From 0ef3c107670f023a581c69ce5d57396a01353152 Mon Sep 17 00:00:00 2001
From: Harold <8866033@gmail.com>
Date: Fri, 24 Apr 2026 16:17:20 +0800
Subject: [PATCH] feat: add PPnix spider
---
py/PPnix.py | 260 +++++++++
.../plans/2026-04-24-ppnix-spider.md | 550 ++++++++++++++++++
.../specs/2026-04-24-ppnix-spider-design.md | 263 +++++++++
py/tests/test_PPnix.py | 171 ++++++
4 files changed, 1244 insertions(+)
create mode 100644 py/PPnix.py
create mode 100644 py/docs/superpowers/plans/2026-04-24-ppnix-spider.md
create mode 100644 py/docs/superpowers/specs/2026-04-24-ppnix-spider-design.md
create mode 100644 py/tests/test_PPnix.py
diff --git a/py/PPnix.py b/py/PPnix.py
new file mode 100644
index 0000000..46f4b92
--- /dev/null
+++ b/py/PPnix.py
@@ -0,0 +1,260 @@
+# coding=utf-8
+import re
+import sys
+from urllib.parse import quote, unquote
+
+from bs4 import BeautifulSoup
+
+from base.spider import Spider as BaseSpider
+
+sys.path.append("..")
+
+
+class Spider(BaseSpider):
+ def __init__(self):
+ self.name = "PPnix[采]"
+ self.host = "https://www.ppnix.com"
+ self.lang_path = "/cn"
+ self.headers = {
+ "User-Agent": (
+ "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+ "AppleWebKit/537.36 (KHTML, like Gecko) "
+ "Chrome/145.0.0.0 Safari/537.36"
+ ),
+ "Referer": self.host + self.lang_path + "/",
+ }
+ self.classes = [
+ {"type_id": "1", "type_name": "电影"},
+ {"type_id": "2", "type_name": "电视剧"},
+ ]
+ sort_values = [
+ {"n": "按时间", "v": "time"},
+ {"n": "按人气", "v": "hits"},
+ {"n": "按评分", "v": "score"},
+ ]
+ self.filters = {
+ "1": [
+ {
+ "key": "class",
+ "name": "类型",
+ "init": "",
+ "value": [
+ {"n": "全部", "v": ""},
+ {"n": "动作", "v": "动作"},
+ {"n": "喜剧", "v": "喜剧"},
+ {"n": "剧情", "v": "剧情"},
+ ],
+ },
+ {"key": "by", "name": "排序", "init": "time", "value": list(sort_values)},
+ ],
+ "2": [
+ {
+ "key": "class",
+ "name": "类型",
+ "init": "",
+ "value": [
+ {"n": "全部", "v": ""},
+ {"n": "爱情", "v": "爱情"},
+ {"n": "古装", "v": "古装"},
+ {"n": "悬疑", "v": "悬疑"},
+ ],
+ },
+ {"key": "by", "name": "排序", "init": "time", "value": list(sort_values)},
+ ],
+ }
+
+ def init(self, extend=""):
+ return None
+
+ def getName(self):
+ return self.name
+
+ def homeContent(self, filter):
+ return {"class": self.classes, "filters": self.filters}
+
+ def _build_url(self, value):
+ raw = str(value or "").strip()
+ if not raw:
+ return self.host + self.lang_path + "/"
+ if raw.startswith(("http://", "https://")):
+ return raw
+ if raw.startswith("/"):
+ return self.host + raw
+ return self.host + "/" + raw
+
+ def _map_type_slug(self, tid):
+ return "tv" if str(tid) == "2" else "movie"
+
+ def _map_sort(self, value):
+ mapping = {"time": "newstime", "hits": "onclick", "score": "rating"}
+ return mapping.get(str(value or "time"), "newstime")
+
+ def _build_category_url(self, tid, pg, extend):
+ values = extend if isinstance(extend, dict) else {}
+ slug = self._map_type_slug(tid)
+ genre = str(values.get("class") or "").strip()
+ page = max(int(pg), 1)
+ page_part = "" if page <= 1 else str(page - 1)
+ sort = self._map_sort(values.get("by", "time"))
+ return self._build_url(f"{self.lang_path}/{slug}/{genre}---{page_part}-{sort}.html")
+
+ def _build_search_url(self, keyword, pg):
+ page = max(int(pg), 1)
+ suffix = "" if page <= 1 else f"-page-{page}"
+ encoded = quote(str(keyword or "").strip())
+ return self._build_url(f"{self.lang_path}/search/{encoded}--.html{suffix}")
+
+ def _text(self, node):
+ return node.get_text(" ", strip=True) if node else ""
+
+ def _request_html(self, path_or_url, referer=None, extra_headers=None):
+ target = path_or_url if str(path_or_url).startswith("http") else self._build_url(path_or_url)
+ headers = dict(self.headers)
+ headers["Referer"] = referer or self.headers["Referer"]
+ if isinstance(extra_headers, dict):
+ headers.update(extra_headers)
+ response = self.fetch(target, headers=headers, timeout=10)
+ if response.status_code != 200:
+ return ""
+ return response.text or ""
+
+ def _dedupe(self, items):
+ seen = set()
+ result = []
+ for item in items:
+ vod_id = str(item.get("vod_id") or "").strip()
+ if not vod_id or vod_id in seen:
+ continue
+ seen.add(vod_id)
+ result.append(item)
+ return result
+
+ def _parse_cards(self, html):
+ soup = BeautifulSoup(html or "", "html.parser")
+ items = []
+ for node in soup.select("li"):
+ anchor = node.select_one("a.thumbnail") or node.select_one("h2 a") or node.select_one("a")
+ image = node.select_one("img.thumb")
+ if not anchor:
+ continue
+ href = str(anchor.get("href") or "").strip()
+ if not re.search(r"/cn/(movie|tv)/\d+\.html", href):
+ continue
+ vod_name = str((image.get("alt") if image else "") or self._text(anchor)).strip()
