Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
File size: 13,789 Bytes
d88a570 970eef1 d88a570 c750639 b1e6db8 970eef1 c750639 970eef1 c750639 970eef1 c750639 81e0b0c b1e6db8 d88a570 c750639 81e0b0c c750639 81e0b0c d88a570 c750639 81e0b0c d88a570 81e0b0c c750639 81e0b0c d88a570 81e0b0c d88a570 ffa4ae8 970eef1 c750639 970eef1 d88a570 970eef1 2a342ed 970eef1 2a342ed 970eef1 2a342ed 970eef1 c750639 970eef1 c750639 d88a570 81e0b0c d88a570 81e0b0c d88a570 81e0b0c d88a570 81e0b0c d88a570 81e0b0c d88a570 c750639 d88a570 81e0b0c d88a570 970eef1 c750639 b1e6db8 c750639 b1e6db8 c750639 b1e6db8 c750639 b1e6db8 c750639 |
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 |
from fastapi import APIRouter, UploadFile, File, HTTPException
import os
import shutil
import uuid
from bs4 import BeautifulSoup
from PyPDF2 import PdfReader
import requests
from fastapi import Form
from typing import Optional, List
import re
from urllib.parse import urlparse
import html
import validators
router = APIRouter(tags=["files"])
# Define file storage by session (imported in main.py)
session_files = {}
# Root folder for uploads
UPLOAD_ROOT = "uploaded_files"
os.makedirs(UPLOAD_ROOT, exist_ok=True)
# Minimum length for any file (in characters)
MIN_FILE_LENGTH = 500
# Configuration des limites de sécurité
MAX_CONTENT_SIZE = 5 * 1024 * 1024 # 5 MB max pour le contenu téléchargé
REQUEST_TIMEOUT = 10 # Timeout pour les requêtes HTTP
# Liste des domaines autorisés (vide = tous autorisés, mais à remplir en production)
ALLOWED_DOMAINS: List[str] = []
# Liste d'extensions de fichiers à bloquer dans les URLs
BLOCKED_EXTENSIONS = ['.exe', '.sh', '.bat', '.dll', '.jar', '.msi']
def validate_pdf(file_path: str) -> bool:
"""Validate if file is a valid PDF."""
try:
reader = PdfReader(file_path)
# Check that the PDF has at least one page
if len(reader.pages) == 0:
return False
# Extract text to check length
text = ""
for page in reader.pages:
text += page.extract_text()
return len(text) >= MIN_FILE_LENGTH
except:
return False
def validate_markdown(file_path: str) -> bool:
"""Validate if file is a valid Markdown file."""
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
# Check minimum length and presence of markdown elements
return len(content) >= MIN_FILE_LENGTH and any(marker in content for marker in ['#', '-', '*', '`', '[', '>'])
except:
return False
def validate_html(file_path: str) -> bool:
"""Validate if file is a valid HTML file."""
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
# Check minimum length and HTML structure
if len(content) < MIN_FILE_LENGTH:
return False
BeautifulSoup(content, 'html.parser')
return True
except:
return False
def validate_txt(file_path: str) -> bool:
"""Validate if file is a valid text file."""
