0Pricing
FastAPI Backend Development Bootcamp · Leçon

Téléversements multiparties et validation du contenu

Acceptez les entrées UploadFile, validez les types MIME et les limites de taille, et protégez-vous contre les charges malveillantes.

Téléversements multiparties et validation du contenu est une leçon FastAPI Backend Development Bootcamp gratuite sur CoddyKit. Ceci est la leçon 1 sur 4. Tu peux lire la leçon complète ci-dessous gratuitement — puis la pratiquer en direct dans le navigateur avec un éditeur de code intégré et un tuteur IA 24/7. Elle fait partie du parcours d'apprentissage FastAPI Backend Development Bootcamp, et ta progression se synchronise sur le web et l'application CoddyKit. Le cours FastAPI Backend Development Bootcamp comprend 4 leçons au total.

Certaines parties de cette leçon n'ont pas encore été traduites et s'affichent en anglais.

Why Multipart Uploads Matter

Regular JSON request bodies cannot carry raw binary files efficiently. To upload an image, PDF, or video, browsers send a multipart/form-data request, which packs each field (text values and file bytes) into separate parts with their own headers.

FastAPI exposes incoming files through two helpers:

  • UploadFile — a spooled file object that keeps small files in memory and large files on disk automatically.
  • File() — a parameter marker that tells FastAPI to read this value from the multipart body.

In this lesson you will accept uploads, validate their MIME type and size, and reject malicious or oversized payloads before they touch your storage.

Your First UploadFile Endpoint

An UploadFile parameter gives you the original filename, the declared content_type, and async methods like read() and seek(). Always declare it with = File(...) so FastAPI parses it from the multipart body.

Note the handler is async because file I/O on UploadFile is awaitable.

from fastapi import FastAPI, UploadFile, File

app = FastAPI()

@app.post("/upload")
async def upload(file: UploadFile = File(...)):
    contents = await file.read()
    return {
        "filename": file.filename,
        "content_type": file.content_type,
        "size_bytes": len(contents),
    }

Never Trust the Declared content_type

The content_type on an UploadFile comes straight from the client. An attacker can label a .exe as image/png. Use it as a cheap first filter, but never as your only check.

A robust pipeline does three things, in order:

  • Reject obviously wrong declared types quickly (cheap).
  • Enforce a hard size limit while streaming (prevents memory exhaustion).
  • Inspect the real file bytes (magic numbers) to confirm the true type.

The next scenes build each layer.

Allowlisting MIME Types

Always use an allowlist, never a blocklist. List exactly the types you support and reject everything else. Return 415 Unsupported Media Type when the declared type is not allowed.

Keep the set small and explicit so new file formats are an intentional decision, not an accident.

from fastapi import FastAPI, UploadFile, File, HTTPException

app = FastAPI()

ALLOWED_TYPES = {"image/jpeg", "image/png", "application/pdf"}

@app.post("/documents")
async def create_document(file: UploadFile = File(...)):
    if file.content_type not in ALLOWED_TYPES:
        raise HTTPException(
            status_code=415,
            detail=f"Unsupported type: {file.content_type}",
        )
    return {"ok": True, "filename": file.filename}

Enforcing a Size Limit by Streaming

Calling await file.read() loads the entire file into memory. A 2 GB upload could crash your worker. Instead, read in fixed-size chunks and abort the moment the running total exceeds your limit.

Return 413 Request Entity Too Large when the cap is breached. This keeps memory bounded no matter how big the client claims the file is.

from fastapi import FastAPI, UploadFile, File, HTTPException

app = FastAPI()

MAX_SIZE = 5 * 1024 * 1024  # 5 MB
CHUNK = 1024 * 1024         # 1 MB

@app.post("/upload")
async def upload(file: UploadFile = File(...)):
    total = 0
    while chunk := await file.read(CHUNK):
        total += len(chunk)
        if total > MAX_SIZE:
            raise HTTPException(413, "File too large")
    return {"filename": file.filename, "size": total}

Streaming Straight to Disk Safely

Once size and type pass, stream the chunks to a destination file instead of holding them in memory. Combine the size check with the write loop so you stop early on oversized payloads and never persist a partial-but-huge file.

