masterots/capstone-mongo

Capstone project for HudlU Mongo

Prerequisites

node
mocha
mongo (remote server is fine)

Steps

clone npm install update database-config.json mongo urls

Commands

npm run build

will compile source files. Compiled source is checked in, this step is not necessary to run project without changes.

npm run generate

Adds data to your prod database. Currently set to add 500K documents to the books collection. Must run npm run build if changing number of documents generated.

npm test

Generates test data and runs tests.

npm start

Will run web server at http://localhost:3000 URLS:

/books/isbn

add ?isbn=<isbn number> to search for a book by isbn

/books/isbnexplain

add ?isbn=<isbn number> to search for a book by isbn. returns mongo's explain results

/books/author

add ?author=<author> to search for a book by author. accepts partial name

/books/publisher

add ?publisher=<publisher> to search for a book by publisher. accepts partial name

/books/title

add ?title=<title> to search for a book by title. accepts partial title

/reports/countBooksReleasedPerYear

/reports/countBooksReleasedPerYearPerPublisher

/reports/booksByAuthor

/reports/avgNumAuthorsPerBookPerYear

Additional info

ISBN-Search-No-Index.json and ISBN-Search-Index.json show the massive improvement in text search by ISBN over 500K documents when an index is added. I wasn't "seeing" any performance issues with my eyes, but the numbers absolutely speak for the improvements a simple index adds to a database.

I did spend time attempting to get "corrupted" data working in my reports, but ultimately didn't feel like it was prudent to spend any more time on it without teammates. I had tried putting a date as a string on a book, and then convert it at aggregation time, but I was getting bad dates out of it. I'd like to spend more time investigating it, outside of this project.