+ if not vod_name:
+ continue
+ items.append(
+ {
+ "vod_id": href.replace("/cn/", "").lstrip("/"),
+ "vod_name": vod_name,
+ "vod_pic": self._build_url((image.get("src") if image else "") or ""),
+ "vod_remarks": self._text(node.select_one("footer .rate") or node.select_one("footer")),
+ }
+ )
+ return items
+
+ def _extract_m3u8_items(self, html):
+ body = str(html or "")
+ info_match = re.search(r"infoid\s*=\s*(\d+)", body)
+ list_match = re.search(r"m3u8\s*=\s*\[(.*?)\]", body, re.S)
+ raw_list = list_match.group(1) if list_match else ""
+ items = []
+ for single, double in re.findall(r"'([^']*)'|\"([^\"]*)\"", raw_list):
+ value = str(single or double or "").strip()
+ if value:
+ items.append(value)
+ return {"info_id": info_match.group(1) if info_match else "", "items": items}
+
+ def _build_play_id(self, info_id, param):
+ return f"{str(info_id or '').strip()}|{quote(str(param or '').strip())}"
+
+ def _parse_play_id(self, play_id):
+ parts = str(play_id or "").split("|", 1)
+ if len(parts) != 2:
+ return {"info_id": "", "param": ""}
+ info_id = parts[0].strip()
+ param = unquote(parts[1].strip())
+ if not info_id or not param:
+ return {"info_id": "", "param": ""}
+ return {"info_id": info_id, "param": param}
+
+ def _extract_excerpt_text(self, soup, label):
+ for node in soup.select(".product-excerpt"):
+ text = self._text(node)
+ if text.startswith(label) or label in text:
+ return text.replace(label, "", 1).strip()
+ return ""
+
+ def homeVideoContent(self):
+ html = self._request_html(self.lang_path + "/")
+ soup = BeautifulSoup(html or "", "html.parser")
+ blocks = soup.select(".lists-content ul")
+ items = []
+ if len(blocks) > 0:
+ items.extend(self._parse_cards(str(blocks[0])))
+ if len(blocks) > 1:
+ items.extend(self._parse_cards(str(blocks[1])))
+ return {"list": self._dedupe(items)[:20]}
+
+ def categoryContent(self, tid, pg, filter, extend):
+ page = max(int(pg), 1)
+ slug = self._map_type_slug(tid)
+ html = self._request_html(self._build_category_url(tid, pg, extend or {}))
+ items = [item for item in self._parse_cards(html) if item["vod_id"].startswith(slug + "/")]
+ return {"page": page, "limit": 24, "total": page * 24 + len(items), "list": items}
+
+ def searchContent(self, key, quick, pg="1"):
+ page = max(int(pg), 1)
+ items = self._parse_cards(self._request_html(self._build_search_url(key, pg)))
+ if str(quick).lower() == "true":
+ items = items[:10]
+ return {"page": page, "limit": 24, "total": page * 24 + len(items), "list": items}
+
+ def detailContent(self, ids):
+ vod_id = str(ids[0] if isinstance(ids, list) and ids else ids or "").strip().lstrip("/")
+ if not vod_id:
+ return {"list": []}
+ html = self._request_html(f"{self.lang_path}/{vod_id}")
+ if not html:
+ return {"list": []}
+ soup = BeautifulSoup(html, "html.parser")
+ title_text = self._text(soup.select_one("h1.product-title") or soup.select_one("title"))
+ info = self._extract_m3u8_items(html)
+ play_urls = []
+ for item in info["items"]:
+ play_urls.append(f"{item}${self._build_play_id(info['info_id'], item)}")
+ actor = self._extract_excerpt_text(soup, "主演：").replace(" / ", ",").replace("/", ",")
+ vod_name = re.sub(r"\s*\([^)]*\)\s*$", "", title_text).strip()
+ year_match = re.search(r"\((\d{4})\)", title_text)
+ return {
+ "list": [
+ {
+ "vod_id": vod_id,
+ "vod_name": vod_name,
+ "vod_pic": self._build_url((soup.select_one("header.product-header img.thumb") or {}).get("src", "")),
+ "type_name": "电视剧" if vod_id.startswith("tv/") else "电影",
+ "vod_year": year_match.group(1) if year_match else "",
+ "vod_director": self._extract_excerpt_text(soup, "导演："),
+ "vod_actor": actor,
+ "vod_content": self._extract_excerpt_text(soup, "简介："),
+ "vod_play_from": "PPnix" if play_urls else "",
+ "vod_play_url": "#".join(play_urls),
+ }
+ ]
+ }
+
+ def playerContent(self, flag, id, vipFlags):
+ meta = self._parse_play_id(id)
+ if not meta["info_id"] or not meta["param"]:
+ return {"parse": 1, "jx": 1, "url": str(id or ""), "header": {}}
+ encoded = quote(meta["param"])
+ return {
+ "parse": 0,
+ "jx": 0,
+ "url": self._build_url(f"/info/m3u8/{meta['info_id']}/{encoded}.m3u8"),
+ "header": {
+ "Referer": self.host + self.lang_path + "/",
+ "Origin": self.host,
+ "User-Agent": self.headers["User-Agent"],
+ },
+ }
diff --git a/py/docs/superpowers/plans/2026-04-24-ppnix-spider.md b/py/docs/superpowers/plans/2026-04-24-ppnix-spider.md
new file mode 100644
index 0000000..2268041
--- /dev/null
+++ b/py/docs/superpowers/plans/2026-04-24-ppnix-spider.md
@@ -0,0 +1,550 @@
+# PPnix Spider Implementation Plan
+
+> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.
