Spaces:
Running
Running
File size: 22,932 Bytes
fba6023 03c8ca2 fba6023 e5a2fe9 fba6023 03c8ca2 fba6023 03c8ca2 fba6023 8155940 fba6023 03c8ca2 fba6023 03c8ca2 f764c05 03c8ca2 fba6023 8155940 19fccd4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 | from __future__ import annotations
import asyncio
import base64
import json
import mimetypes
from pathlib import Path
from typing import Any
from uuid import uuid4
import aiofiles
from fastapi import Request
from starlette.datastructures import FormData, UploadFile
from app.core.config import Settings
from app.core.exceptions import InputError
from app.models.media import InputMedia, MediaSource, OperationResult, ResolvedRequest
from app.services.cleanup import CleanupService, RequestWorkspace
from app.services.downloader import Downloader
from app.services.validator import MediaValidator
from app.services.ytdlp_service import YTDLPService
from app.security.assets import CanonicalAssetNotFoundError, CanonicalAssetService
from app.security.context import auth_context
class InputResolver:
"""Normalizes multipart, JSON, Base64, n8n, URL, and raw-byte inputs."""
INPUT_KEYS = {
"url",
"urls",
"base64",
"binary",
"input",
"inputs",
"files",
"file",
"path",
"temp_path",
"asset_id",
}
def __init__(
self,
settings: Settings,
cleanup: CleanupService,
downloader: Downloader,
ytdlp: YTDLPService,
validator: MediaValidator,
assets: CanonicalAssetService | None = None,
) -> None:
self.settings = settings
self.cleanup = cleanup
self.downloader = downloader
self.ytdlp = ytdlp
self.validator = validator
self.assets = assets
async def resolve(self, request: Request, *, require_input: bool = True) -> ResolvedRequest:
request_id = request.state.request_id
workspace = await self.cleanup.create_workspace(request_id)
content_type = request.headers.get("content-type", "").lower()
is_ytdlp_request = request.url.path.startswith("/v1/ytdlp/")
try:
if content_type.startswith("multipart/form-data"):
inputs, params = await self._multipart(request, workspace, is_ytdlp_request)
elif "json" in content_type:
inputs, params = await self._json(request, workspace, is_ytdlp_request)
else:
inputs, params = await self._raw(request, workspace, content_type)
except Exception:
await self.cleanup.complete(request_id)
raise
if require_input and not inputs:
await self.cleanup.complete(request_id)
raise InputError(
"No media input found. Send multipart file(s), raw bytes, url, base64, or n8n binary data."
)
params = {**dict(request.query_params), **params}
return ResolvedRequest(request_id=request_id, inputs=inputs, params=params)
async def resolve_payload(
self,
payload: dict[str, Any],
request_id: str,
*,
require_input: bool = True,
ytdlp_options: dict[str, Any] | None = None,
) -> ResolvedRequest:
"""Resolve an in-process JSON payload through the same REST input pipeline.
MCP and other non-HTTP transports use this method so URL, Base64, n8n
binary, managed temporary paths, validation, and workspace handling stay
centralized in one resolver.
"""
workspace = await self.cleanup.create_workspace(request_id)
try:
descriptors = self._collect_descriptors(payload)
params = self._params_from_payload(payload)
inputs = [
await self._resolve_descriptor(
descriptor,
workspace,
ytdlp_options=ytdlp_options,
)
for descriptor in descriptors
]
except Exception:
await self.cleanup.complete(request_id)
raise
if require_input and not inputs:
await self.cleanup.complete(request_id)
raise InputError(
"No media input found. Provide url, base64, binary, temp_path, or inputs."
