291 lines
10 KiB
Python
291 lines
10 KiB
Python
from flask import render_template, request
|
|
from datetime import datetime, timedelta
|
|
import geoip2.database
|
|
from auth import require_secret
|
|
import os
|
|
import psycopg2
|
|
|
|
file_access_temp = []
|
|
|
|
dbname = os.environ.get('DB_NAME')
|
|
user = os.environ.get('DB_USER')
|
|
password = os.environ.get('DB_PASSWORD')
|
|
host = os.environ.get('DB_HOST')
|
|
port = int(os.environ.get('DB_PORT', 5432))
|
|
|
|
connection = psycopg2.connect(dbname=dbname,
|
|
user=user,
|
|
password=password,
|
|
host=host,
|
|
port=port
|
|
)
|
|
# Enable autocommit
|
|
connection.autocommit = True
|
|
log_db = connection
|
|
|
|
# Function to initialize the database.
|
|
def init_log_db():
|
|
with log_db.cursor() as cursor:
|
|
cursor.execute('''
|
|
CREATE TABLE IF NOT EXISTS file_access_log (
|
|
id SERIAL PRIMARY KEY,
|
|
timestamp TIMESTAMP,
|
|
rel_path TEXT,
|
|
filesize BIGINT,
|
|
mime TEXT,
|
|
ip_address TEXT,
|
|
user_agent TEXT,
|
|
device_id TEXT,
|
|
cached BOOLEAN
|
|
)
|
|
''')
|
|
|
|
init_log_db()
|
|
|
|
|
|
|
|
def lookup_location(ip, reader):
|
|
try:
|
|
response = reader.city(ip)
|
|
country = response.country.name if response.country.name else "Unknown"
|
|
city = response.city.name if response.city.name else "Unknown"
|
|
return country, city
|
|
except Exception:
|
|
return "Unknown", "Unknown"
|
|
|
|
def get_device_type(user_agent):
|
|
"Classify device type based on user agent string"
|
|
if 'Android' in user_agent:
|
|
return 'Android'
|
|
elif 'iPhone' in user_agent or 'iPad' in user_agent:
|
|
return 'iOS'
|
|
elif 'Windows' in user_agent:
|
|
return 'Windows'
|
|
elif 'Macintosh' in user_agent or 'Mac OS' in user_agent:
|
|
return 'MacOS'
|
|
elif 'Linux' in user_agent:
|
|
return 'Linux'
|
|
else:
|
|
return 'Other'
|
|
|
|
# Logging function that uses the singleton connection.
|
|
def log_file_access(rel_path, filesize, mime, ip_address, user_agent, device_id, cached):
|
|
global file_access_temp
|
|
timestamp = datetime.now() # Use datetime object directly
|
|
with log_db.connection.cursor() as cursor:
|
|
cursor.execute('''
|
|
INSERT INTO file_access_log (timestamp, rel_path, filesize, mime, ip_address, user_agent, device_id, cached)
|
|
VALUES (%s, %s, %s, %s, %s, %s, %s, %s)
|
|
''', (timestamp, rel_path, filesize, mime, ip_address, user_agent, device_id, cached))
|
|
file_access_temp.insert(0, [timestamp.isoformat(), rel_path, filesize, mime, ip_address, user_agent, device_id, cached])
|
|
return timestamp.isoformat()
|
|
|
|
def return_file_access():
|
|
global file_access_temp
|
|
if file_access_temp:
|
|
cutoff_time = datetime.now() - timedelta(minutes=10)
|
|
file_access_temp[:] = [
|
|
entry for entry in file_access_temp
|
|
if datetime.fromisoformat(entry[0]) >= cutoff_time
|
|
]
|
|
return file_access_temp
|
|
else:
|
|
return []
|
|
|
|
@require_secret
|
|
def connections():
|
|
return render_template('connections.html')
|
|
|
|
@require_secret
|
|
def dashboard():
|
|
filetype_arg = request.args.get('filetype', 'audio')
|
|
timeframe = request.args.get('timeframe', 'today')
|
|
now = datetime.now()
|
|
|
|
# Determine which file type we're filtering by.
|
|
filetype = 'other'
|
|
|
|
allowed_list = ['mp3', 'wav', 'audio']
|
|
if filetype_arg.lower() in allowed_list:
|
|
filetype = 'audio/'
|
|
|
|
allowed_list = ['jpg', 'jpeg', 'image', 'photo']
|
|
if filetype_arg.lower() in allowed_list:
|
|
filetype = 'image/'
|
|
|
|
allowed_list = ['mp4', 'mov', 'wmv', 'avi']
|
|
if filetype_arg.lower() in allowed_list:
|
|
filetype = 'video/'
|
|
|
|
|
|
|
|
# Determine the start time based on timeframe.
|
|
if timeframe == 'today':
|
|
start = now.replace(hour=0, minute=0, second=0, microsecond=0)
|
|
elif timeframe == '7days':
|
|
start = now - timedelta(days=7)
|
|
elif timeframe == '30days':
|
|
start = now - timedelta(days=30)
|
|
elif timeframe == '365days':
|
|
start = now - timedelta(days=365)
|
|
else:
|
|
start = now.replace(hour=0, minute=0, second=0, microsecond=0)
|
|
|
|
# Build the SQL filter for mime
|
|
if filetype == 'other':
|
|
# Exclude audio, image, and video mimes
|
|
filetype_filter_sql = "AND mime NOT LIKE 'audio/%' AND mime NOT LIKE 'image/%' AND mime NOT LIKE 'video/%'"
|
|
params = (start,)
|
|
else:
