Project: Logs Analysis

About

From Udacity:

You will analyze data from the logs of a web service to answer questions such as "What is the most popular page?" and "When was the error rate high?" using advanced SQL queries.

Supporting courses:

• Intro to Relational Databases

Usage

usage: logs-analysis [options]

Tool for analyzing news data provided by Udacity

optional arguments:
  -h, --help            show this help message and exit
  -a, --all             show answers to all analysis questions
  -ta, --top3-articles  show the three most popular articles of all time
  -tu, --top-authors    show the most popular authors of all time
  -et, --error-thresh   show the days where >1% of requests led to errors

Views

Postgres views used for logs-analysis script.

CREATE VIEW requests_by_day AS
SELECT time::date AS day, count(*) AS requests
FROM log
GROUP BY day;

CREATE VIEW errors_by_day AS
SELECT time::date AS day, count(*) AS errors
FROM log
WHERE status = '404 NOT FOUND'
GROUP BY day;

Requirements

• Vagrant or postgres and psycopy2
• The news database and news data are installed (see Install Newsdata)

Install

There are a number of ways to install logs-analysis depending on your current setup. To install from scratch:

1. Install the VM
2. Install the news data
3. Install logs-analysis

Install VM

1. git clone https://github.com/udacity/fullstack-nanodegree-vm.git fullstack && cd fullstack/vagrant
2. vagrant up
3. vagrant ssh

Install Newsdata

1. wget https://github.com/brenj/logs-analysis/blob/master/newsdata.zip?raw=true -O newsdata.zip
2. unzip newsdata.zip
3. psql -d news -f newsdata.sql

Note: newsdata.zip from brenj/logs-analysis contains the necessary views to run logs-analysis. If you are using the original newsdata.zip data then make sure to create the required views (see Views).

Install logs-analysis

1. git clone https://github.com/brenj/logs-analysis.git && cd logs-analysis
2. ./logs-analysis.py --all

Code Quality

This code base adheres to the PEP8 standard.

Grading (by Udacity)

Criteria	Highest Grade Possible	Grade Recieved
Functionality	Meets Specifications	Meets Specifications
Code Quality	Meets Specifications	Meets Specifications
README File	Meets Specifications	Meets Specifications