csc326-project

---

AMAZON WEB SERVICES INFORMATION:

Public IP Address: 54.174.107.175

To Access:

• Go to http://ec2-54-174-107-175.compute-1.amazonaws.com/

Instance setup script:

• Found in /AWSInstances/InitializeInstances.py

---

BENCHMARK SETUP:

The following commands were used to perform preliminary benchmarking of our application:

RESULTS AVAILABLE AT:

• Lab2: RESULTS.pdf
• Lab3: RESULTS-LAB3.pdf

1. Analyzing Application Performance:
   $ ab -n 318 -c 318 http://54.174.107.175/?keywords=helloworld+foo+bar

2. Anyalyzing CPU performance:
   $ dstat --cpu -d --mem --net -io
   $ ab -n 1000 -c318 http://54.174.107.175/?keywords=helloworld+foo+bar

NOTE: The results of our performance analysis can be found in RESULTS.docx

---

HOW TO USE:

DOWNLOADING EXTERNAL PACKAGES:

Enter the following commands in:
Terminal (Mac) Command Prompt (Windows)

1. Installing Bottle Framework:
   $ pip install bottle

2. Installing BeautifulSoup Library (V 4.6.0)
   $ pip install beautifulsoup4

3. Installing NumPy (V 1.13.3)
   $ pip install numpy

4. Installing Beaker
   $ pip install Beaker

5. Installing OAuth2 Library
   $ pip install oauth2client

6. Instaling Google API Client for Python
   $ pip install --upgrade google-api-python-client

7. Install http2 (AMI Linux dependency)
   $ pip install httplib2

8. Install PyMongo
   $ pip install pymongo

NOTE: For MAC users installing OAuth2 Library

• It is recommended to use a virtual environment for the command line so as not to conflict with existing files during the oauth2client installation process
• Can install virtual environment with the following command: $ pip install virtualenv

---

RUNNING WEB APPLICATION ON LOCAL MACHINE:

Code Alterations:

1. In /MainApp.py:

• Change Bottle Script to run with host='localhost' and port='8000'
• Change Google Redirect URL in Flow class initialization to redirect_uri='http://localhost:8000/redirect'
• Change Google CLIENT_ID and CLIENT_SECRET to correct values

Running Application: 2. Enter ' $ python MainApp.py ' in: Terminal (Mac) Command Prompt (Windows) 2. Open up web browser. Navigate to 'localhost:8000' 3. Enter search string in input box.

Anonymous Mode:

• The results table with words and their word counts will be displayed

Signed_in Mode:

• Results table, 20 most popular keyword searches, and 10 most recent words will be displayed
• Your search history will be saved for every subsequent login

---

TESTING:

• There are three files used for testing the application, one for front-end and two for back-end.

Front-end:

• Lab 1: Tests the results obtained from the user's input query (ie words and their word counts)
• Lab 1: Tests the history of user inputs (ie correctness of the top twenty searched keywords)

Back-end:

• Lab 1: Tests get_inverted_index() and get_resolved_inverted_index() functions.
• Lab 2: Tests functionality of session management class

---

How to Test:

1. On the command line, navigate to the project directory.
   $ cd /path/to/project/directory

2. Testing Front-end and Back-end functionalities
   Run the following commands:

   LAB 3: PAGERANK TESTS:
   $ python -m UnitTests.PageRankCrawlerTests

   LAB 1 & 2: CRAWLER & RESULTSPAGE TESTS:
   

• Front-end:
  $ python -m UnitTests.ResultsPageServicesTest
• Back-end:
  $ python -m UnitTests.WebScrapingServicesTest
• Session Management:
  $ python -m UnitTests.UserSessionManagerTests

---

IMPORTANT: before run Back-end unit test, start application by running (from project root):

• $ python MainApp.py & (only necessary for lab 1 & 2 unittests)