From 53571a890a2e2ffd99914e1fe28faf5e0e4e90de Mon Sep 17 00:00:00 2001 From: Harold <8866033@gmail.com> Date: Sun, 19 Apr 2026 16:43:15 +0800 Subject: [PATCH] feat: add ddys detail parsing --- py/tests/test_低端影视.py | 45 +++++++++++++++++++ py/低端影视.py | 91 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 136 insertions(+) diff --git a/py/tests/test_低端影视.py b/py/tests/test_低端影视.py index f565e47..f7b3f8f 100644 --- a/py/tests/test_低端影视.py +++ b/py/tests/test_低端影视.py @@ -140,6 +140,51 @@ class TestDDYSSpider(unittest.TestCase): self.assertEqual(result["list"][0]["vod_id"], "https://ddys.io/anime/result/") self.assertEqual(result["pagecount"], 2) + def test_parse_detail_page_merges_direct_and_pan_sources(self): + html = """ + +

低端示例DDYS Example

+ 低端示例 +
2025 · 日本 · 动画
+
导演：导演甲
+
主演：主演乙
+

第一段简介。

第二段简介。

+ +
+ +
+
+ +
+ + """ + vod = self.spider._parse_detail_page(html, "https://ddys.io/anime/demo/") + self.assertEqual(vod["vod_name"], "低端示例") + self.assertEqual(vod["vod_pic"], "https://ddys.io/poster-detail.jpg") + self.assertEqual(vod["vod_year"], "2025") + self.assertEqual(vod["vod_area"], "日本") + self.assertEqual(vod["vod_class"], "动画") + self.assertEqual(vod["vod_director"], "导演甲") + self.assertEqual(vod["vod_actor"], "主演乙") + self.assertEqual(vod["vod_content"], "第一段简介。\n第二段简介。") + self.assertEqual(vod["vod_play_from"], "DDYS$$$quark$$$baidu") + self.assertIn("第2集$/play/ep2", vod["vod_play_url"]) + self.assertIn("夸克查看$https://pan.quark.cn/s/abc123", vod["vod_play_url"]) + self.assertIn("百度查看$https://pan.baidu.com/s/demo", vod["vod_play_url"]) + + @patch.object(Spider, "_request_html") + def test_detail_content_reads_detail_page_and_returns_single_vod(self, mock_request_html): + mock_request_html.return_value = """ +

详情标题

+ + """ + result = self.spider.detailContent(["https://ddys.io/movie/demo/"]) + self.assertEqual(mock_request_html.call_args.args[0], "https://ddys.io/movie/demo/") + self.assertEqual(result["list"][0]["vod_id"], "https://ddys.io/movie/demo/") + self.assertEqual(result["list"][0]["vod_name"], "详情标题") + self.assertEqual(result["list"][0]["vod_play_from"], "直连") + self.assertEqual(result["list"][0]["vod_play_url"], "全集$/play/detail-demo") + if __name__ == "__main__": unittest.main() diff --git a/py/低端影视.py b/py/低端影视.py index 54759e7..d19b0b6 100644 --- a/py/低端影视.py +++ b/py/低端影视.py @@ -1,4 +1,5 @@ # coding=utf-8 +import base64 import json import re import sys @@ -176,3 +177,93 @@ class Spider(BaseSpider): if not items: items = self._parse_movie_cards(html) return {"page": page, "pagecount": page + 1 if items else page, "total": len(items), "list": items} + + def _extract_switch_sources(self, html): + root = self.html(html) + if root is None: + return [] + groups = [] + for button in root.xpath("//button[contains(@onclick,'switchSource')]"): + onclick = (button.xpath("./@onclick") or [""])[0] + matched = re.search(r"switchSource\(\d+,\s*'([^']*)',\s*'[^']*'\)", onclick, re.S) + if not matched: + continue + name = self._clean_text("".join(button.xpath(".//text()"))) or self.name + payload = self._clean_text(matched.group(1)) + if not payload: + continue + if "$" not in payload: + payload = f"全集${payload}" + groups.append({"from": name, "urls": payload}) + return groups + + def _extract_pan_sources(self, html): + root = self.html(html) + if root is None: + return [] + groups = [] + for panel in root.xpath("//*[contains(@class,'download-type-content')]"): + panel_id = ((panel.xpath("./@id") or [""])[0]).replace("download-type-", "").strip() + if panel_id not in ("quark", "xunlei", "baidu"): + continue + entries = [] + for button in panel.xpath(".//button[@onclick]"): + onclick = (button.xpath("./@onclick") or [""])[0] + matched = re.search(r"atob\('([^']+)'\)", onclick) + if not matched: + continue + try: + link = base64.b64decode(matched.group(1)).decode("utf-8").strip() + except Exception: + continue + title = self._clean_text("".join(button.xpath(".//text()"))) or panel_id + if link: + entries.append(f"{title}${link}") + if entries: + groups.append({"from": panel_id, "urls": "#".join(dict.fromkeys(entries))}) + return groups + + def _parse_detail_page(self, html, vod_id): + root = self.html(html) + if root is None: + return {"vod_id": vod_id, "vod_name": "", "vod_play_from": "", "vod_play_url": ""} + title = self._clean_text("".join(root.xpath("//h1[1]/text()"))) + pic = ((root.xpath("//img[@alt][1]/@src") or [""])[0]).strip() + meta = self._clean_text("".join(root.xpath("(//*[contains(@class,'text-gray-600')])[1]//text()"))) + parts = [self._clean_text(item) for item in meta.split("·")] if meta else [] + director = "" + actor = "" + for text in root.xpath("//*[contains(@class,'text-gray-700')]"): + joined = self._clean_text("".join(text.xpath(".//text()"))) + if joined.startswith("导演："): + director = joined.split("：", 1)[-1].strip() + if joined.startswith("主演："): + actor = joined.split("：", 1)[-1].strip() + content = "\n".join( + [self._clean_text("".join(node.xpath(".//text()"))) for node in root.xpath("//*[contains(@class,'prose')]//p")] + ).strip() + play_groups = self._extract_switch_sources(html) + self._extract_pan_sources(html) + return { + "vod_id": vod_id, + "vod_name": title, + "vod_pic": self._build_url(pic), + "vod_content": content, + "vod_remarks": " · ".join([item for item in parts if item]), + "vod_year": parts[0] if len(parts) > 0 else "", + "vod_area": parts[1] if len(parts) > 1 else "", + "vod_class": parts[2] if len(parts) > 2 else "", + "vod_director": director, + "vod_actor": actor, + "vod_play_from": "$$$".join([item["from"] for item in play_groups]), + "vod_play_url": "$$$".join([item["urls"] for item in play_groups]), + } + + def detailContent(self, ids): + result = {"list": []} + for raw_id in ids: + vod_id = self._stringify(raw_id).strip() + if not vod_id: + continue + vod = self._parse_detail_page(self._request_html(vod_id), vod_id) + result["list"].append(vod) + return result