- We will be looking at the readability and cleanliness of your code
- Provide unit tests. The cleanliness and readbility of tests is just as important as your production code.
- Please provide 80% test coverage for your code.
- Think simple. Readability and modularity are better than being clever
- Make a github repository and commit your work in small cohesive chunks.
Create a command line app that takes as input a file with a set of records in one of three formats described below, and outputs (to the screen) the set of records sorted in one of three ways.
A record consists of the following 5 fields: last name, first name, gender, date of birth, and favorite color. The input is 3 files, each containing records stored in a different format. You may generate these files yourself, and you can make certain assumptions if it makes solving your problem easier.
-
The pipe-delimited file lists each record as follows:
LastName | FirstName | Gender | FavoriteColor | DateOfBirth
-
The comma-delimited file looks like this:
LastName, FirstName, Gender, FavoriteColor, DateOfBirth
-
The space-delimited file looks like this:
LastName FirstName Gender FavoriteColor DateOfBirth
Create and display 3 different views of the data you read in:
- Output 1 - sorted by gender (females before males) then by last name ascending
- Output 2 - sorted by birth date, ascending
- Output 3 - sorted by last name, descending
Display dates in the format M/D/YYYY
To check output for step 1, execute the command lein run
Tests for this section are required as well.
Within the same code base, build a standalone REST API with the following endpoints:
- POST /records - Post a single data line in any of the 3 formats supported by your existing code
- GET /records/gender - return records sorted by gender
- GET /records/birthdate - returns records sorted by birthdate
- GET /records/name - returns records sorted by name
It's your choice how you render the output from these endpoints as long as it is well structured data. These endpoints should return JSON.
To keep it simple, don't worry about using a persistent datastore.
All endpoints are implemented and respond with proper status code. Improperly
formatted POST bodies respond with a status 400 and spec/explain
against the
expected regular expression.
To check output for step 2, run the web server with the command
lein ring server
and utilize the generated Swagger page
When you are invited to interview in person, we will sit down with you and offer feedback and guidance in refactoring your code. We would like to see you pick up these skills quickly and apply them during the interview to improve the structure of your solution.
Please check-in your project files or bring your own laptop for this part.
- "input" files can have data stored in format of my choice, e.g. date can be stored as ISO8601 so long as the display adheres to the homework spec
- favorite color and date of birth will be provided in a format that does not conflict with any chosen delimiting character
- responses won't be so large that a conversion step before response matters
- responses do not have to be paginated for size considerations
- command-line arguments to specify delimiter
- more use of spec
- allow POST to use m/d/yyyy format. Easiest to convert line->record->line
- make functions figure out delimiter instead of being passed
- could optimize how frequently a full read of all data is done
- memoize + track inserts?, i.e. in-memory store
- I think tests are writing the same entry to the datastore which is pretty annoying and something I'd like to fix