)
return ResolvedRequest(request_id=request_id, inputs=inputs, params=params)
async def _multipart(
self,
request: Request,
workspace: RequestWorkspace,
is_ytdlp_request: bool,
) -> tuple[list[InputMedia], dict[str, Any]]:
try:
form: FormData = await request.form(
max_files=100,
max_fields=500,
max_part_size=self.settings.max_upload_size,
)
except TypeError:
form = await request.form()
except Exception as exc:
raise InputError("Invalid multipart form data") from exc
inputs: list[InputMedia] = []
fields: dict[str, Any] = {}
for key, value in form.multi_items():
if isinstance(value, UploadFile):
inputs.append(await self._save_upload(value, workspace))
continue
if key in fields:
current = fields[key]
fields[key] = current + [value] if isinstance(current, list) else [current, value]
else:
fields[key] = value
descriptors: list[dict[str, Any]] = []
for key, value in fields.items():
if key == "url":
descriptors.extend({"url": item} for item in self._as_list(value))
elif key == "urls":
parsed = self._parse_field(value)
descriptors.extend({"url": item} for item in self._as_list(parsed))
elif key == "base64" or key.startswith("binary."):
for item in self._as_list(value):
parsed = self._parse_field(item)
descriptors.append(parsed if isinstance(parsed, dict) else {"base64": parsed})
elif key in {"input", "inputs", "files", "binary"}:
parsed = self._parse_field(value)
descriptors.extend(self._collect_descriptors(parsed, binary=key == "binary"))
params = self._params_from_payload(fields)
ytdlp_options = params if is_ytdlp_request else None
for descriptor in descriptors:
inputs.append(
await self._resolve_descriptor(descriptor, workspace, ytdlp_options=ytdlp_options)
)
return inputs, params
async def _json(
self,
request: Request,
workspace: RequestWorkspace,
is_ytdlp_request: bool,
) -> tuple[list[InputMedia], dict[str, Any]]:
content_length = self._int(request.headers.get("content-length"))
maximum_json = int(self.settings.max_upload_size * 1.5) + 1_048_576
if content_length and content_length > maximum_json:
raise InputError("JSON request body is too large")
body = await request.body()
if len(body) > maximum_json:
raise InputError("JSON request body is too large")
try:
payload = await asyncio.to_thread(json.loads, body)
except (json.JSONDecodeError, UnicodeDecodeError) as exc:
raise InputError("Request body is not valid JSON") from exc
if isinstance(payload, str):
stripped = payload.strip()
payload = {"url": stripped} if stripped.startswith(("http://", "https://")) else {"base64": stripped}
elif isinstance(payload, list):
payload = {"inputs": payload}
elif not isinstance(payload, dict):
raise InputError("JSON request body must be an object, array, URL string, or Base64 string")
descriptors = self._collect_descriptors(payload)
params = self._params_from_payload(payload)
ytdlp_options = params if is_ytdlp_request else None
inputs = [
await self._resolve_descriptor(descriptor, workspace, ytdlp_options=ytdlp_options)
for descriptor in descriptors
]
return inputs, params
async def _raw(
self, request: Request, workspace: RequestWorkspace, content_type: str
) -> tuple[list[InputMedia], dict[str, Any]]:
declared = self._int(request.headers.get("content-length"))
if declared and declared > self.settings.max_upload_size:
raise InputError("The request body exceeds MAX_UPLOAD_SIZE")
filename = self.validator.safe_filename(
request.headers.get("x-filename") or request.query_params.get("filename"),
"input.bin",
)
path = workspace.uploads / f"{uuid4().hex}_{filename}"
size = 0
async with aiofiles.open(path, "wb") as output:
async for chunk in request.stream():
size += len(chunk)
if size > self.settings.max_upload_size:
await output.close()
path.unlink(missing_ok=True)
raise InputError("The request body exceeds MAX_UPLOAD_SIZE")
await output.write(chunk)
if size == 0:
path.unlink(missing_ok=True)
return [], dict(request.query_params)
mime_type = content_type.split(";", 1)[0] or self.validator.infer_mime(path)
self.validator.validate_declared(filename, mime_type, size)
return [
InputMedia(
source=MediaSource.OCTET_STREAM,
filename=filename,
mime_type=mime_type,
temp_path=path,
size=size,
)
], dict(request.query_params)
async def _save_upload(self, upload: UploadFile, workspace: RequestWorkspace) -> InputMedia:
filename = self.validator.safe_filename(upload.filename, "upload.bin")
path = workspace.uploads / f"{uuid4().hex}_{filename}"
size = 0
async with aiofiles.open(path, "wb") as output:
while chunk := await upload.read(1024 * 1024):
size += len(chunk)
if size > self.settings.max_upload_size:
await output.close()
path.unlink(missing_ok=True)
raise InputError("The uploaded media exceeds MAX_UPLOAD_SIZE")
await output.write(chunk)
await upload.close()