|
|
# Filter for mimes that start with the given type.
|
|
filetype_filter_sql = "AND mime LIKE %s"
|
|
params = (start, filetype + '%')
|
|
|
|
with log_db.connection.cursor() as cursor:
|
|
# Raw file access counts (top files)
|
|
query = f'''
|
|
SELECT rel_path, COUNT(*) as access_count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY rel_path
|
|
ORDER BY access_count DESC
|
|
LIMIT 20
|
|
'''
|
|
cursor.execute(query, params)
|
|
rows = cursor.fetchall()
|
|
|
|
# Daily access trend for a line chart
|
|
query = f'''
|
|
SELECT CAST(timestamp AS DATE) as date, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY CAST(timestamp AS DATE)
|
|
ORDER BY date
|
|
'''
|
|
cursor.execute(query, params)
|
|
daily_access_data = [dict(date=str(row[0]), count=row[1]) for row in cursor.fetchall()]
|
|
|
|
# Aggregate download counts by time bucket according to the timeframe.
|
|
if timeframe == 'today':
|
|
query = f'''
|
|
SELECT to_char(timestamp, 'HH24') as bucket, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY bucket
|
|
ORDER BY bucket
|
|
'''
|
|
cursor.execute(query, params)
|
|
elif timeframe in ('7days', '30days'):
|
|
query = f'''
|
|
SELECT CAST(timestamp AS DATE) as bucket, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY bucket
|
|
ORDER BY bucket
|
|
'''
|
|
cursor.execute(query, params)
|
|
elif timeframe == '365days':
|
|
query = f'''
|
|
SELECT to_char(timestamp, 'YYYY-MM') as bucket, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY bucket
|
|
ORDER BY bucket
|
|
'''
|
|
cursor.execute(query, params)
|
|
else:
|
|
query = f'''
|
|
SELECT CAST(timestamp AS DATE) as bucket, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY bucket
|
|
ORDER BY bucket
|
|
'''
|
|
cursor.execute(query, params)
|
|
timeframe_data = [dict(bucket=row[0], count=row[1]) for row in cursor.fetchall()]
|
|
|
|
# User agent distribution (aggregate by device type)
|
|
query = f'''
|
|
SELECT user_agent, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY user_agent
|
|
ORDER BY count DESC
|
|
'''
|
|
cursor.execute(query, params)
|
|
raw_user_agents = [dict(user_agent=row[0], count=row[1]) for row in cursor.fetchall()]
|
|
device_counts = {}
|
|
for entry in raw_user_agents:
|
|
device = get_device_type(entry['user_agent'])
|
|
device_counts[device] = device_counts.get(device, 0) + entry['count']
|
|
user_agent_data = [dict(device=device, count=count) for device, count in device_counts.items()]
|
|
|
|
# Parent folder distribution
|
|
query = f'''
|
|
SELECT rel_path, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY rel_path
|
|
ORDER BY count DESC
|
|
'''
|
|
cursor.execute(query, params)
|
|
folder_data = {}
|
|
for row in cursor.fetchall():
|
|
rel_path = row[0]
|
|
parent_folder = rel_path.rsplit('/', 1)[0] if '/' in rel_path else "Root"
|
|
folder_data[parent_folder] = folder_data.get(parent_folder, 0) + row[1]
|
|
folder_data = [dict(folder=folder, count=count) for folder, count in folder_data.items()]
|
|
folder_data.sort(key=lambda x: x['count'], reverse=True)
|
|
folder_data = folder_data[:10]
|
|
|
|
# Aggregate IP addresses with counts
|
|
query = f'''
|
|
SELECT ip_address, COUNT(*) as count
|
|
FROM file_access_log
|
|
WHERE timestamp >= %s {filetype_filter_sql}
|
|
GROUP BY ip_address
|
|
ORDER BY count DESC
|
|
'''
|
|
cursor.execute(query, params)
|
|
ip_rows = cursor.fetchall()
|
|
|
|
# Summary stats using separate SQL queries
|
|
query = f'SELECT COUNT(*) FROM file_access_log WHERE timestamp >= %s {filetype_filter_sql}'
|
|
cursor.execute(query, params)
|
|
total_accesses = cursor.fetchone()[0]
|
|
|
|
query = f'SELECT COUNT(DISTINCT rel_path) FROM file_access_log WHERE timestamp >= %s {filetype_filter_sql}'
|
|
cursor.execute(query, params)
|
|
unique_files = cursor.fetchone()[0]
|
|
|
|
query = f'SELECT COUNT(DISTINCT device_id) FROM file_access_log WHERE timestamp >= %s {filetype_filter_sql}'
|
|
cursor.execute(query, params)
|
|
unique_user = cursor.fetchone()[0]
|
|
|
|
# Process location data with GeoIP2.
|
|
reader = geoip2.database.Reader('GeoLite2-City.mmdb')
|
|
location_data = {}
|
|
for ip, count in ip_rows:
|
|
country, city = lookup_location(ip, reader)
|
|
key = (country, city)
|
|
location_data[key] = location_data.get(key, 0) + count
|
|
reader.close()
|
|
|
|
location_data = [dict(country=key[0], city=key[1], count=value) for key, value in location_data.items()]
|
|
location_data.sort(key=lambda x: x['count'], reverse=True)
|
|
location_data = location_data[:20]
|
|
|
|
return render_template("dashboard.html",
|
|
timeframe=timeframe,
|
|
rows=rows,
|
|
daily_access_data=daily_access_data,
|
|
user_agent_data=user_agent_data,
|
|
folder_data=folder_data,
|
|
location_data=location_data,
|
|
total_accesses=total_accesses,
|
|
unique_files=unique_files,
|
|
unique_user=unique_user,
|
|
timeframe_data=timeframe_data)
|
|
|
|
|