+
+**Goal:** Build a PPnix spider in `py/PPnix.py` with static categories and stable filters, deterministic HTML parsing, short IDs, direct m3u8 playback, and matching unit tests.
+
+**Architecture:** Keep the spider in one file following the repository's existing adapter pattern. Use small parsing helpers for URL building, card extraction, detail metadata extraction, and compact play-ID encoding so each behavior can be tested in isolation with embedded HTML fixtures.
+
+**Tech Stack:** Python 3.14, `unittest`, `unittest.mock`, `requests` via `base.spider.Spider.fetch`, `beautifulsoup4`, `urllib.parse`, `re`
+
+---
+
+### Task 1: Add failing tests for static metadata, URL builders, and card parsing
+
+**Files:**
+- Create: `py/tests/test_PPnix.py`
+- Create: `py/PPnix.py`
+- Test: `py/tests/test_PPnix.py`
+
+- [ ] **Step 1: Write the failing tests**
+
+```python
+import unittest
+from importlib.machinery import SourceFileLoader
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[1]
+MODULE = SourceFileLoader("ppnix_spider", str(ROOT / "PPnix.py")).load_module()
+Spider = MODULE.Spider
+
+
+class TestPPnixSpider(unittest.TestCase):
+ def setUp(self):
+ Spider._instance = None
+ self.spider = Spider()
+ self.spider.init()
+
+ def test_home_content_exposes_expected_categories_and_filter_keys(self):
+ content = self.spider.homeContent(False)
+ self.assertEqual([item["type_id"] for item in content["class"]], ["1", "2"])
+ self.assertEqual([item["key"] for item in content["filters"]["1"]], ["class", "by"])
+ self.assertEqual([item["key"] for item in content["filters"]["2"]], ["class", "by"])
+
+ def test_build_category_url_maps_first_page_and_sort_values(self):
+ self.assertEqual(
+ self.spider._build_category_url("1", "1", {}),
+ "https://www.ppnix.com/cn/movie/----newstime.html",
+ )
+ self.assertEqual(
+ self.spider._build_category_url("2", "3", {"class": "爱情", "by": "hits"}),
+ "https://www.ppnix.com/cn/tv/爱情---2-onclick.html",
+ )
+
+ def test_build_search_url_uses_ppnix_pattern(self):
+ self.assertEqual(
+ self.spider._build_search_url("繁花", "1"),
+ "https://www.ppnix.com/cn/search/%E7%B9%81%E8%8A%B1--.html",
+ )
+ self.assertEqual(
+ self.spider._build_search_url("繁花", "2"),
+ "https://www.ppnix.com/cn/search/%E7%B9%81%E8%8A%B1--.html-page-2",
+ )
+
+ def test_parse_cards_extracts_short_vod_id_title_cover_and_remarks(self):
+ html = '''
+
+ '''
+ self.assertEqual(
+ self.spider._parse_cards(html),
+ [
+ {
+ "vod_id": "movie/123.html",
+ "vod_name": "示例影片",
+ "vod_pic": "https://www.ppnix.com/poster.jpg",
+ "vod_remarks": "HD",
+ }
+ ],
+ )
+```
+
+- [ ] **Step 2: Run test to verify it fails**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: FAIL because `py/PPnix.py` does not exist yet or required methods are missing
+
+- [ ] **Step 3: Write minimal implementation**
+
+```python
+# coding=utf-8
+import re
+import sys
+from urllib.parse import quote
+
+from bs4 import BeautifulSoup
+from base.spider import Spider as BaseSpider
+
+sys.path.append("..")
