Usage: ruby parser.rb [options] -h, --help Help -f, --format=FORMAT Output Format [stdout] default: stdout -a, --aggregator=AGGREGATOR Aggregator to use [simple|valid-only] default: simple -s, --sort-order=SORT_ORDER Sort Order [unique-visits|visits] default: visits
bundle exec ruby parser.rb spec/fixtures/full_webserver_valid_ips.log
ruby parser.rb spec/fixtures/full_webserver_valid_ips.log -s unique-visits
ruby parser.rb spec/fixtures/full_webserver_invalid_ips.log -a simple
- Invalid IP Addresses will not be counted
- IPv6 is not a requirement but should be considered
- Only URL paths will be supported i.e.
/pathj
It is a subclass of the BaseParser. This enables a different parser to be implemented by implementing the run method
It is a subclass of the BaseAggregator. This enables a different parser to be implemented by implementing the add method
It is a subclass of the BaseFormatter. This enables a new formatter to be implemented by implementing the display method
ipaddr stdlib is used for validation which supports IPv6
The built in CSVParser processes line by line and does not load the whole document into memory. If a faster CSV parser comes along later this can be implemented by subclassing the BaseParser
Unique ip addresses are saved in a set to conserve memory (as opposed to putting them all into an array and counting the unique instances)
These are the features that would get added if more time was given:
- Improve
test coverage - More comprehensive logging
- Windows batch file to run program
- More output formats (PDF, XML, HTML)
- Command Line Option to allow invalid IP addresses to be included in the aggregate data (needs to be discussed)
- Consider moving the validation to use ActiveModel