Replace batch DAG system with streaming architecture
- Remove legacy_tasks.py, hybrid_state.py, render.py - Remove old task modules (analyze, execute, execute_sexp, orchestrate) - Add streaming interpreter from test repo - Add sexp_effects with primitives and video effects - Add streaming Celery task with CID-based asset resolution - Support both CID and friendly name references for assets - Add .dockerignore to prevent local clones from conflicting Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
giles
303
tasks/streaming.py
Normal file
303
tasks/streaming.py
Normal file
@@ -0,0 +1,303 @@
|
||||
"""
|
||||
Streaming video rendering task.
|
||||
|
||||
Executes S-expression recipes for frame-by-frame video processing.
|
||||
Supports CID and friendly name references for assets.
|
||||
"""
|
||||
|
||||
import hashlib
|
||||
import logging
|
||||
import os
|
||||
import sys
|
||||
import tempfile
|
||||
from pathlib import Path
|
||||
from typing import Dict, Optional
|
||||
|
||||
from celery import current_task
|
||||
|
||||
# Add parent directory to path for imports
|
||||
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
||||
|
||||
from celery_app import app
|
||||
from cache_manager import get_cache_manager
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
def resolve_asset(ref: str, actor_id: Optional[str] = None) -> Optional[Path]:
|
||||
"""
|
||||
Resolve an asset reference (CID or friendly name) to a file path.
|
||||
|
||||
Args:
|
||||
ref: CID or friendly name (e.g., "my-video" or "QmXyz...")
|
||||
actor_id: User ID for friendly name resolution
|
||||
|
||||
Returns:
|
||||
Path to the asset file, or None if not found
|
||||
"""
|
||||
cache_mgr = get_cache_manager()
|
||||
|
||||
# Try as direct CID first
|
||||
path = cache_mgr.get_by_cid(ref)
|
||||
if path and path.exists():
|
||||
logger.info(f"Resolved {ref[:16]}... as CID to {path}")
|
||||
return path
|
||||
|
||||
# Try as friendly name if actor_id provided
|
||||
if actor_id:
|
||||
import asyncio
|
||||
from database import resolve_friendly_name
|
||||
|
||||
try:
|
||||
loop = asyncio.new_event_loop()
|
||||
asyncio.set_event_loop(loop)
|
||||
try:
|
||||
cid = loop.run_until_complete(resolve_friendly_name(actor_id, ref))
|
||||
finally:
|
||||
loop.close()
|
||||
|
||||
if cid:
|
||||
path = cache_mgr.get_by_cid(cid)
|
||||
if path and path.exists():
|
||||
logger.info(f"Resolved '{ref}' via friendly name to {path}")
|
||||
return path
|
||||
except Exception as e:
|
||||
logger.warning(f"Failed to resolve friendly name '{ref}': {e}")
|
||||
|
||||
logger.warning(f"Could not resolve asset reference: {ref}")
|
||||
return None
|
||||
|
||||
|
||||
class CIDVideoSource:
|
||||
"""
|
||||
Video source that resolves CIDs to file paths.
|
||||
|
||||
Wraps the streaming VideoSource to work with cached assets.
|
||||
"""
|
||||
|
||||
def __init__(self, cid: str, fps: float = 30, actor_id: Optional[str] = None):
|
||||
self.cid = cid
|
||||
self.fps = fps
|
||||
self.actor_id = actor_id
|
||||
self._source = None
|
||||
|
||||
def _ensure_source(self):
|
||||
if self._source is None:
|
||||
path = resolve_asset(self.cid, self.actor_id)
|
||||
if not path:
|
||||
raise ValueError(f"Could not resolve video source: {self.cid}")
|
||||
|
||||
from streaming.stream_sexp_generic import VideoSource
|
||||
# Import from primitives where VideoSource is defined
|
||||
from sexp_effects.primitive_libs.streaming import VideoSource
|
||||
self._source = VideoSource(str(path), self.fps)
|
||||
|
||||
def read_at(self, t: float):
|
||||
self._ensure_source()
|
||||
return self._source.read_at(t)
|
||||
|
||||
def read(self):
|
||||
self._ensure_source()
|
||||
return self._source.read()
|
||||
|
||||
@property
|
||||
def size(self):
|
||||
self._ensure_source()
|
||||
return self._source.size
|
||||
|
||||
@property
|
||||
def duration(self):
|
||||
self._ensure_source()
|
||||
return self._source._duration
|
||||
|
||||
def close(self):
|
||||
if self._source:
|
||||
self._source.close()
|
||||
|
||||
|
||||
class CIDAudioAnalyzer:
|
||||
"""
|
||||
Audio analyzer that resolves CIDs to file paths.