+
+
+class Spider(BaseSpider):
+ def __init__(self):
+ self.name = "PPnix[采]"
+ self.host = "https://www.ppnix.com"
+ self.lang_path = "/cn"
+ self.headers = {
+ "User-Agent": (
+ "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
+ "AppleWebKit/537.36 (KHTML, like Gecko) "
+ "Chrome/145.0.0.0 Safari/537.36"
+ ),
+ "Referer": self.host + self.lang_path + "/",
+ }
+ self.classes = [
+ {"type_id": "1", "type_name": "电影"},
+ {"type_id": "2", "type_name": "电视剧"},
+ ]
+ self.filters = {
+ "1": [
+ {"key": "class", "name": "类型", "init": "", "value": [{"n": "全部", "v": ""}, {"n": "动作", "v": "动作"}]},
+ {"key": "by", "name": "排序", "init": "time", "value": [{"n": "按时间", "v": "time"}, {"n": "按人气", "v": "hits"}, {"n": "按评分", "v": "score"}]},
+ ],
+ "2": [
+ {"key": "class", "name": "类型", "init": "", "value": [{"n": "全部", "v": ""}, {"n": "爱情", "v": "爱情"}]},
+ {"key": "by", "name": "排序", "init": "time", "value": [{"n": "按时间", "v": "time"}, {"n": "按人气", "v": "hits"}, {"n": "按评分", "v": "score"}]},
+ ],
+ }
+
+ def init(self, extend=""):
+ return None
+
+ def getName(self):
+ return self.name
+
+ def homeContent(self, filter):
+ return {"class": self.classes, "filters": self.filters}
+
+ def _build_url(self, value):
+ raw = str(value or "").strip()
+ if not raw:
+ return self.host + self.lang_path + "/"
+ if raw.startswith(("http://", "https://")):
+ return raw
+ if raw.startswith("/"):
+ return self.host + raw
+ return self.host + "/" + raw
+
+ def _map_type_slug(self, tid):
+ return "tv" if str(tid) == "2" else "movie"
+
+ def _map_sort(self, value):
+ return {"time": "newstime", "hits": "onclick", "score": "rating"}.get(str(value or "time"), "newstime")
+
+ def _build_category_url(self, tid, pg, extend):
+ extend = extend or {}
+ slug = self._map_type_slug(tid)
+ genre = str(extend.get("class", "") or "")
+ page = max(int(pg), 1)
+ page_part = "" if page <= 1 else str(page - 1)
+ sort = self._map_sort(extend.get("by", "time"))
+ return self._build_url(f"{self.lang_path}/{slug}/{genre}---{page_part}-{sort}.html")
+
+ def _build_search_url(self, keyword, pg):
+ page = max(int(pg), 1)
+ suffix = "" if page <= 1 else f"-page-{page}"
+ return self._build_url(f"{self.lang_path}/search/{quote(str(keyword or '').strip())}--.html{suffix}")
+
+ def _parse_cards(self, html):
+ soup = BeautifulSoup(html or "", "html.parser")
+ items = []
+ for node in soup.select("li"):
+ anchor = node.select_one("a.thumbnail") or node.select_one("h2 a") or node.select_one("a")
+ image = node.select_one("img.thumb")
+ if not anchor:
+ continue
+ href = (anchor.get("href") or "").strip()
+ match = re.search(r"/cn/(movie|tv)/\d+\.html", href)
+ if not match:
+ continue
+ items.append(
+ {
+ "vod_id": href.replace("/cn/", "").lstrip("/"),
+ "vod_name": ((image.get("alt") if image else "") or anchor.get_text(" ", strip=True)).strip(),
+ "vod_pic": self._build_url((image.get("src") if image else "") or ""),
+ "vod_remarks": (node.select_one("footer .rate") or node.select_one("footer") or "").get_text(" ", strip=True),
+ }
+ )
+ return items
+```
+
+- [ ] **Step 4: Run test to verify it passes**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: PASS for the four tests above
+
+- [ ] **Step 5: Commit**
+
+```bash
+git add py/PPnix.py py/tests/test_PPnix.py
+git commit -m "feat: add PPnix spider skeleton"
+```
+
+### Task 2: Add failing tests for homepage, category, and search flows
+
+**Files:**
+- Modify: `py/tests/test_PPnix.py`
+- Modify: `py/PPnix.py`
+- Test: `py/tests/test_PPnix.py`
+
+- [ ] **Step 1: Write the failing tests**
+
+```python
+from unittest.mock import patch
+
+ @patch.object(Spider, "_request_html")
+ def test_home_video_content_merges_movie_and_tv_cards(self, mock_request_html):
+ mock_request_html.return_value = '''
+
+
+ 
+
+
+ 
+
+ 
+ 
示例剧 (2025)
+
+ 导演：导演甲
+ 主演：演员甲 / 演员乙
+ 简介：一段剧情简介
+
+ '''
+ result = self.spider.detailContent(["tv/8899.html"])
+ vod = result["list"][0]
+ self.assertEqual(vod["vod_id"], "tv/8899.html")
+ self.assertEqual(vod["vod_name"], "示例剧")
+ self.assertEqual(vod["vod_pic"], "https://www.ppnix.com/poster.jpg")
+ self.assertEqual(vod["vod_year"], "2025")
+ self.assertEqual(vod["vod_director"], "导演甲")
