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
76 lines
2.6 KiB
Python
76 lines
2.6 KiB
Python
"""MEDIA_ROOT enforcement. Every path a request names goes through here.
|
|
|
|
Inputs and outputs must be absolute and stay inside the real MEDIA_ROOT after
|
|
symlinks are resolved. A path that does not exist yet is checked through its
|
|
nearest existing ancestor (realpath resolves the existing prefix).
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import os
|
|
import stat
|
|
|
|
from .errors import InvalidPath, MediaError
|
|
|
|
|
|
def _inside(root: str, candidate: str) -> bool:
|
|
return candidate == root or candidate.startswith(root.rstrip(os.sep) + os.sep)
|
|
|
|
|
|
def resolve_inside(root: str, requested: str) -> str:
|
|
"""Return the real absolute location of `requested`, or raise InvalidPath."""
|
|
if not isinstance(requested, str) or not requested or "\0" in requested:
|
|
raise InvalidPath()
|
|
if not os.path.isabs(requested):
|
|
raise InvalidPath("path must be absolute")
|
|
real_root = os.path.realpath(root)
|
|
lexical = os.path.normpath(requested)
|
|
# Reject lexical traversal even when it would land back inside the root.
|
|
if ".." in requested.split("/"):
|
|
raise InvalidPath()
|
|
real = os.path.realpath(lexical)
|
|
if not _inside(real_root, real):
|
|
raise InvalidPath()
|
|
return real
|
|
|
|
|
|
def input_file(root: str, requested: str, max_bytes: int | None = None) -> str:
|
|
real = resolve_inside(root, requested)
|
|
try:
|
|
info = os.stat(real)
|
|
except FileNotFoundError as error:
|
|
raise InvalidPath("input file does not exist") from error
|
|
if not stat.S_ISREG(info.st_mode):
|
|
raise InvalidPath("input is not a regular file")
|
|
if max_bytes is not None and info.st_size > max_bytes:
|
|
raise MediaError("too_large", "input exceeds MEDIA_MAX_INPUT_BYTES")
|
|
return real
|
|
|
|
|
|
def input_dir(root: str, requested: str) -> str:
|
|
real = resolve_inside(root, requested)
|
|
if not os.path.isdir(real):
|
|
raise InvalidPath("directory does not exist")
|
|
return real
|
|
|
|
|
|
def output_file(root: str, requested: str) -> str:
|
|
"""Resolve an output file path and create its parent directories inside the root."""
|
|
real = resolve_inside(root, requested)
|
|
if os.path.isdir(real):
|
|
raise InvalidPath("output path is a directory")
|
|
parent = os.path.dirname(real)
|
|
os.makedirs(parent, exist_ok=True)
|
|
# The parent may have been swapped for a symlink in between; check again.
|
|
resolve_inside(root, parent)
|
|
return real
|
|
|
|
|
|
def output_dir(root: str, requested: str) -> str:
|
|
real = resolve_inside(root, requested)
|
|
os.makedirs(real, exist_ok=True)
|
|
resolve_inside(root, real)
|
|
if not os.path.isdir(real):
|
|
raise InvalidPath("output directory is not a directory")
|
|
return real
|