implement multisecret function

This commit is contained in:
lelo 2025-03-22 01:55:12 +01:00
parent 5ace803cc0
commit 663d5b50f9
4 changed files with 123 additions and 92 deletions

2
.gitignore vendored
View File

@ -2,4 +2,4 @@
/filecache /filecache
/__pycache__ /__pycache__
/access_log.db /access_log.db
/allowed_secrets.json /folder_config.json

View File

@ -1,6 +0,0 @@
{
"dev_key_f83745ft0g5rg3": {
"expiry" : "31.12.2000",
"file_root": "\\\\path\\if\\using\\windows"
}
}

192
app.py
View File

@ -1,4 +1,4 @@
from flask import Flask, render_template, send_file, url_for, jsonify, request, session, send_from_directory from flask import Flask, render_template, send_file, url_for, jsonify, request, session, send_from_directory, abort
import os import os
from PIL import Image from PIL import Image
import io import io
@ -12,7 +12,7 @@ import geoip2.database
from functools import lru_cache from functools import lru_cache
from urllib.parse import urlparse, unquote from urllib.parse import urlparse, unquote
from werkzeug.middleware.proxy_fix import ProxyFix from werkzeug.middleware.proxy_fix import ProxyFix
cache = diskcache.Cache('./filecache', size_limit= 48 * 1024**3) # 32 GB limit cache = diskcache.Cache('./filecache', size_limit= 48 * 1024**3) # 48 GB limit
app = Flask(__name__) app = Flask(__name__)
app.wsgi_app = ProxyFix(app.wsgi_app, x_for=1, x_proto=1) app.wsgi_app = ProxyFix(app.wsgi_app, x_for=1, x_proto=1)
@ -23,58 +23,74 @@ if os.environ.get('FLASK_ENV') == 'production':
app.config['SESSION_COOKIE_SAMESITE'] = 'None' app.config['SESSION_COOKIE_SAMESITE'] = 'None'
app.config['SESSION_COOKIE_SECURE'] = True app.config['SESSION_COOKIE_SECURE'] = True
def load_allowed_secrets(filename='allowed_secrets.json'): with open('folder_config.json') as file:
with open(filename) as f: app.config['folder_config'] = json.load(file)
secrets = json.load(f)
for key, value in secrets.items():
if 'expiry' in value:
value['expiry'] = datetime.strptime(value['expiry'], '%d.%m.%Y').date()
return secrets
def require_secret(f): def require_secret(f):
@wraps(f) @wraps(f)
def decorated_function(*args, **kwargs): def decorated_function(*args, **kwargs):
allowed_secrets = load_allowed_secrets() # Your config list:
today = date.today() folder_config = app.config['folder_config']
def is_valid(secret_data): def is_valid(config_item, provided_secret):
expiry_date = secret_data.get('expiry') """
is_valid = expiry_date and today <= expiry_date Checks if today's date is <= validity date
return is_valid AND if the provided secret matches config_item['secret'].
"""
folder_validity = config_item['validity']
# Convert string to a date if necessary:
if isinstance(folder_validity, str):
folder_validity = datetime.strptime(folder_validity, '%d.%m.%Y').date()
# Check if a secret was provided via GET parameter # Return whether it's still valid and secrets match:
get_secret = request.args.get('secret') return (
if get_secret is not None: date.today() <= folder_validity and
secret_data = allowed_secrets.get(get_secret) provided_secret == config_item['secret']
if secret_data: )
if is_valid(secret_data):
# Valid secret provided in URL: update session and config
session['secret'] = get_secret
session.permanent = True
app.config['FILE_ROOT'] = secret_data.get('file_root')
print("session:", session['secret'])
return f(*args, **kwargs)
else:
# Secret provided via URL is expired or invalid
return render_template('error.html', message="Invalid or expired secret."), 403
# 1) Get secret from query params (if any)
args_secret = request.args.get('secret')
# 2) Initialize 'allowed_secrets' in the session if missing
if 'allowed_secrets' not in session:
session['allowed_secrets'] = []
# If no secret provided via GET, check the session # 3) If a new secret is provided, check if its valid, and add to session if so
session_secret = session.get('secret') if args_secret:
if session_secret is not None: for config_item in folder_config:
secret_data = allowed_secrets.get(session_secret) if is_valid(config_item, args_secret):
if secret_data: if args_secret not in session['allowed_secrets']:
if is_valid(secret_data): session['allowed_secrets'].append(args_secret)
session.permanent = True session.permanent = True # Make the session permanent
app.config['FILE_ROOT'] = secret_data.get('file_root')
return f(*args, **kwargs)
else:
# Session secret exists but is expired
return render_template('error.html', message="Invalid or expired secret."), 403
# No secret provided at all; show the public index page # 4) Re-check validity of each secret in session['allowed_secrets']
return render_template('index.html') # If a secret is no longer valid (or not in config), remove it.
for secret_in_session in session['allowed_secrets'][:]:
# Find the current config item with matching secret
config_item = next(
(c for c in folder_config if c['secret'] == secret_in_session),