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
return len(content.strip()) >= MIN_FILE_LENGTH
except:
return False
# Initialize session files dictionary with pre-calculated documents
precalculated_docs = ["the-bitter-lesson", "hurricane-faq", "pokemon-guide"]
for doc_id in precalculated_docs:
doc_dir = os.path.join(UPLOAD_ROOT, doc_id)
if os.path.exists(doc_dir):
doc_files_dir = os.path.join(doc_dir, "uploaded_files")
if os.path.exists(doc_files_dir):
for filename in os.listdir(doc_files_dir):
if filename.endswith((".pdf", ".txt", ".html", ".md")):
file_path = os.path.join(doc_files_dir, filename)
session_files[doc_id] = file_path
print(f"Added pre-calculated document to session_files: {doc_id} -> {file_path}")
break
else:
# Search directly in the doc_dir
for filename in os.listdir(doc_dir):
if filename.endswith((".pdf", ".txt", ".html", ".md")):
file_path = os.path.join(doc_dir, filename)
session_files[doc_id] = file_path
print(f"Added pre-calculated document to session_files: {doc_id} -> {file_path}")
break
@router.post("/upload")
async def upload_file(file: UploadFile = File(...)):
"""
Upload a file to the server and generate a session ID
Args:
file: The file to upload
Returns:
Dictionary with filename, status and session_id
"""
# Check if the file is a PDF, TXT, HTML or MD
if not file.filename.endswith(('.pdf', '.txt', '.html', '.md')):
raise HTTPException(status_code=400, detail="Only PDF, TXT, HTML and MD files are accepted")
# Get the file extension
file_extension = os.path.splitext(file.filename)[1].lower()
# Generate a session ID for this file
session_id = str(uuid.uuid4())
# Create the session directory structure
session_dir = os.path.join(UPLOAD_ROOT, session_id)
uploaded_files_dir = os.path.join(session_dir, "uploaded_files")
os.makedirs(uploaded_files_dir, exist_ok=True)
# Create standardized filename
standardized_filename = f"document{file_extension}"
# Create the full path to save the file
file_path = os.path.join(uploaded_files_dir, standardized_filename)
# Save the file
with open(file_path, "wb") as buffer:
shutil.copyfileobj(file.file, buffer)
# Validate the file according to its type
is_valid = False
error_detail = ""
if file_extension == '.pdf':
try:
reader = PdfReader(file_path)
if len(reader.pages) == 0:
error_detail = "PDF must contain at least one page"
is_valid = False
else:
text = ""
for page in reader.pages:
text += page.extract_text()
if len(text) < MIN_FILE_LENGTH:
error_detail = f"PDF contains {len(text)} characters but must contain at least {MIN_FILE_LENGTH}"
is_valid = False
else:
is_valid = True
except:
error_detail = "Invalid PDF format"
is_valid = False
elif file_extension == '.md':
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
if len(content) < MIN_FILE_LENGTH:
error_detail = f"Markdown file contains {len(content)} characters but must contain at least {MIN_FILE_LENGTH}"
is_valid = False
elif not any(marker in content for marker in ['#', '-', '*', '`', '[', '>']):
error_detail = "Markdown file does not contain any valid Markdown elements"
is_valid = False
else:
is_valid = True
except:
error_detail = "Invalid Markdown format"
is_valid = False
elif file_extension == '.html':
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
if len(content) < MIN_FILE_LENGTH:
error_detail = f"HTML file contains {len(content)} characters but must contain at least {MIN_FILE_LENGTH}"
is_valid = False
else:
BeautifulSoup(content, 'html.parser')
is_valid = True
except:
error_detail = "Invalid HTML format"
is_valid = False
elif file_extension == '.txt':
try:
with open(file_path, 'r', encoding='utf-8') as f:
content = f.read()
content_length = len(content.strip())
if content_length < MIN_FILE_LENGTH:
error_detail = f"Text file contains {content_length} characters but must contain at least {MIN_FILE_LENGTH}"
is_valid = False
else:
is_valid = True
except:
error_detail = "Invalid text format"
is_valid = False
if not is_valid:
# Delete the invalid file
os.remove(file_path)
raise HTTPException(status_code=400, detail=error_detail or f"Invalid {file_extension[1:].upper()} file")
# Store file path for later use
session_files[session_id] = file_path
return {"filename": standardized_filename, "status": "uploaded", "session_id": session_id}
@router.post("/upload-url")
async def upload_url(url: str = Form(...)):
"""
Upload content from a URL, extract text and store it as a document
Args:
url: The URL to download content from
Returns:
Dictionary with status and session_id
"""
try:
# Valider que l'URL est bien formée
if not validators.url(url):
raise HTTPException(status_code=400, detail="Invalid URL format")
# Vérifier si l'URL a une extension bloquée
parsed_url = urlparse(url)
path = parsed_url.path.lower()
if any(path.endswith(ext) for ext in BLOCKED_EXTENSIONS):
raise HTTPException(status_code=400, detail="This file type is not allowed")
# Vérifier si le domaine est autorisé (si la liste n'est pas vide)
domain = parsed_url.netloc
if ALLOWED_DOMAINS and domain not in ALLOWED_DOMAINS:
raise HTTPException(status_code=403, detail="This domain is not in the allowed list")
# Retrieve the content from the URL with proper headers to mimic a browser
headers = {
'User-Agent': 'Mozilla/5.0 (compatible; YourBenchBot/1.0; +https://yourbench.example.com)',
'Accept': 'text/html,application/xhtml+xml',
'Accept-Language': 'en-US,en;q=0.5',
}
response = requests.get(
url,
timeout=REQUEST_TIMEOUT,
headers=headers,
stream=True # Pour vérifier la taille avant de télécharger tout le contenu
)
response.raise_for_status()
# Vérifier le Content-Type
content_type = response.headers.get('Content-Type', '')
if not content_type.startswith(('text/html', 'text/plain', 'application/xhtml+xml')):
raise HTTPException(
status_code=400,
detail=f"Unsupported content type: {content_type}. Only HTML and text formats are supported."
