Refactor for daily usage output

.gitignore

@@ -2,3 +2,4 @@ config.yml
 *.pyc
 __pycache__
 logs/
+output/

README.md

@@ -59,15 +59,15 @@ This script scans one or more `idp-process.log` files to see details about which
 
 ### Filenames
 
-In both cases, the default file to analyze is the current (live) log file, as it would be named on the idpv4 servers: `/var/log/tomcat9/access.log` and `/opt/shibboleth-idp/logs/idp-process.log`. Use `-f` to specify a different filename.
+The default file to analyze is the current (live) log file, as it would be named on the IdP v4 or IdP v5 server: `/opt/shibboleth-idp/logs/idp-process.log`. Use `-f` to specify a different filename.
 
 Multiple filenames are allowed (separate by spaces), wildcards are allowed, and filenames that end in `.gz` can be processed without unzipping them.
 
 **Example**
 
-To get every SP that has used this IdP for authentication over the entire `logs/` directory:
+To get every user and service provider that has used this IdP for authentication over the entire `logs/` directory:
 ```bash
-./logscan.py sp -f /opt/shibboleth-idp/logs/idp-process*
+./logscan.py -f /opt/shibboleth-idp/logs/idp-process*
 ```
 
 ### Options

logscan.py

@@ -2,6 +2,7 @@
 
 from argparse import ArgumentParser
 from parsers import ShibbolethLog
+import os
 
 
 def help(args):
@@ -11,24 +12,74 @@ def help(args):
 def scan(args):
     kwargs = {
         'principal': args.principal,
-        'requester': args.requester
+        'requester': args.requester,
+        # 'sso': args.sso,
+        'daily': args.daily,
+        'output': args.output,
     }
     log = ShibbolethLog(**kwargs)
     for filename in args.filename:
         log.load(filename)
     log.command_scan()
 
 
+def main(args):
+    if args.output:
+        # If we specified an output directory, make sure it exists.
+        output_dir = os.path.join(os.getcwd(), args.output)
+        if not os.path.exists(output_dir):
+            os.mkdir(output_dir)
+        args.output = output_dir
+    scan(args)
+
+
 if __name__ == '__main__':
     argp = ArgumentParser(
-        epilog='Specify neither -n nor -r to show all service providers. Specify both to see IP address and timestamp of logins.')
-    argp.add_argument('-f', '--filename', type=str, nargs='*',
-                      default=['/opt/shibboleth-idp/logs/idp-process.log'],
-                      help='Log filename(s) to process, accepts wildcards')
-    argp.add_argument('-n', '--principal', default=None, nargs='+',
-                      help='Restrict to this username and list service providers')
-    argp.add_argument('-r', '--requester', default=None, nargs='+',
-                      help='Restrict to this service provider and list usernames')
+        description='''
+            Scans Shib logs for instances of specified (or all) usernames
+            logging into specified (or all) service providers, and returns
+            counts as CSV of either totals or daily traffic.
+        ''',
+        epilog='''
+            Specify neither -n nor -r to show all usernames and service
+            providers. Specify both to see IP address and timestamp of
+            all logins.''',
+    )
+
+    subject = argp.add_argument_group('Subjects to scan for')
+    subject.add_argument(
+        '-n', '--principal', default=None, nargs='+',
+        help='Limit scan to the username(s) provided')
+    subject.add_argument(
+        '-r', '--requester', default=None, nargs='+',
+        help='Limit scan to the service provider(s) provided')
+    # subject.add_argument(
+    #     '-s', '--sso', action='store_true',
+    #     help='Determine if SSO was used within above limits')
+
+    output = argp.add_argument_group('Output options')
+    # TODO: -d needs exactly one of -n or -r.
+    output.add_argument(
+        '-d', '--daily', action='store_true',
+        help='Provide daily usage as CSV for exactly one of -n or -r')
+    output.add_argument(
+        '-o', '--output', default=None, nargs='?',
+        help='Create logs of results in this output directory')
+    # output.add_argument(
+    #     '-v', '--verbose', action='store_true',
+    #     help='Provide verbose output')
+
+    targets = argp.add_argument_group('Which log files to scan')
+    targets.add_argument(
+        '-f', '--filename', type=str, nargs='*',
+        default=['/opt/shibboleth-idp/logs/idp-process.log'],
+        help='Log filename(s) to process, accepts wildcards')
 
     args = argp.parse_args()
-    scan(args)
+    if args.daily:
+        if ((args.principal and args.requester)
+            or (not args.principal and not args.requester)):
+            print('The -d/--daily option requires exactly one of -n/--principal or -r/--requester')
+            exit(1)
+
+    main(args)