Use await file.seek(0) if you read the stream earlier and need to start over.

import aiofiles
from fastapi import UploadFile, File, HTTPException

MAX_SIZE = 5 * 1024 * 1024

async def save_upload(file: UploadFile, dest: str) -> int:
    total = 0
    async with aiofiles.open(dest, "wb") as out:
        while chunk := await file.read(1024 * 1024):
            total += len(chunk)
            if total > MAX_SIZE:
                raise HTTPException(413, "File too large")
            await out.write(chunk)
    return total

Verifying Real Content with Magic Numbers

The most reliable type check inspects the file's leading bytes, its magic number. A PNG always starts with \x89PNG\r\n\x1a\n; a JPEG with \xff\xd8\xff; a PDF with %PDF.

This pure-Python function maps a byte prefix to a real MIME type. You can run it on an online judge with no framework at all.

def sniff_mime(head: bytes) -> str | None:
    signatures = {
        b"\x89PNG\r\n\x1a\n": "image/png",
        b"\xff\xd8\xff": "image/jpeg",
        b"%PDF": "application/pdf",
    }
    for magic, mime in signatures.items():
        if head.startswith(magic):
            return mime
    return None


if __name__ == "__main__":
    print(sniff_mime(b"\x89PNG\r\n\x1a\nrest"))  # image/png
    print(sniff_mime(b"%PDF-1.7"))               # application/pdf
    print(sniff_mime(b"MZ\x90\x00"))             # None (rejected)

Cross-Checking Declared vs Real Type

Combine the layers: read just enough bytes to sniff the magic number, confirm it is in your allowlist, and verify it matches what the client declared. A mismatch (declared image/png but real bytes say PDF) is a strong signal of a malicious or buggy client, so reject it.

After sniffing, call await file.seek(0) so the full file can still be saved.

from fastapi import UploadFile, File, HTTPException

ALLOWED = {"image/png", "image/jpeg", "application/pdf"}

async def validate_type(file: UploadFile) -> str:
    head = await file.read(8)
    await file.seek(0)
    real = sniff_mime(head)
    if real not in ALLOWED:
        raise HTTPException(415, "Content not allowed")
    if file.content_type != real:
        raise HTTPException(415, "Declared type mismatch")
    return real

Sanitizing Filenames

Never use the client-supplied filename directly as a storage path. Names like ../../etc/passwd enable path-traversal, and odd characters break filesystems. Strip the directory part, keep a safe character set, and prefer a generated name plus a validated extension.

This helper is pure Python and judge-runnable.

import re
import uuid
from pathlib import PurePosixPath

EXT_FOR = {"image/png": ".png", "image/jpeg": ".jpg", "application/pdf": ".pdf"}

def safe_name(original: str, mime: str) -> str:
    base = PurePosixPath(original).name          # drop any path parts
    base = re.sub(r"[^A-Za-z0-9._-]", "_", base)  # keep safe chars
    ext = EXT_FOR.get(mime, "")
    return f"{uuid.uuid4().hex}{ext}"


if __name__ == "__main__":
    print(safe_name("../../etc/passwd", "image/png").endswith(".png"))
    print("/" not in safe_name("weird name!.jpg", "image/jpeg"))

Handling Multiple Files at Once

To accept several files in one request, declare the parameter as a list[UploadFile]. The client sends the same form field name repeatedly. Validate each file independently and fail the whole request if any one is invalid, so partial uploads never leave inconsistent state.

from fastapi import FastAPI, UploadFile, File, HTTPException

app = FastAPI()
ALLOWED = {"image/png", "image/jpeg"}

@app.post("/gallery")
async def gallery(files: list[UploadFile] = File(...)):
    if len(files) > 10:
        raise HTTPException(400, "Too many files (max 10)")
    for f in files:
        if f.content_type not in ALLOWED:
            raise HTTPException(415, f"{f.filename}: bad type")
    return {"received": [f.filename for f in files]}

Packaging Validation as a Dependency

Repeating type and size checks in every endpoint is error-prone. Wrap them in a reusable FastAPI dependency. The dependency runs the full pipeline and returns a clean, validated UploadFile, so your route stays focused on business logic.