+ self.assertEqual(vod["vod_actor"], "演员甲,演员乙")
+ self.assertEqual(vod["vod_content"], "一段剧情简介")
+ self.assertEqual(vod["vod_play_from"], "PPnix")
+ self.assertEqual(vod["vod_play_url"], "第1集$8899|%E7%AC%AC1%E9%9B%86#第2集$8899|%E7%AC%AC2%E9%9B%86")
+
+ def test_player_content_returns_direct_m3u8_url(self):
+ result = self.spider.playerContent("PPnix", "8899|%E7%AC%AC1%E9%9B%86", {})
+ self.assertEqual(result["parse"], 0)
+ self.assertEqual(result["jx"], 0)
+ self.assertEqual(result["url"], "https://www.ppnix.com/info/m3u8/8899/%E7%AC%AC1%E9%9B%86.m3u8")
+ self.assertEqual(result["header"]["Origin"], "https://www.ppnix.com")
+
+ def test_player_content_falls_back_when_play_id_is_invalid(self):
+ result = self.spider.playerContent("PPnix", "broken", {})
+ self.assertEqual(result["parse"], 1)
+ self.assertEqual(result["jx"], 1)
+ self.assertEqual(result["url"], "broken")
+```
+
+- [ ] **Step 2: Run test to verify it fails**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: FAIL because detail and player helpers are still missing
+
+- [ ] **Step 3: Write minimal implementation**
+
+```python
+from urllib.parse import quote, unquote
+
+ def _extract_m3u8_items(self, html):
+ body = html or ""
+ info_match = re.search(r"infoid\s*=\s*(\d+)", body)
+ list_match = re.search(r"m3u8\s*=\s*\[(.*?)\]", body, re.S)
+ items = re.findall(r"'([^']*)'|\"([^\"]*)\"", list_match.group(1) if list_match else "")
+ return {
+ "info_id": info_match.group(1) if info_match else "",
+ "items": [a or b for a, b in items if (a or b).strip()],
+ }
+
+ def _build_play_id(self, info_id, param):
+ return f"{str(info_id).strip()}|{quote(str(param).strip())}"
+
+ def _parse_play_id(self, play_id):
+ parts = str(play_id or "").split("|", 1)
+ if len(parts) != 2 or not parts[0] or not parts[1]:
+ return {"info_id": "", "param": ""}
+ return {"info_id": parts[0].strip(), "param": unquote(parts[1].strip())}
+
+ def detailContent(self, ids):
+ vod_id = str(ids[0] if isinstance(ids, list) and ids else ids or "").strip().lstrip("/")
+ if not vod_id:
+ return {"list": []}
+ html = self._request_html(self._build_url("/" + self.lang_path.strip("/") + "/" + vod_id))
+ soup = BeautifulSoup(html or "", "html.parser")
+ title_raw = (soup.select_one("h1.product-title") or soup.select_one("title"))
+ title_text = title_raw.get_text(" ", strip=True) if title_raw else ""
+ info = self._extract_m3u8_items(html)
+ play_urls = []
+ for item in info["items"]:
+ play_urls.append(f"{item}${self._build_play_id(info['info_id'], item)}")
+ return {
+ "list": [
+ {
+ "vod_id": vod_id,
+ "vod_name": re.sub(r"\s*\([^)]*\)\s*$", "", title_text).strip(),
+ "vod_pic": self._build_url((soup.select_one('header.product-header img.thumb') or {}).get("src", "")),
+ "type_name": "电视剧" if vod_id.startswith("tv/") else "电影",
+ "vod_year": (re.search(r"\((\d{4})\)", title_text) or [None, ""])[1],
+ "vod_director": self._extract_excerpt_text(soup, "导演："),
+ "vod_actor": self._extract_excerpt_text(soup, "主演：").replace(" / ", ","),
+ "vod_content": self._extract_excerpt_text(soup, "简介："),
+ "vod_play_from": "PPnix" if play_urls else "",
+ "vod_play_url": "#".join(play_urls),
+ }
+ ]
+ }
+
+ def _extract_excerpt_text(self, soup, label):
+ for node in soup.select(".product-excerpt"):
+ text = node.get_text(" ", strip=True)
+ if label in text:
+ return text.replace(label, "", 1).strip()
+ return ""
+
+ def playerContent(self, flag, id, vipFlags):
+ meta = self._parse_play_id(id)
+ if not meta["info_id"] or not meta["param"]:
+ return {"parse": 1, "jx": 1, "url": str(id or ""), "header": {}}
+ encoded = quote(meta["param"])
+ return {
+ "parse": 0,
+ "jx": 0,
+ "url": self._build_url(f"/info/m3u8/{meta['info_id']}/{encoded}.m3u8"),
+ "header": {
+ "Referer": self.host + self.lang_path + "/",
+ "Origin": self.host,
+ "User-Agent": self.headers["User-Agent"],
+ },
+ }
+```
+
+- [ ] **Step 4: Run test to verify it passes**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: PASS for detail and player tests
+
+- [ ] **Step 5: Commit**
+
+```bash
+git add py/PPnix.py py/tests/test_PPnix.py
+git commit -m "feat: add PPnix detail and player parsing"
+```
+