|
||||
"""
|
||||
|
||||
def __init__(self, cid: str, actor_id: Optional[str] = None):
|
||||
self.cid = cid
|
||||
self.actor_id = actor_id
|
||||
self._analyzer = None
|
||||
|
||||
def _ensure_analyzer(self):
|
||||
if self._analyzer is None:
|
||||
path = resolve_asset(self.cid, self.actor_id)
|
||||
if not path:
|
||||
raise ValueError(f"Could not resolve audio source: {self.cid}")
|
||||
|
||||
from sexp_effects.primitive_libs.streaming import AudioAnalyzer
|
||||
self._analyzer = AudioAnalyzer(str(path))
|
||||
|
||||
def get_energy(self, t: float) -> float:
|
||||
self._ensure_analyzer()
|
||||
return self._analyzer.get_energy(t)
|
||||
|
||||
def get_beat(self, t: float) -> bool:
|
||||
self._ensure_analyzer()
|
||||
return self._analyzer.get_beat(t)
|
||||
|
||||
def get_beat_count(self, t: float) -> int:
|
||||
self._ensure_analyzer()
|
||||
return self._analyzer.get_beat_count(t)
|
||||
|
||||
@property
|
||||
def duration(self):
|
||||
self._ensure_analyzer()
|
||||
return self._analyzer.duration
|
||||
|
||||
|
||||
def create_cid_primitives(actor_id: Optional[str] = None):
|
||||
"""
|
||||
Create CID-aware primitive functions.
|
||||
|
||||
Returns dict of primitives that resolve CIDs before creating sources.
|
||||
"""
|
||||
def prim_make_video_source_cid(cid: str, fps: float = 30):
|
||||
return CIDVideoSource(cid, fps, actor_id)
|
||||
|
||||
def prim_make_audio_analyzer_cid(cid: str):
|
||||
return CIDAudioAnalyzer(cid, actor_id)
|
||||
|
||||
return {
|
||||
'streaming:make-video-source': prim_make_video_source_cid,
|
||||
'streaming:make-audio-analyzer': prim_make_audio_analyzer_cid,
|
||||
}
|
||||
|
||||
|
||||
@app.task(bind=True, name='tasks.run_stream')
|
||||
def run_stream(
|
||||
self,
|
||||
recipe_sexp: str,
|
||||
output_name: str = "output.mp4",
|
||||
duration: Optional[float] = None,
|
||||
fps: Optional[float] = None,
|
||||
actor_id: Optional[str] = None,
|
||||
sources_sexp: Optional[str] = None,
|
||||
audio_sexp: Optional[str] = None,
|
||||
) -> dict:
|
||||
"""
|
||||
Execute a streaming S-expression recipe.