mime_type = upload.content_type or self.validator.infer_mime(path)
self.validator.validate_declared(filename, mime_type, size)
return InputMedia(
source=MediaSource.MULTIPART,
filename=filename,
mime_type=mime_type,
temp_path=path,
size=size,
)
async def _resolve_descriptor(
self,
descriptor: dict[str, Any],
workspace: RequestWorkspace,
*,
ytdlp_options: dict[str, Any] | None = None,
) -> InputMedia:
if "asset_id" in descriptor:
return await self._resolve_asset(descriptor, workspace)
if "temp_path" in descriptor or "path" in descriptor:
return await self._resolve_local_path(descriptor, workspace)
if "url" in descriptor:
url = descriptor["url"]
if not isinstance(url, str) or not url.strip():
raise InputError("Media URL must be a non-empty string")
await self.downloader.validate_url(url)
if ytdlp_options is not None or await asyncio.to_thread(self.ytdlp.supports_url, url):
options = ytdlp_options or {}
mode = str(options.get("mode", "video"))
if options.get("playlist") or options.get("include_formats"):
max_entries = self._int(options.get("max_entries", 100))
if max_entries is None:
raise InputError("max_entries must be an integer")
metadata = await self.ytdlp.extract_metadata(
url,
playlist=bool(options.get("playlist")),
include_formats=bool(options.get("include_formats")),
max_entries=max_entries,
)
result = OperationResult(metadata=metadata)
mode = "metadata"
else:
result = await self.ytdlp.download(
url,
workspace.uploads,
mode=mode,
format_selector=options.get("format"),
audio_format=str(options.get("audio_format", "mp3")),
)
if result.path is None:
if mode != "metadata":
raise InputError("yt-dlp did not return a media file")
result.path = workspace.uploads / "metadata.json"
encoded = json.dumps(
result.metadata, ensure_ascii=False, separators=(",", ":")
).encode("utf-8")
async with aiofiles.open(result.path, "wb") as output:
await output.write(encoded)
result.filename = result.path.name
result.mime_type = "application/json"
return InputMedia(
source=MediaSource.YTDLP,
filename=result.filename or result.path.name,
mime_type=result.mime_type or self.validator.infer_mime(result.path),
temp_path=result.path,
size=result.path.stat().st_size,
duration=result.metadata.get("duration"),
metadata=result.metadata,
)
path, mime_type = await self.downloader.download(
url, workspace.uploads, descriptor.get("filename")
)
return InputMedia(
source=MediaSource.JSON_URL,
filename=self._original_name(path),
mime_type=mime_type,
temp_path=path,
size=path.stat().st_size,
metadata={"url": url},
)
data = descriptor.get("base64", descriptor.get("data"))
if not isinstance(data, str) or not data:
raise InputError("Input descriptor must contain a url or Base64 data")
return await self._decode_base64(data, descriptor, workspace)
async def _resolve_asset(
self, descriptor: dict[str, Any], workspace: RequestWorkspace
) -> InputMedia:
asset_id = descriptor.get("asset_id")
if not isinstance(asset_id, str) or not asset_id.strip():
raise InputError("asset_id must be a non-empty string")
context = auth_context.get()
if self.assets is None or context is None or not context.workspace_id or not context.user_id:
raise InputError("asset_id requires an authenticated workspace")
try:
record = await self.assets.get_owned_by_id(
workspace_id=context.workspace_id,
user_id=context.user_id,
asset_id=asset_id,
)
source = self.cleanup.resolve_download(record.request_id, record.filename)
await self.assets.verify_file(record, source)
except (CanonicalAssetNotFoundError, FileNotFoundError, ValueError) as exc:
raise InputError("Media asset was not found or is no longer available") from exc
destination = workspace.uploads / f"{uuid4().hex}_{record.filename}"
# Stream canonical media in bounded chunks. Asset resolution must never
# load an entire video/audio file into the worker process memory.
copied = 0
try:
async with (
aiofiles.open(source, "rb") as input_stream,
aiofiles.open(destination, "wb") as output_stream,
):
while chunk := await input_stream.read(1024 * 1024):
copied += len(chunk)
if copied > self.settings.max_upload_size:
raise InputError("Media asset exceeds MAX_UPLOAD_SIZE")
await output_stream.write(chunk)
except Exception:
destination.unlink(missing_ok=True)
raise
if copied != record.file_size:
destination.unlink(missing_ok=True)
raise InputError("Media asset size changed while being resolved")
return InputMedia(
source=MediaSource.ASSET,
filename=record.filename,
mime_type=record.mime_type,
temp_path=destination,
size=record.file_size,
metadata={"asset_id": record.id, "sha256": record.sha256},
)
async def _resolve_local_path(
self, descriptor: dict[str, Any], workspace: RequestWorkspace
) -> InputMedia:
raw_path = descriptor.get("temp_path", descriptor.get("path"))
if not isinstance(raw_path, str) or not raw_path.strip():