+### Task 4: Run focused verification and repo-style cleanup
+
+**Files:**
+- Modify: `py/PPnix.py`
+- Modify: `py/tests/test_PPnix.py`
+- Test: `py/tests/test_PPnix.py`
+
+- [ ] **Step 1: Run the focused PPnix test module**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: all PPnix tests pass
+
+- [ ] **Step 2: Tighten static filter values and helper names if test-driven cleanup is needed**
+
+```python
+self.filters = {
+ "1": [
+ {
+ "key": "class",
+ "name": "类型",
+ "init": "",
+ "value": [{"n": "全部", "v": ""}, {"n": "动作", "v": "动作"}, {"n": "喜剧", "v": "喜剧"}, {"n": "剧情", "v": "剧情"}],
+ },
+ {
+ "key": "by",
+ "name": "排序",
+ "init": "time",
+ "value": [{"n": "按时间", "v": "time"}, {"n": "按人气", "v": "hits"}, {"n": "按评分", "v": "score"}],
+ },
+ ],
+ "2": [
+ {
+ "key": "class",
+ "name": "类型",
+ "init": "",
+ "value": [{"n": "全部", "v": ""}, {"n": "爱情", "v": "爱情"}, {"n": "古装", "v": "古装"}, {"n": "悬疑", "v": "悬疑"}],
+ },
+ {
+ "key": "by",
+ "name": "排序",
+ "init": "time",
+ "value": [{"n": "按时间", "v": "time"}, {"n": "按人气", "v": "hits"}, {"n": "按评分", "v": "score"}],
+ },
+ ],
+}
+```
+
+- [ ] **Step 3: Re-run the focused PPnix test module**
+
+Run: `uv run python -m unittest py/tests/test_PPnix.py -v`
+Expected: all PPnix tests still pass after cleanup
+
+- [ ] **Step 4: Commit**
+
+```bash
+git add py/PPnix.py py/tests/test_PPnix.py
+git commit -m "test: verify PPnix spider behavior"
+```
diff --git a/py/docs/superpowers/specs/2026-04-24-ppnix-spider-design.md b/py/docs/superpowers/specs/2026-04-24-ppnix-spider-design.md
new file mode 100644
index 0000000..ccfc317
--- /dev/null
+++ b/py/docs/superpowers/specs/2026-04-24-ppnix-spider-design.md
@@ -0,0 +1,263 @@
+# PPnix Spider Design
+
+## Goal
+
+Implement a new Python spider at `py/PPnix.py` for PPnix based on the provided Node reference, adapted to this repository's existing `Spider` interface and testing conventions.
+
+Scope for this iteration:
+
+- Support two categories: movie and tv
+- Expose stable filters only: type/class and sort/by
+- Parse homepage recommendations
+- Parse category listings
+- Parse keyword search
+- Parse detail pages and playlist data
+- Return direct playback URLs from PPnix m3u8 endpoints
+
+Explicitly out of scope for this iteration:
+
+- Dynamic filter discovery and caching
+- TVBox-specific `data:m3u8` rewriting
+- Subtitle aggregation in `playerContent`
+- JavaScript server route parity with the Node plugin
+
+## Target API Shape
+
+The spider will implement the standard methods used in this repo:
+
+- `homeContent(filter)`
+- `homeVideoContent()`
+- `categoryContent(tid, pg, filter, extend)`
+- `searchContent(key, quick, pg="1")`
+- `detailContent(ids)`
+- `playerContent(flag, id, vipFlags)`
+
+The spider will follow current repo conventions:
+
+- Single-file site adapter
+- Deterministic parsing helpers
+- Short site-local IDs rather than full URLs
+- No `pagecount` field in list/search responses
+
+## Data Model
+
+### Categories
+
+Use stable repo-style numeric category IDs instead of the Node plugin's string IDs:
+
+- `1` => movie
+- `2` => tv
+
+Display names:
+
+- `1` => `电影`
+- `2` => `电视剧`
+
+### Filters
+
+Only expose stable filters:
+
+- `class`
+- `by`
+
+For both category IDs, `homeContent` will return:
+
+- `class`: values extracted from the Node reference and hard-coded per category
+- `by`: `time`, `hits`, `score`
+
+Sort mapping for request URLs:
+
+- `time` => `newstime`
+- `hits` => `onclick`
+- `score` => `rating`
+
+Default sort:
+
+- `time`
+
+`filter_def` caching is unnecessary because filters are static.
+
+### IDs
+
+To match repository guidance, the spider will keep IDs short.
+
+- List/detail `vod_id`: relative detail path such as `movie/123.html` or `tv/456.html`
+- Play IDs: compact encoded payload containing only `infoId` and `param`
+
+Recommended play ID format:
+
+- `infoId|urlencoded(param)`
+
+This is sufficient because the final playback URL is deterministic:
+
+- `https://www.ppnix.com/info/m3u8/{infoId}/{encoded_param}.m3u8`
+
+No referer or display name needs to be stored in the ID for this scope.
+
+## Architecture
+
+`PPnix.py` will keep logic in a single class with focused helpers.
+
+### Core helpers
+
+- `_build_url(path)`
+ - Normalizes relative paths against `https://www.ppnix.com`
+ - Supports raw absolute URLs and root-relative paths
+
+- `_request_html(path_or_url, referer=None, extra_headers=None)`
+ - Performs GET requests through `fetch`
+ - Applies browser-like headers
+ - Returns response text or empty string on non-200
+
+- `_clean_text(value)`
+ - Strips HTML whitespace noise and entities used in visible text
+
+- `_fix_image(url)`
+ - Converts relative poster URLs to absolute URLs
+
+### List/category/search helpers
+
+- `_map_type_slug(tid)`
+ - `1 -> movie`, `2 -> tv`
+
+- `_build_category_url(tid, pg, extend)`
+ - Converts repo category/filter input into PPnix path format
+ - First page omits the page index segment
+ - Path format:
+ - `/cn/{type_slug}/{genre}---{page_index}-{sort}.html`
+ - For this scope, unsupported filters are ignored
+
+- `_build_search_url(keyword, pg)`
+ - Uses the Node reference pattern:
+ - `/cn/search/{encoded_keyword}--.html`
+ - page > 1 appends `-page-{pg}`
+
+- `_parse_cards(html, type_hint="")`
+ - Parses list cards from homepage/category/search HTML
+ - Extracts:
+ - `vod_id`
+ - `vod_name`
+ - `vod_pic`
+ - `vod_remarks`
+ - Filters out malformed entries
+ - Deduplicates by `vod_id`
+
+### Detail/play helpers
+
+- `_extract_m3u8_items(html)`
+ - Reads:
+ - `infoid = 123`
+ - `m3u8 = [...]`
+ - Returns `infoId` plus ordered episode/item labels
+
+- `_build_play_id(info_id, param)`
+ - Returns compact playback ID
+
+- `_parse_play_id(play_id)`
+ - Safely reverses the compact playback ID
+
+- `_parse_detail_page(html, vod_id)`
+ - Extracts title, poster, year, director, actor, content
+ - Determines `type_name` from `vod_id`
+ - Builds a single `PPnix` play group using the extracted m3u8 items
+
+## Request and Parsing Flow
+
+### homeContent
+
+Returns static categories and static filters only.
+
+### homeVideoContent
+
+Requests `/cn/`, parses homepage blocks, merges movie and tv cards, and caps to 20-24 items. The implementation should prefer deterministic extraction over mirroring every homepage section.
+
+### categoryContent
+
+1. Build category URL from `tid`, `pg`, and `extend`
+2. Fetch category page
+3. Parse cards
+4. Filter cards so returned `vod_id` matches the requested type slug
+5. Return:
+ - `page`
+ - `limit`
+ - `total`
+ - `list`
+
+`total` can follow the repo’s common approximate pattern:
+
+- `page * limit + len(list)`
+
+This is acceptable because the site does not expose a clean total count in the provided reference.
+
+### searchContent
+
+1. Build search URL
+2. Parse cards from search result page
+3. Keep only IDs matching `movie/.html` or `tv/.html`
+4. Return `page`, `limit`, `total`, `list`
+
+### detailContent
+
+1. Normalize incoming ID into a relative detail path
+2. Fetch detail page
+3. Parse metadata
+4. Parse `infoid` and `m3u8` item names from inline JS
+5. Build:
+ - `vod_play_from = "PPnix"` if episodes exist
+ - `vod_play_url = "名称$playId#名称$playId..."`
+
+### playerContent
+
+1. Decode `play_id` into `infoId` and `param`
+2. If either is missing, fall back to parse-required response
+3. Build direct source URL:
+ - `https://www.ppnix.com/info/m3u8/{infoId}/{encoded_param}.m3u8`
+4. Return direct-play payload with headers:
+ - `Referer: https://www.ppnix.com/cn/`
+ - `Origin: https://www.ppnix.com`
+ - `User-Agent: browser UA`
+
+No m3u8 text rewriting will be attempted.
+
+## Error Handling
+
+- Request helpers return empty string on non-200 to keep parser code simple
+- Parser helpers should tolerate missing nodes and return empty/default fields
+- `detailContent` returns `{"list": []}` on fetch/parse failure
+- `playerContent` returns parse-required fallback when `play_id` is malformed
+
+This matches the current repo’s defensive style better than raising exceptions.
+
+## Testing Strategy
+
+Tests will be added at `py/tests/test_PPnix.py` using `unittest` and `unittest.mock`.
+
+### Red tests to write first
+
+1. `homeContent` returns categories `1` and `2`, and only stable filter keys
+2. `_build_category_url` maps default and selected sort/class values correctly
+3. `_parse_cards` extracts short IDs, names, posters, and remarks
+4. `homeVideoContent` merges homepage movie/tv sections and truncates result size
+5. `categoryContent` requests the expected PPnix listing URL and returns repo-style page payload without `pagecount`
+6. `searchContent` requests the expected PPnix search URL and filters valid movie/tv IDs
+7. `_extract_m3u8_items` reads `infoid` and ordered item names from inline JS
+8. `detailContent` builds metadata and `PPnix` play group from fixture HTML
+9. `playerContent` returns direct m3u8 URL for a valid compact play ID
+10. `playerContent` falls back to parse-required response when play ID is malformed
+
+### Fixture design
+
+Use embedded HTML snippets rather than live requests. Fixtures should cover:
+
+- Homepage lists
+- Category/search cards
+- Detail metadata block
+- Inline script with `infoid` and `m3u8`
+
+No network access will be used in tests.
