5e2094f2ad
FastAPI service in infra/media-worker implementing every MediaInferenceProvider operation behind HTTP API v1, plus capabilities. Bearer auth, MEDIA_ROOT realpath enforcement, strict request schemas, cancellation on disconnect, GPU/CPU gates with 503 overflow, input size/duration limits, lazy SHA256-pinned model downloads, fail-closed redaction with atomic output, offline pytest suite with fake backends, CPU/CUDA/ROCm Dockerfile and a license review of every model. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0138YqDj3ttfayRwuMKKrqVj
63 lines
1.8 KiB
Python
63 lines
1.8 KiB
Python
"""Helpers shared by the operation modules."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import threading
|
|
from dataclasses import dataclass
|
|
from typing import Any
|
|
|
|
import numpy as np
|
|
|
|
from .. import media_io, paths
|
|
from ..config import Settings
|
|
from ..errors import BadRequest
|
|
from ..media_io import VideoInfo
|
|
from ..models import Registry
|
|
|
|
|
|
@dataclass
|
|
class Ctx:
|
|
settings: Settings
|
|
registry: Registry
|
|
|
|
|
|
def open_input(
|
|
ctx: Ctx, path: str, cancel: threading.Event, need_video: bool = True
|
|
) -> tuple[str, VideoInfo]:
|
|
"""Resolve, size-check, probe, and duration-check an input file."""
|
|
real = paths.input_file(ctx.settings.media_root, path, ctx.settings.max_input_bytes)
|
|
info = media_io.probe(real, cancel)
|
|
media_io.check_duration(info, ctx.settings.max_seconds)
|
|
if need_video and not info.has_video:
|
|
raise BadRequest("input has no video stream")
|
|
return real, info
|
|
|
|
|
|
def clamp01(value: float) -> float:
|
|
return float(min(max(value, 0.0), 1.0))
|
|
|
|
|
|
def norm_box(x: float, y: float, w: float, h: float, width: int, height: int) -> dict[str, float]:
|
|
"""Pixel box -> normalized box clipped to the frame."""
|
|
x0, y0 = clamp01(x / width), clamp01(y / height)
|
|
x1, y1 = clamp01((x + w) / width), clamp01((y + h) / height)
|
|
return {"x": x0, "y": y0, "w": max(x1 - x0, 0.0), "h": max(y1 - y0, 0.0)}
|
|
|
|
|
|
def norm_point(x: float, y: float, width: int, height: int) -> dict[str, float]:
|
|
return {"x": clamp01(x / width), "y": clamp01(y / height)}
|
|
|
|
|
|
def unique_sorted(times: list[int]) -> list[int]:
|
|
return sorted(set(int(t) for t in times))
|
|
|
|
|
|
def jsonable(value: Any) -> Any:
|
|
if isinstance(value, dict):
|
|
return {k: jsonable(v) for k, v in value.items()}
|
|
if isinstance(value, (list, tuple)):
|
|
return [jsonable(v) for v in value]
|
|
if isinstance(value, np.generic):
|
|
return value.item()
|
|
return value
|