_detail_text(detail_root, html, "年份"),
+            "vod_area": self._extract_detail_text(detail_root, html, "地区"),
+            "vod_lang": "",
+            "vod_director": self._extract_detail_text(detail_root, html, "导演", joiner=","),
+            "vod_actor": self._extract_detail_text(detail_root, html, "主演", joiner=","),
        }
        return {"list": [vod]}

    def detailContent(self, ids):
        vod_id = ids[0]
-        html = self._request_html(vod_id, expect_xpath="//*[contains(@class,'title')]|//a[contains(@href,'/vodplay/')]")
+        html = self._request_html(
+            self._build_detail_request_url(vod_id),
+            expect_xpath="//*[contains(@class,'title')]|//a[contains(@href,'/vodplay/')]",
+        )
        return self._parse_detail_page(html, vod_id)

    def _parse_player_data(self, html):
@@ -256,7 +352,8 @@ class Spider(BaseSpider):
    def playerContent(self, flag, id, vipFlags):
        current = id
        for _ in range(3):
-            html = self._request_html(current, referer=self.host)
+            request_url = self._build_play_request_url(current)
+            html = self._request_html(request_url, referer=self.host)
            data = self._parse_player_data(html)
            if not data:
                return {"parse": 0, "playUrl": "", "url": ""}
@@ -271,6 +368,6 @@ class Spider(BaseSpider):
                "parse": 0,
                "playUrl": "",
                "url": play_url,
-                "header": self._build_player_headers(current),
+                "header": self._build_player_headers(request_url),
            }
        return {"parse": 0, "playUrl": "", "url": ""}