Introducing the GoodreadsBooks CLI gem!
This command line application lists books that are Winners of Goodreads Choice Awards. By default, the app will list the awards winning books from the most recent year as published by Goodreads.
Users have the option to select a different Choice Awards year to get a list of the winning books from that chosen year. The options avaliable for scraping are from 2010 to the most recent year, which is usually the previous calendar year. From the list, users have the option to choose a book and get detailed information about their selected book. In addition, users can select to open a link to Goodreads website to read the reviews & information about that book.
This is the second Ruby Gem that I’ve built for the CLI Data Gem Project assessment. Having completed the first application early February, I figured it would take me a while to peruse and refactor the first application to function better (hint: the first Upcoming Book Releases data gem is slow and needs to be refactored). I might as well build another gem, keeping in mind the performance issue that I might face in scraping the data.
It was so much easier to build the CLI gem a second time. Initially, I planned on coding the user interface prompts and using heredoc to simulate the data. However, once I started and after examining the html codes and identifying the data for scraping, I decided to change my approach and get the data scraping methods working first.
When testing the completed GoodreadsBooks CLI gem, I realized there was a performance issue. It took around 60 seconds to scrape the 20 winning books from year 2017. This performance issue would surface each time users selected a new awards year, which called the Nokogiri parser to access and scrape the best book of each category data from Goodreads web page . For each category’s best book, Nokogiri parser is called again to access and scrape detailed information of that book (2nd level of html page). The advantage in getting the detailed information of the best books upfront, was that GoodreadsBooks would efficiently display the book information each time users selected a best book for viewing.
A trade off was to fix the slow upfront loading time, and having a slower display time when users selected a new best book for detailed viewing. To change the code, refactoring was needed to only scrape the detailed book information when users selected that book for viewing.
This modification sped up the initial loading time to just under 10 seconds each time a new awards year was selected. When users then selected a best book for detailed viewing, if the book details have not been scraped from Goodreads web page (i.e. first time viewing this book), Nokogiri parser was called to access and scrape this book’s details, resulting in a slower screen display when this function is performed.
Overall, I was pleased with the improved performance of the GoodreadsBooks CLI gem.
GoodreadsBooks CLI gem is on RubyGems
- To install the gem, type
gem install goodreads-bookson the command prompt - To run the application, type
goodreads-books
GoodreadsBooks code is on github.