raise InputError("temp_path must be a non-empty string")
candidate = Path(raw_path).expanduser().resolve()
allowed_roots = (self.settings.temp_dir.resolve(), self.settings.output_dir.resolve())
if not any(candidate == root or root in candidate.parents for root in allowed_roots):
raise InputError("temp_path must be inside TEMP_DIR or OUTPUT_DIR")
if not candidate.is_file():
raise InputError("The managed temporary file does not exist")
filename = self.validator.safe_filename(descriptor.get("filename"), candidate.name)
mime_type = str(
descriptor.get("mime_type")
or descriptor.get("mimeType")
or self.validator.infer_mime(candidate)
)
size = candidate.stat().st_size
self.validator.validate_declared(filename, mime_type, size)
destination = workspace.uploads / f"{uuid4().hex}_{filename}"
async with (
aiofiles.open(candidate, "rb") as source,
aiofiles.open(destination, "wb") as output,
):
while chunk := await source.read(1024 * 1024):
await output.write(chunk)
return InputMedia(
source=MediaSource.LOCAL_PATH,
filename=filename,
mime_type=mime_type,
temp_path=destination,
size=size,
metadata={"managed_source": str(candidate)},
)
async def _decode_base64(
self, value: str, descriptor: dict[str, Any], workspace: RequestWorkspace
) -> InputMedia:
mime_type = str(
descriptor.get("mime_type") or descriptor.get("mimeType") or "application/octet-stream"
)
if value.startswith("data:"):
try:
header, value = value.split(",", 1)
except ValueError as exc:
raise InputError("Invalid Base64 data URI") from exc
if ";base64" not in header:
raise InputError("Only Base64 data URIs are supported")
mime_type = header[5:].split(";", 1)[0] or mime_type
compact = "".join(value.split())
estimated_size = len(compact) * 3 // 4
if estimated_size > self.settings.max_upload_size:
raise InputError("Decoded Base64 media exceeds MAX_UPLOAD_SIZE")
try:
decoded = await asyncio.to_thread(base64.b64decode, compact, validate=True)
except ValueError as exc:
raise InputError("Media contains invalid Base64 data") from exc
extension = mimetypes.guess_extension(mime_type) or ".bin"
supplied_name = descriptor.get("filename") or descriptor.get("fileName")
filename = self.validator.safe_filename(supplied_name, f"decoded{extension}")
self.validator.validate_declared(filename, mime_type, len(decoded))
path = workspace.uploads / f"{uuid4().hex}_{filename}"
async with aiofiles.open(path, "wb") as output:
await output.write(decoded)
source = MediaSource.N8N_BINARY if "data" in descriptor else MediaSource.JSON_BASE64
return InputMedia(
source=source,
filename=filename,
mime_type=mime_type,
temp_path=path,
size=len(decoded),
)
def _collect_descriptors(self, value: Any, *, binary: bool = False) -> list[dict[str, Any]]:
if isinstance(value, list):
result: list[dict[str, Any]] = []
for item in value:
result.extend(self._collect_descriptors(item, binary=binary))
return result
if not isinstance(value, dict):
if binary and isinstance(value, str):
return [{"data": value}]
if isinstance(value, str):
stripped = value.strip()
if stripped.startswith(("http://", "https://")):
return [{"url": stripped}]
return []
if any(key in value for key in ("url", "base64", "temp_path", "path", "asset_id")):
return [value]
if "data" in value and (
binary or any(key in value for key in ("fileName", "filename", "mimeType", "mime_type"))
):
return [value]
result: list[dict[str, Any]] = []
for key, item in value.items():
if key in self.INPUT_KEYS:
result.extend(self._collect_descriptors(item, binary=(key == "binary")))
elif binary and isinstance(item, (dict, list)):
result.extend(self._collect_descriptors(item, binary=True))
return result
def _params_from_payload(self, payload: dict[str, Any]) -> dict[str, Any]:
"""Keep operation parameters while excluding fields consumed as media inputs."""
if not isinstance(payload, dict):
return {}
return {key: value for key, value in payload.items() if key not in self.INPUT_KEYS}
@staticmethod
def _as_list(value: Any) -> list[Any]:
return value if isinstance(value, list) else [value]
@staticmethod
def _parse_field(value: Any) -> Any:
if not isinstance(value, str):
return value
stripped = value.strip()
if not stripped:
return stripped
try:
return json.loads(stripped)
except json.JSONDecodeError:
return value
@staticmethod
def _int(value: Any) -> int | None:
if value is None or isinstance(value, bool):
return None
try:
return int(value)
except (TypeError, ValueError, OverflowError):
return None
@staticmethod
def _original_name(path: Path) -> str:
name = path.name
# Downloader prefixes names with a random identifier for collision safety.
if "_" in name:
prefix, remainder = name.split("_", 1)
if len(prefix) == 32 and all(ch in "0123456789abcdef" for ch in prefix.lower()):
return remainder
return name
|