None
)
# If the config item doesnt exist or is invalid, remove secret
if config_item is None or not is_valid(config_item, secret_in_session):
session['allowed_secrets'].remove(secret_in_session)
# 5) Build session['folders'] fresh from the valid secrets
session['folders'] = {}
for secret_in_session in session.get('allowed_secrets', []):
config_item = next(
(c for c in folder_config if c['secret'] == secret_in_session),
None
)
if config_item:
for folder_info in config_item['folders']:
session['folders'][folder_info['foldername']] = folder_info['folderpath']
# 6) If we have folders, proceed; otherwise show index
if session['folders']:
return f(*args, **kwargs)
else:
return render_template('index.html')
return decorated_function return decorated_function
@ -165,7 +181,7 @@ def list_directory_contents(directory, subpath):
file_entry['has_transcript'] = True file_entry['has_transcript'] = True
transcript_rel_path = os.path.join(subpath, "Transkription", transcript_filename) if subpath else os.path.join("Transkription", transcript_filename) transcript_rel_path = os.path.join(subpath, "Transkription", transcript_filename) if subpath else os.path.join("Transkription", transcript_filename)
transcript_rel_path = transcript_rel_path.replace(os.sep, '/') transcript_rel_path = transcript_rel_path.replace(os.sep, '/')
file_entry['transcript_url'] = url_for('get_transcript', filename=transcript_rel_path) file_entry['transcript_url'] = url_for('get_transcript', subpath=transcript_rel_path)
else: else:
file_entry['has_transcript'] = False file_entry['has_transcript'] = False
else: else:
@ -176,7 +192,7 @@ def list_directory_contents(directory, subpath):
return directories, files return directories, files
def generate_breadcrumbs(subpath): def generate_breadcrumbs(subpath=None):
breadcrumbs = [{'name': 'Home', 'path': ''}] breadcrumbs = [{'name': 'Home', 'path': ''}]
if subpath: if subpath:
parts = subpath.split('/') parts = subpath.split('/')
@ -191,8 +207,20 @@ def generate_breadcrumbs(subpath):
@app.route('/api/path/<path:subpath>') @app.route('/api/path/<path:subpath>')
@require_secret @require_secret
def api_browse(subpath): def api_browse(subpath):
file_root = app.config['FILE_ROOT'] if subpath == '': # root directory
directory = os.path.join(file_root, subpath.replace('/', os.sep)) foldernames = []
for foldername, folderpath in session['folders'].items():
foldernames.append({'name': foldername, 'path': foldername})
return jsonify({
'breadcrumbs': generate_breadcrumbs(),
'directories': foldernames,
'files': []
})
root, *relative_parts = subpath.split('/')
base_path = session['folders'][root]
directory = os.path.join(base_path, *relative_parts)
if not os.path.isdir(directory): if not os.path.isdir(directory):
return jsonify({'error': 'Directory not found'}), 404 return jsonify({'error': 'Directory not found'}), 404
@ -399,11 +427,12 @@ def log_file_access(full_path):
conn.commit() conn.commit()
conn.close() conn.close()
@app.route("/media/<path:filename>") @app.route("/media/<path:subpath>")
@require_secret @require_secret
def serve_file(filename): def serve_file(subpath):
decoded_filename = unquote(filename).replace('/', os.sep) root, *relative_parts = subpath.split('/')
full_path = os.path.normpath(os.path.join(app.config['FILE_ROOT'], decoded_filename)) base_path = session['folders'][root]
full_path = os.path.join(base_path, *relative_parts)
if not os.path.isfile(full_path): if not os.path.isfile(full_path):
app.logger.error(f"File not found: {full_path}") app.logger.error(f"File not found: {full_path}")
@ -424,7 +453,7 @@ def serve_file(filename):
# Check cache first (using diskcache) # Check cache first (using diskcache)
response = None response = None
cached = cache.get(filename) cached = cache.get(subpath)
if cached: if cached:
cached_file_bytes, mime = cached cached_file_bytes, mime = cached
cached_file = io.BytesIO(cached_file_bytes) cached_file = io.BytesIO(cached_file_bytes)
@ -438,20 +467,20 @@ def serve_file(filename):
img_bytes = io.BytesIO() img_bytes = io.BytesIO()
img.save(img_bytes, format='PNG', quality=85) img.save(img_bytes, format='PNG', quality=85)
img_bytes = img_bytes.getvalue() img_bytes = img_bytes.getvalue()
cache.set(filename, (img_bytes, mime)) cache.set(subpath, (img_bytes, mime))
response = send_file(io.BytesIO(img_bytes), mimetype=mime) response = send_file(io.BytesIO(img_bytes), mimetype=mime)
except Exception as e: except Exception as e:
app.logger.error(f"Image processing failed for {filename}: {e}") app.logger.error(f"Image processing failed for {subpath}: {e}")
abort(500) abort(500)
else: else:
# Cache non-image files: read bytes and cache # Cache non-image files: read bytes and cache
try: try:
with open(full_path, 'rb') as f: with open(full_path, 'rb') as f:
file_bytes = f.read() file_bytes = f.read()