This is the production-grade shape: allowlist, streamed size cap, magic-number sniff, and filename safety all in one place.

from fastapi import Depends, UploadFile, File, HTTPException

MAX_SIZE = 5 * 1024 * 1024

async def validated_upload(file: UploadFile = File(...)) -> UploadFile:
    head = await file.read(8)
    if sniff_mime(head) not in {"image/png", "image/jpeg", "application/pdf"}:
        raise HTTPException(415, "Unsupported content")
    total = len(head)
    while chunk := await file.read(1024 * 1024):
        total += len(chunk)
        if total > MAX_SIZE:
            raise HTTPException(413, "File too large")
    await file.seek(0)
    return file

@app.post("/secure-upload")
async def secure_upload(file: UploadFile = Depends(validated_upload)):
    return {"filename": file.filename}

Quick Check: Choosing the Right Guard

An endpoint accepts profile pictures. A user uploads a 3 GB file whose content_type header claims image/png, but the bytes are actually an executable. Which single combination of checks reliably protects the server?

Recap: A Layered Upload Defense

You built a complete, defensive upload pipeline for FastAPI:

  • UploadFile + File() accept multipart data with streaming-friendly I/O.
  • Allowlist the declared MIME type and return 415 for anything unexpected, but never trust that header alone.
  • Stream in chunks and abort with 413 once a size cap is exceeded, keeping memory bounded.
  • Sniff magic numbers to confirm the real content type and reject declared-vs-real mismatches.
  • Sanitize filenames with generated names to stop path traversal.
  • Package it as a dependency so every endpoint reuses the same guard.

Layered checks, ordered cheap-to-expensive, give you robust protection against oversized and malicious uploads.

Questions Fréquemment Posées

La leçon « Téléversements multiparties et validation du contenu » est-elle gratuite ?

Oui — le texte complet de « Téléversements multiparties et validation du contenu » est gratuit à lire ici sur le web. Pour la pratiquer de manière interactive (un éditeur de code intégré et un tuteur IA 24/7) et déverrouiller le reste du cours FastAPI Backend Development Bootcamp, passe à CoddyKit PRO. Le cours FastAPI Backend Development Bootcamp comprend 4 leçons au total.

Qu'est-ce que j'apprendrai dans « Téléversements multiparties et validation du contenu » ?

Acceptez les entrées UploadFile, validez les types MIME et les limites de taille, et protégez-vous contre les charges malveillantes. Tu pratiques FastAPI Backend Development Bootcamp avec du code pratique que tu exécutes directement dans le navigateur, et un tuteur IA 24/7 répond à tes questions au fur et à mesure que tu avances dans la leçon.

Dois-je avoir de l'expérience pour commencer FastAPI Backend Development Bootcamp ?

Aucune expérience préalable n'est requise. FastAPI Backend Development Bootcamp sur CoddyKit est structuré pour les débutants jusqu'aux apprenants avancés, donc tu peux commencer ici ou depuis le début et avancer à ton rythme. Ceci est la leçon 1 sur 4.

Combien de temps prend la leçon « Téléversements multiparties et validation du contenu » ?

La plupart des leçons CoddyKit prennent environ 5–10 minutes. Chacune est courte et interactive, tu progresses régulièrement et tu repiques exactement où tu t'es arrêté sur le web et l'app.

Peux-tu écrire et exécuter du code dans cette leçon FastAPI Backend Development Bootcamp ?

Oui. Chaque leçon FastAPI Backend Development Bootcamp inclut un éditeur de code intégré, tu écris et exécutes du vrai code directement dans ton navigateur et tu reçois des retours IA instantanés — aucune configuration locale requise.

Toutes les leçons de ce cours

  1. Téléversements multiparties et validation du contenu
  2. Réponses en flux continu et requêtes par plages
  3. Délégation du stockage vers des compartiments compatibles S3
  4. Transformation asynchrone d’images et de documents
← Retour à FastAPI Backend Development Bootcamp