feat(workers): ingest.video via yt-dlp + Whisper
yt-dlp pulls metadata (title, description, uploader, thumbnail) and bestaudio (opus). faster-whisper transcribes; audio file removed after. Creates a refs row with kind='video' and source_kind='youtube' for YouTube URLs, generic 'video' otherwise. Idempotent on sha256(space_id + url) via refs.external_id. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
59
workers/tests/test_video.py
Normal file
59
workers/tests/test_video.py
Normal file
@@ -0,0 +1,59 @@
|
||||
import subprocess
|
||||
from unittest.mock import patch
|
||||
from void_workers.handlers.video import handle as handle_video
|
||||
|
||||
|
||||
def _reset_void_schema(conn):
|
||||
conn.execute("DROP SCHEMA IF EXISTS public CASCADE")
|
||||
conn.execute("CREATE SCHEMA public")
|
||||
conn.execute("CREATE EXTENSION IF NOT EXISTS pgcrypto")
|
||||
conn.execute("CREATE EXTENSION IF NOT EXISTS vector")
|
||||
|
||||
|
||||
def _run_node_migrations():
|
||||
subprocess.run(
|
||||
["node", "lib/db/migrate.js", "up"],
|
||||
cwd="/project/src/void-v2",
|
||||
check=True
|
||||
)
|
||||
|
||||
|
||||
def test_video_creates_ref_with_transcript_and_metadata(conn):
|
||||
_reset_void_schema(conn)
|
||||
_run_node_migrations()
|
||||
sp = conn.execute(
|
||||
"INSERT INTO spaces(slug, name) VALUES('plan4-vid', 'V') RETURNING id"
|
||||
).fetchone()[0]
|
||||
|
||||
info = {
|
||||
"title": "Sample video",
|
||||
"description": "a description",
|
||||
"duration": 90,
|
||||
"uploader": "Channel",
|
||||
"thumbnail": "https://i.ytimg.com/t.jpg"
|
||||
}
|
||||
with patch("void_workers.handlers.video._yt_dlp_info", return_value=info), \
|
||||
patch("void_workers.handlers.video._yt_dlp_audio", return_value="/tmp/fake.opus"), \
|
||||
patch("void_workers.handlers.video.whisper_transcribe", return_value="hello world transcript"), \
|
||||
patch("os.unlink"):
|
||||
out = handle_video({"space_id": str(sp), "url": "https://youtu.be/abc"})
|
||||
|
||||
assert "ref_id" in out
|
||||
row = conn.execute(
|
||||
"SELECT title, body_text, source_kind FROM refs WHERE id=%s",
|
||||
(out["ref_id"],)
|
||||
).fetchone()
|
||||
assert row[0] == "Sample video"
|
||||
assert "hello world" in row[1]
|
||||
assert row[2] == "youtube"
|
||||
|
||||
|
||||
def test_video_skipped_when_yt_dlp_fails(conn):
|
||||
_reset_void_schema(conn)
|
||||
_run_node_migrations()
|
||||
sp = conn.execute(
|
||||
"INSERT INTO spaces(slug, name) VALUES('plan4-vid2', 'V2') RETURNING id"
|
||||
).fetchone()[0]
|
||||
with patch("void_workers.handlers.video._yt_dlp_info", return_value=None):
|
||||
out = handle_video({"space_id": str(sp), "url": "https://youtu.be/gone"})
|
||||
assert out.get("skipped") == "yt-dlp"
|
||||
Reference in New Issue
Block a user