cache.set(filename, (file_bytes, mime)) cache.set(subpath, (file_bytes, mime))
response = send_file(io.BytesIO(file_bytes), mimetype=mime) response = send_file(io.BytesIO(file_bytes), mimetype=mime)
except Exception as e: except Exception as e:
app.logger.error(f"Failed to read file {filename}: {e}") app.logger.error(f"Failed to read file {subpath}: {e}")
abort(500) abort(500)
# Set Cache-Control header (browser caching for 1 day) # Set Cache-Control header (browser caching for 1 day)
@ -459,11 +488,13 @@ def serve_file(filename):
return response return response
@app.route("/transcript/<path:filename>") @app.route("/transcript/<path:subpath>")
@require_secret @require_secret
def get_transcript(filename): def get_transcript(subpath):
fs_filename = filename.replace('/', os.sep)
full_path = os.path.join(app.config['FILE_ROOT'], fs_filename) root, *relative_parts = subpath.split('/')
base_path = session['folders'][root]
full_path = os.path.join(base_path, *relative_parts)
if not os.path.isfile(full_path): if not os.path.isfile(full_path):
return "Transcription not found", 404 return "Transcription not found", 404
@ -472,34 +503,27 @@ def get_transcript(filename):
content = f.read() content = f.read()
return content, 200, {'Content-Type': 'text/markdown; charset=utf-8'} return content, 200, {'Content-Type': 'text/markdown; charset=utf-8'}
@app.route("/crawl/<path:start_relative_path>") @app.route("/crawl/<path:subpath>")
@require_secret @require_secret
def crawl_and_cache(start_relative_path): def crawl_and_cache(subpath):
""" """
Crawls through a directory (relative to app.config['FILE_ROOT']) and caches each file. Crawls through a directory and caches each file.
For images, it creates a thumbnail (max 1200x1200) and caches the processed image. For images, it creates a thumbnail (max 1200x1200) and caches the processed image.
For non-images, it simply reads and caches the file bytes. For non-images, it simply reads and caches the file bytes.
:param start_relative_path: The folder (relative to FILE_ROOT) to start crawling.
""" """
# Compute the absolute path for the starting directory root, *relative_parts = subpath.split('/')
base_dir = os.path.normpath(os.path.join(app.config['FILE_ROOT'], start_relative_path)) base_path = session['folders'][root]
full_path = os.path.join(base_path, *relative_parts)
# Check that base_dir is under FILE_ROOT to prevent directory traversal
if not base_dir.startswith(os.path.abspath(app.config['FILE_ROOT'])):
return jsonify({"error": "Invalid path"}), 400
cached_files = [] # List to hold cached file relative paths cached_files = [] # List to hold cached file relative paths
# Walk through all subdirectories and files # Walk through all subdirectories and files
for root, dirs, files in os.walk(base_dir): for root, dirs, files in os.walk(full_path):
for filename in files: for filename in files:
full_path = os.path.join(root, filename) full_path_file = os.path.join(root, filename)
# Compute the relative key used for caching
rel_key = os.path.relpath(full_path, app.config['FILE_ROOT'])
# Skip if this file is already in the cache # Skip if this file is already in the cache
if cache.get(rel_key): if cache.get(full_path_file):
continue continue
# Determine the MIME type # Determine the MIME type
@ -517,19 +541,19 @@ def crawl_and_cache(start_relative_path):
img.save(img_bytes_io, format='PNG', quality=85) img.save(img_bytes_io, format='PNG', quality=85)
img_bytes = img_bytes_io.getvalue() img_bytes = img_bytes_io.getvalue()
# Cache the processed image bytes along with its mime type # Cache the processed image bytes along with its mime type
cache.set(rel_key, (img_bytes, mime)) cache.set(full_path_file, (img_bytes, mime))
cached_files.append(rel_key) cached_files.append(full_path_file)
except Exception as e: except Exception as e:
app.logger.error(f"Image processing failed for {rel_key}: {e}") app.logger.error(f"Image processing failed for {full_path_file}: {e}")
else: else:
# Process non-image files # Process non-image files
try: try:
with open(full_path, 'rb') as f: with open(full_path_file, 'rb') as f:
file_bytes = f.read() file_bytes = f.read()
cache.set(rel_key, (file_bytes, mime)) cache.set(full_path_file, (file_bytes, mime))
cached_files.append(rel_key) cached_files.append(full_path_file)
except Exception as e: except Exception as e:
app.logger.error(f"Failed to read file {rel_key}: {e}") app.logger.error(f"Failed to read file {full_path_file}: {e}")
# Return the list of cached files as a JSON response # Return the list of cached files as a JSON response
return json.dumps({"cached_files": cached_files}, indent=4), 200 return json.dumps({"cached_files": cached_files}, indent=4), 200

View File

@ -0,0 +1,13 @@
[
{
"secret" : "dev_key_f83745ft0g5rg3",
"validity" : "31.12.2030",
"folders": [
{
"foldername": "My Folder",
"folderpath": "\\\\path\\if\\using\\windows"
}
]
}
]