Write a python script/module/package, that for a given flight data in a form of csv
file (check the examples), prints out a structured list of all flight combinations for a selected route between airports A -> B, sorted by the final price for the trip.
You've been provided with some semi-randomly generated example csv datasets you can use to test your solution. The datasets have following columns:
flight_no
: Flight number.origin
,destination
: Airport codes.departure
,arrival
: Dates and times of the departures/arrivals.base_price
,bag_price
: Prices of the ticket and one piece of baggage.bags_allowed
: Number of allowed pieces of baggage for the flight.
In addition to the dataset, your script will take some additional arguments as input:
Argument name | type | Description | Notes |
---|---|---|---|
origin |
string | Origin airport code | |
destination |
string | Destination airport code |
- By default you're performing search on ALL available combinations, according to search parameters.
- In case of a combination of A -> B -> C, the layover time in B should not be less than 1 hour and more than 6 hours.
- No repeating airports in the same trip!
- A -> B -> A -> C is not a valid combination for search A -> C.
- Output is sorted by the final price of the trip.
You may add any number of additional search parameters to boost your chances to attend. Here are 2 recommended ones:
Argument name | type | Description | Notes |
---|---|---|---|
bags |
integer | Number of requested bags | Optional (defaults to 0) |
return |
boolean | Is it a return flight? | Optional (defaults to false) |
Example input (assuming solution.py
is the main module):
python -m solution example/example0.csv RFZ WIW --bags=1 --return
will perform a search RFZ -> WIW -> RFZ for flights which allow at least 1 piece of baggage.
- NOTE: Since WIW is in this case the final destination for one part of the trip, the layover rule does not apply.
The output will be a json-compatible structured list of trips sorted by price. The trip has the following schema:
Field | Description |
---|---|
flights |
A list of flights in the trip according to the input dataset. |
origin |
Origin airport of the trip. |
destination |
The final destination of the trip. |
bags_allowed |
The number of allowed bags for the trip. |
bags_count |
The searched number of bags. |
total_price |
The total price for the trip. |
travel_time |
The total travel time. |
For more information, check the example section.
Assuming your solution is working, we'll be additionally judging based on following skills:
- input, output - what if we input garbage?
- modules, packages & code structure (hint: it's easy to overdo it)
- usage of standard library and built-in data structures
- code readability, clarity, used conventions, documentation and comments
- Your solution needs to contain a README file describing what it does and how to run it.
- Only the standard library is allowed, no 3rd party packages, notebooks, specialized distros (Conda) etc.
- The code should run as is, no environment setup should be required.
Follow the instructions you received in the email.
Let's imagine we wrote our solution into one file, solution.py
and our datatset is in data.csv
.
We want to test the script by performing a flight search on route BTW -> REJ (we know the airports are present in the dataset) with one bag. We run the thing:
python -m solution data.csv BTW REJ --bags=1
and get the following result:
[
{
"flights": [
{
"flight_no": "XC233",
"origin": "BTW",
"destination": "WTF",
"departure": "2021-09-02T05:50:00",
"arrival": "2021-09-02T8:20:00",
"base_price": 67.0,
"bag_price": 7.0,
"bags_allowed": 2
},
{
"flight_no": "VJ832",
"origin": "WTF",
"destination": "REJ",
"departure": "2021-09-02T11:05:00",
"arrival": "2021-09-02T12:45:00",
"base_price": 31.0,
"bag_price": 5.0,
"bags_allowed": 1
}
],
"bags_allowed": 1,
"bags_count": 1,
"destination": "REJ",
"origin": "BTW",
"total_price": 110.0,
"travel_time": "6:55:00"
},
{
"flights": [
{
"flight_no": "JV042",
"origin": "BTW",
"destination": "REJ",
"departure": "2021-09-01T17:35:00",
"arrival": "2021-09-01T21:05:00",
"base_price": 216.0,
"bag_price": 11.0,
"bags_allowed": 2
}
],
"bags_allowed": 2,
"bags_count": 1,
"destination": "REJ",
"origin": "BTW",
"total_price": 227.0,
"travel_time": "3:30:00"
}
]