)
# Vérifier la taille du contenu
content_length = int(response.headers.get('Content-Length', 0))
if content_length > MAX_CONTENT_SIZE:
raise HTTPException(
status_code=400,
detail=f"Content too large ({content_length} bytes). Maximum size: {MAX_CONTENT_SIZE} bytes."
)
# Lire le contenu avec une limite de taille
content = ""
bytes_read = 0
for chunk in response.iter_content(chunk_size=8192, decode_unicode=True):
bytes_read += len(chunk.encode('utf-8') if isinstance(chunk, str) else chunk)
if bytes_read > MAX_CONTENT_SIZE:
raise HTTPException(
status_code=400,
detail=f"Content exceeded maximum allowed size of {MAX_CONTENT_SIZE} bytes"
)
content += chunk if isinstance(chunk, str) else chunk.decode('utf-8', errors='replace')
# Extract text from HTML with BeautifulSoup using the lxml parser for better security
soup = BeautifulSoup(content, 'html.parser')
# Remove potentially dangerous elements
for element in soup(['script', 'style', 'iframe', 'object', 'embed', 'noscript']):
element.extract()
# Remove on* attributes (event handlers) from all tags
for tag in soup.find_all(True):
for attr in list(tag.attrs):
if attr.startswith('on'):
del tag[attr]
# Extract the text
text = soup.get_text()
# Clean the text (remove multiple spaces and empty lines)
lines = (line.strip() for line in text.splitlines())
chunks = (phrase.strip() for line in lines for phrase in line.split(" "))
text = '\n'.join(chunk for chunk in chunks if chunk)
# Sanitize the text to prevent any potential stored XSS
text = html.escape(text)
# Limit to 25000 characters if necessary
if len(text) > 25000:
text = text[:25000]
# Check if the text is long enough
if len(text.strip()) < MIN_FILE_LENGTH:
raise HTTPException(
status_code=400,
detail=f"The content is too short ({len(text.strip())} characters). Minimum required: {MIN_FILE_LENGTH} characters."
)
# Generate a session ID
session_id = str(uuid.uuid4())
# Create the directory structure for the session
session_dir = os.path.join(UPLOAD_ROOT, session_id)
uploaded_files_dir = os.path.join(session_dir, "uploaded_files")
os.makedirs(uploaded_files_dir, exist_ok=True)
# Path of the file to save
file_path = os.path.join(uploaded_files_dir, "document.txt")
# Save the text
with open(file_path, "w", encoding="utf-8") as f:
f.write(text)
# Store the file path for later use
session_files[session_id] = file_path
return {
"status": "uploaded",
"session_id": session_id,
"filename": "document.txt",
"text_length": len(text),
"source_url": url
}
except requests.exceptions.RequestException as e:
raise HTTPException(status_code=400, detail=f"Error retrieving the URL: {str(e)}")
except Exception as e:
raise HTTPException(status_code=500, detail=f"Error processing the URL: {str(e)}") |