conductor(cs229): Phase 1 Acquire - transcript.json (5397 segments via yt-dlp VTT fallback) + video.log (yt-dlp success for 336MB mp4, R5 verified)

Fix extract_transcript.py: YouTubeTranscriptApi.get_transcript() (not .fetch()). youtube-transcript-api v1.2.4 uses class method get_transcript(video_id), not instance .fetch().

R5 mitigation: yt-dlp's VTT auto-sub extraction works where youtube-transcript-api fails (XML parse error on empty response). 5397 segments recovered.

Add gitignore patterns for video_analysis artifacts: *.mp4, *.vtt (regenerable). video.log intentionally tracked.

conductor/tracks/video_analysis_cs229_building_llms_20260621/artifacts/video.log

@@ -0,0 +1,10 @@
+# yt-dlp log
+# url: https://youtu.be/9vM4p9NN0Ts
+# output: C:\projects\manual_slop\conductor\tracks\video_analysis_cs229_building_llms_20260621\artifacts\video.mp4
+# returncode: 0
+
+stdout:
+
+
+stderr:
+

scripts/tier2/artifacts/video_analysis_campaign_20260621/phase1_acquire_cs229.py

@@ -0,0 +1,117 @@
+"""Phase 1 Acquire driver for video_analysis_cs229_building_llms_20260621.
+
+Strategy: youtube-transcript-api fails for this video (R5: XML parse error on empty response,
+likely a YouTube API restriction). Fall back to yt-dlp's own subtitle extraction.
+"""
+from __future__ import annotations
+
+import json
+import re
+import subprocess
+import sys
+import time
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parents[4]
+sys.path.insert(0, str(ROOT))
+
+from scripts.video_analysis.download_video import download_video
+from scripts.video_analysis.extract_transcript import _fetch_raw_transcript
+
+URL = "https://youtu.be/9vM4p9NN0Ts"
+ARTIFACTS = ROOT / "conductor" / "tracks" / "video_analysis_cs229_building_llms_20260621" / "artifacts"
+ARTIFACTS.mkdir(parents=True, exist_ok=True)
+
+
+def _parse_vtt_segments(vtt_path: Path) -> list[dict]:
+ text = vtt_path.read_text(encoding="utf-8")
+ segments: list[dict] = []
+ pattern = re.compile(r"(\d{2}):(\d{2}):(\d{2})\.(\d{3})\s+-->", re.MULTILINE)
+ blocks = re.split(r"\n\n+", text)
+ for block in blocks:
+  match = pattern.search(block)
+  if not match:
+   continue
+  h, m, s, ms = match.groups()
+  start = int(h) * 3600 + int(m) * 60 + int(s) + int(ms) / 1000.0
+  lines = [ln.strip() for ln in block.split("\n") if ln.strip() and not pattern.match(ln) and "-->" not in ln]
+  text_content = " ".join(lines)
+  if text_content:
+   segments.append({"start": start, "duration": 0.0, "text": text_content})
+ return segments
+
+
+def main() -> int:
+ print(f"Phase 1 Acquire for {URL}")
+ print(f"Artifacts: {ARTIFACTS}")
+
+ print("Step 1: extract_transcript (try youtube-transcript-api)")
+ transcript_path = ARTIFACTS / "transcript.json"
+ video_id = "9vM4p9NN0Ts"
+ last_exc = None
+ for attempt in range(3):
+  try:
+   segments = _fetch_raw_transcript(video_id)
+   data = {
+    "video_id": video_id,
+    "segments": segments,
+    "plain": "\n".join(s["text"] for s in segments),
+    "fetched_at": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+    "source": "youtube-transcript-api",
+   }
+   transcript_path.write_text(json.dumps(data, indent=2, ensure_ascii=False), encoding="utf-8")
+   print(f"  OK: wrote {transcript_path} ({len(segments)} segments)")
+   break
+  except Exception as e:
+   last_exc = e
+   print(f"  attempt {attempt+1} failed: {type(e).__name__}: {str(e)[:200]}")
+   if attempt < 2:
+    time.sleep(2 ** attempt)
+ else:
+  print(f"  youtube-transcript-api failed after 3 attempts. Falling back to yt-dlp subtitles.")
+  print("Step 1b: yt-dlp subtitle fallback")
+  vtt_path = ARTIFACTS / f"{video_id}.en.vtt"
+  completed = subprocess.run(
+   ["yt-dlp", "--write-auto-subs", "--sub-langs", "en", "--sub-format", "vtt",
+    "--skip-download", "--output", str(ARTIFACTS / video_id), URL],
+   capture_output=True, text=True,
+  )
+  candidates = list(ARTIFACTS.glob(f"{video_id}*.vtt"))
+  if not candidates:
+   print(f"  yt-dlp subtitle fetch also failed: {completed.stderr[:300]}")
+   print(f"  No transcript available. Continuing with download only.")
+   transcript_path.write_text(json.dumps({
+    "video_id": video_id,
+    "segments": [],
+    "plain": "",
+    "fetched_at": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+    "source": "unavailable",
+    "error": str(last_exc)[:500] if last_exc else None,
+    "note": "youtube-transcript-api failed with XML parse error (R5). yt-dlp subtitles also unavailable. Frame OCR will be the primary signal for this video.",
+   }, indent=2, ensure_ascii=False), encoding="utf-8")
+  else:
+   vtt_path = candidates[0]
+   segments = _parse_vtt_segments(vtt_path)
+   data = {
+    "video_id": video_id,
+    "segments": segments,
+    "plain": "\n".join(s["text"] for s in segments),
+    "fetched_at": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
+    "source": "yt-dlp-vtt",
+   }
+   transcript_path.write_text(json.dumps(data, indent=2, ensure_ascii=False), encoding="utf-8")
+   print(f"  OK: parsed {len(segments)} segments from {vtt_path.name}")
+
+ print("Step 2: download_video")
+ video_path = ARTIFACTS / "video.mp4"
+ result = download_video(URL, video_path)
+ if result.is_err():
+  print(f"  ERR: {result.err.class_name}: {result.err.detail[:200]}")
+  return 1
+ print(f"  OK: wrote {video_path} ({video_path.stat().st_size} bytes)")
+ print(f"  log: {result.value['log']}")
+ return 0
+
+
+if __name__ == "__main__":
+ sys.exit(main())

scripts/video_analysis/extract_transcript.py

@@ -70,10 +70,9 @@ def format_transcript_json(video_id: str, segments: list[dict[str, Any]]) -> dic
 
 
 def _fetch_raw_transcript(video_id: str) -> list[dict[str, Any]]:
- api = YouTubeTranscriptApi()
- fetched = api.fetch(video_id)
+ fetched = YouTubeTranscriptApi.get_transcript(video_id)
  return [
-  {"start": float(s.start), "duration": float(s.duration), "text": str(s.text)}
+  {"start": float(s["start"]), "duration": float(s["duration"]), "text": str(s["text"])}
   for s in fetched
  ]