+
+## Implementation Notes
+
+- Prefer `BeautifulSoup` because the project already depends on it and many spiders use HTML-string parsing without heavy DOM abstractions
+- Keep helper methods small and deterministic so test failures identify one parser responsibility at a time
+- Avoid introducing caching or feature flags until they are required by a failing test
diff --git a/py/tests/test_PPnix.py b/py/tests/test_PPnix.py
new file mode 100644
index 0000000..2120c5c
--- /dev/null
+++ b/py/tests/test_PPnix.py
@@ -0,0 +1,171 @@
+import unittest
+from importlib.machinery import SourceFileLoader
+from pathlib import Path
+from unittest.mock import patch
+
+
+ROOT = Path(__file__).resolve().parents[1]
+MODULE = SourceFileLoader("ppnix_spider", str(ROOT / "PPnix.py")).load_module()
+Spider = MODULE.Spider
+
+
+class TestPPnixSpider(unittest.TestCase):
+ def setUp(self):
+ Spider._instance = None
+ self.spider = Spider()
+ self.spider.init()
+
+ def test_home_content_exposes_expected_categories_and_filter_keys(self):
+ content = self.spider.homeContent(False)
+ self.assertEqual([item["type_id"] for item in content["class"]], ["1", "2"])
+ self.assertEqual([item["key"] for item in content["filters"]["1"]], ["class", "by"])
+ self.assertEqual([item["key"] for item in content["filters"]["2"]], ["class", "by"])
+
+ def test_build_category_url_maps_first_page_and_sort_values(self):
+ self.assertEqual(
+ self.spider._build_category_url("1", "1", {}),
+ "https://www.ppnix.com/cn/movie/----newstime.html",
+ )
+ self.assertEqual(
+ self.spider._build_category_url("2", "3", {"class": "爱情", "by": "hits"}),
+ "https://www.ppnix.com/cn/tv/爱情---2-onclick.html",
+ )
+
+ def test_build_search_url_uses_ppnix_pattern(self):
+ self.assertEqual(
+ self.spider._build_search_url("繁花", "1"),
+ "https://www.ppnix.com/cn/search/%E7%B9%81%E8%8A%B1--.html",
+ )
+ self.assertEqual(
+ self.spider._build_search_url("繁花", "2"),
+ "https://www.ppnix.com/cn/search/%E7%B9%81%E8%8A%B1--.html-page-2",
+ )
+
+ def test_parse_cards_extracts_short_vod_id_title_cover_and_remarks(self):
+ html = """
+
+ """
+ self.assertEqual(
+ self.spider._parse_cards(html),
+ [
+ {
+ "vod_id": "movie/123.html",
+ "vod_name": "示例影片",
+ "vod_pic": "https://www.ppnix.com/poster.jpg",
+ "vod_remarks": "HD",
+ }
+ ],
+ )
+
+ @patch.object(Spider, "_request_html")
+ def test_home_video_content_merges_movie_and_tv_cards(self, mock_request_html):
+ mock_request_html.return_value = """
+
+
+
+
+
+
+
+
+
示例剧 (2025)
+
+ 导演：导演甲
+ 主演：演员甲 / 演员乙
+ 简介：一段剧情简介
+
+ """
+ result = self.spider.detailContent(["tv/8899.html"])
+ vod = result["list"][0]
+ self.assertEqual(vod["vod_id"], "tv/8899.html")
+ self.assertEqual(vod["vod_name"], "示例剧")
+ self.assertEqual(vod["vod_pic"], "https://www.ppnix.com/poster.jpg")
+ self.assertEqual(vod["vod_year"], "2025")
+ self.assertEqual(vod["vod_director"], "导演甲")
+ self.assertEqual(vod["vod_actor"], "演员甲,演员乙")
+ self.assertEqual(vod["vod_content"], "一段剧情简介")
+ self.assertEqual(vod["vod_play_from"], "PPnix")
+ self.assertEqual(
+ vod["vod_play_url"],
+ "第1集$8899|%E7%AC%AC1%E9%9B%86#第2集$8899|%E7%AC%AC2%E9%9B%86",
+ )
+
+ def test_player_content_returns_direct_m3u8_url(self):
+ result = self.spider.playerContent("PPnix", "8899|%E7%AC%AC1%E9%9B%86", {})
+ self.assertEqual(result["parse"], 0)
+ self.assertEqual(result["jx"], 0)
+ self.assertEqual(result["url"], "https://www.ppnix.com/info/m3u8/8899/%E7%AC%AC1%E9%9B%86.m3u8")
+ self.assertEqual(result["header"]["Origin"], "https://www.ppnix.com")
+
+ def test_player_content_falls_back_when_play_id_is_invalid(self):
+ result = self.spider.playerContent("PPnix", "broken", {})
+ self.assertEqual(result["parse"], 1)
+ self.assertEqual(result["jx"], 1)
+ self.assertEqual(result["url"], "broken")
+
+
+if __name__ == "__main__":
+ unittest.main()
+
+
+