|
||||
|
||||
Args:
|
||||
recipe_sexp: The recipe S-expression content
|
||||
output_name: Name for the output file
|
||||
duration: Optional duration override (seconds)
|
||||
fps: Optional FPS override
|
||||
actor_id: User ID for friendly name resolution
|
||||
sources_sexp: Optional sources config S-expression
|
||||
audio_sexp: Optional audio config S-expression
|
||||
|
||||
Returns:
|
||||
Dict with output_cid, output_path, and status
|
||||
"""
|
||||
task_id = self.request.id
|
||||
logger.info(f"Starting stream task {task_id}")
|
||||
|
||||
self.update_state(state='INITIALIZING', meta={'progress': 0})
|
||||
|
||||
# Get the app directory for primitive/effect paths
|
||||
app_dir = Path(__file__).parent.parent # celery/
|
||||
sexp_effects_dir = app_dir / "sexp_effects"
|
||||
effects_dir = app_dir / "effects"
|
||||
templates_dir = app_dir / "templates"
|
||||
|
||||
# Create temp directory for work
|
||||
work_dir = Path(tempfile.mkdtemp(prefix="stream_"))
|
||||
recipe_path = work_dir / "recipe.sexp"
|
||||
output_path = work_dir / output_name
|
||||
|
||||
# Create symlinks to effect directories so relative paths work
|
||||
(work_dir / "sexp_effects").symlink_to(sexp_effects_dir)
|
||||
(work_dir / "effects").symlink_to(effects_dir)
|
||||
(work_dir / "templates").symlink_to(templates_dir)
|
||||
|
||||
try:
|
||||
# Write recipe to temp file
|
||||
recipe_path.write_text(recipe_sexp)
|
||||
|
||||
# Write optional config files
|
||||
sources_path = None
|
||||
if sources_sexp:
|
||||
sources_path = work_dir / "sources.sexp"
|
||||
sources_path.write_text(sources_sexp)
|
||||
|
||||
audio_path = None
|
||||
if audio_sexp:
|
||||
audio_path = work_dir / "audio.sexp"
|
||||
audio_path.write_text(audio_sexp)
|
||||
|
||||
self.update_state(state='RENDERING', meta={'progress': 5})
|
||||
|
||||
# Import the streaming interpreter
|
||||
from streaming.stream_sexp_generic import StreamInterpreter
|
||||
|
||||
# Create interpreter
|
||||
interp = StreamInterpreter(str(recipe_path))
|
||||
|
||||
# Set primitive library directory explicitly
|
||||
interp.primitive_lib_dir = sexp_effects_dir / "primitive_libs"
|
||||
|
||||
if fps:
|
||||
interp.config['fps'] = fps
|
||||
if sources_path:
|
||||
interp.sources_config = sources_path
|
||||
if audio_path:
|
||||
interp.audio_config = audio_path
|
||||
|
||||
# Override primitives with CID-aware versions
|
||||
cid_prims = create_cid_primitives(actor_id)
|
||||
interp.primitives.update(cid_prims)
|
||||
|
||||
# Run rendering to file
|
||||
logger.info(f"Rendering to {output_path}")
|
||||
interp.run(duration=duration, output=str(output_path))
|
||||
|
||||
self.update_state(state='CACHING', meta={'progress': 90})
|
||||
|
||||
# Store output in cache
|
||||
if output_path.exists():
|
||||
cache_mgr = get_cache_manager()
|
||||
cached_file, ipfs_cid = cache_mgr.put(
|
||||
source_path=output_path,
|
||||
node_type="STREAM_OUTPUT",
|
||||
node_id=f"stream_{task_id}",
|
||||
)
|
||||
|
||||
logger.info(f"Stream output cached: CID={cached_file.cid}, IPFS={ipfs_cid}")
|
||||
|
||||
return {
|
||||
"status": "completed",
|
||||
"task_id": task_id,
|
||||
"output_cid": cached_file.cid,
|
||||
"ipfs_cid": ipfs_cid,
|
||||
"output_path": str(cached_file.path),
|
||||
}
|
||||
else:
|
||||
return {
|
||||
"status": "failed",
|
||||
"task_id": task_id,
|
||||
"error": "Output file not created",
|
||||
}
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"Stream task {task_id} failed: {e}")
|
||||
import traceback
|
||||
traceback.print_exc()
|
||||
|
||||
return {
|
||||
"status": "failed",
|
||||
"task_id": task_id,
|
||||
"error": str(e),
|
||||
}
|
||||
|
||||
finally:
|
||||
# Cleanup temp directory
|
||||
import shutil
|
||||
if work_dir.exists():
|
||||
shutil.rmtree(work_dir, ignore_errors=True)
|
||||
Reference in New Issue
Block a user