From f2bc9519035c398d8ca7562e7c11de02fd6c1ec4 Mon Sep 17 00:00:00 2001
From: xiaoxia <dev@xiaoxiajianji.com>
Date: Thu, 17 Sep 2026 16:39:49 +0800
Subject: [PATCH] fix: add A0 path using App Feed API for Douyin extraction

---
 apps/api/app/api/routes/scripts_ai.py | 65 ++++++++++++++++++---------
 1 file changed, 45 insertions(+), 20 deletions(-)

diff --git a/apps/api/app/api/routes/scripts_ai.py b/apps/api/app/api/routes/scripts_ai.py
index 2fd9078a1..929560698 100644
--- a/apps/api/app/api/routes/scripts_ai.py
+++ b/apps/api/app/api/routes/scripts_ai.py
@@ -356,7 +356,7 @@ def _fetch_ttwid():
     # 方式一：直接调 ttwid 注册接口
     try:
         with httpx.Client(timeout=8, follow_redirects=True, verify=False) as http:
-            http.post(
+            r = http.post(
                 "https://ttwid.bytedance.com/ttwid/union/register/",
                 json={
                     "region": "cn",
@@ -769,7 +769,7 @@ def _ytdlp_download_and_local_asr(page_url, temp_dir, cookiefile=None):
         cookie_files_to_try.append(cookiefile)
 
     # 2. 合成 cookie 重试（复用 _ytdlp_extract 的 cookie 生成逻辑，8次重试）
-    import time as _time
+    import secrets, time as _time
     ttwid = _fetch_ttwid()
     for i in range(8):
         if i % 3 == 0:
@@ -945,10 +945,9 @@ def _html_scrape_direct_url(page_url, timeout=15):
 @router.get("/douyin/__debug_diag")
 def douyin_diag():
     """[Staging only] 容器内网络/yt-dlp诊断"""
-    import tempfile as _tf
-    import time as _t
-
     import httpx as _httpx
+    import time as _t
+    import tempfile as _tf
     import yt_dlp as _ytdlp
 
     results = {}
@@ -1085,6 +1084,22 @@ def douyin_diag():
 
     # 6. yt-dlp version
     results["ytdlp_version"] = _ytdlp.version.__version__
+
+    # 7. App Feed API 直连测试（路径 A0）
+    try:
+        from packages.douyin_parser import fetch_douyin_video_url as _feed_fetch
+        ft0 = _t.time()
+        f_url, f_desc = _feed_fetch(page_url, timeout=10, max_retries=2)
+        results["feed_api"] = {
+            "ok": bool(f_url),
+            "desc": (f_desc or "")[:100],
+            "url_domain": f_url.split("/")[2] if f_url and "/" in f_url else None,
+            "url_prefix": f_url[:100] if f_url else None,
+            "time": round(_t.time() - ft0, 2),
+        }
+    except Exception as e:
+        results["feed_api"] = {"ok": False, "error": str(e)[:200]}
+
     results["total_time"] = round(_t.time() - t0, 2)
     return results
 
@@ -1101,6 +1116,18 @@ def extract_from_douyin(
     page_url = _canonicalize_douyin_url(page_url)
     _dbg("page_url", page_url)
 
+    # 路径 A0：抖音 App Feed API 直连（零依赖、无需 cookies/签名、最稳定，优先级最高）
+    # 使用 Android 客户端 aid=1128 接口，直接返回 aweme_list 含无水印直链
+    direct_url = None
+    feed_desc = ""
+    try:
+        from packages.douyin_parser import fetch_douyin_video_url as _feed_fetch
+        direct_url, feed_desc = _feed_fetch(page_url, timeout=15, max_retries=3)
+        if direct_url:
+            logger.info("抖音 App Feed API 直连成功: url=%s play_len=%d", page_url, len(direct_url))
+    except Exception as exc:  # noqa: BLE001
+        logger.warning("抖音 App Feed API 异常: %s", exc)
+
     text = ""
     duration = 0.0
     cookiefile = _resolve_cookies_file()
@@ -1108,23 +1135,21 @@ def extract_from_douyin(
 
     # 整体重试：外层最多 2 轮完整链路，应对第三方 API 瞬时限流/CDN 抖动
     max_rounds = 2
-    direct_url = None
     meta_duration = 0.0
     for round_idx in range(max_rounds):
-        direct_url = None
-        meta_duration = 0.0
-
-        # 路径 A1：第三方解析 API（最稳定，apizero 付费时 99.88% 可用）
-        logger.info("抖音解析第%d轮开始: url=%s", round_idx + 1, page_url)
-        try:
-            direct_url, meta_duration = _parser_extract_direct_url(page_url)
-            if direct_url:
-                logger.info(
-                    "第三方解析 API 成功（第%d轮）: url_domain=%s",
-                    round_idx + 1, direct_url.split("/")[2] if "/" in direct_url else "?",
-                )
-        except Exception as exc:  # noqa: BLE001
-            logger.warning("第三方解析 API 异常（第%d轮）: %s", round_idx + 1, exc)
+        # 路径 A0 已成功则跳过 A1/A2/A3；否则走兜底链路
+        if not direct_url:
+            # 路径 A1：第三方解析 API（apizero 付费时可用）
+            logger.info("抖音解析第%d轮开始: url=%s a0_ok=%s", round_idx + 1, page_url, bool(direct_url))
+            try:
+                direct_url, meta_duration = _parser_extract_direct_url(page_url)
+                if direct_url:
+                    logger.info(
+                        "第三方解析 API 成功（第%d轮）: url_domain=%s",
+                        round_idx + 1, direct_url.split("/")[2] if "/" in direct_url else "?",
+                    )
+            except Exception as exc:  # noqa: BLE001
+                logger.warning("第三方解析 API 异常（第%d轮）: %s", round_idx + 1, exc)
 
         # 路径 A2：yt-dlp 拿直链（作为兜底，需要 cookie 概率性成功）
         if not direct_url: