https://github.com/GeneDer/coding-challenge
Pure Python3 solution to Insight Coding Challenge, with updated ./data-gen/get-tweets.py for Python3 compliance.
Structures used to solve problem are described in ./src/challenge/challibs.py. challibs.TimeGraph is a structure that mantains a sliding window of time for edges of a graph. This will not produce useful results against edges with timestamps occuring before the Unix Epock.
challibs.clean_tweets provides a data ellis converting raw tweets in json into simple dictionaries with only critical information.
./run.sh wraps ./src/challenge-runner.py, the executable, python, CLI to the program.
always-reset-time : list of parsable tweets that are all 60 seconds appart
empty : process over empty file
gradual-reset-time : list of 120 parsable tweets seperated by a second
malformed-input : list of garbage input, treated same as limit-json entries
never-reset-time : list of 60 parsable tweets seperated by a second
one-tweet : simplified single tweet
time-without-edge : update graph with new tweet that has no edge
Large test was not included, due to size. Large test had
- 306011 lines of text pulled using
./data/get-tweets.py(874M) - contained 16484 limits, which registered as malformed input
- produced 289527 lines of text in output in 5m27.199s with trivial memory profile
.
├── data-gen
│ ├── get-tweets.py
│ ├── README.md
│ └── tweets.txt
├── insight_testsuite
│ ├── results.txt
│ ├── run_tests.sh
│ └── tests
│ ├── always-reset-time
│ │ └── ...
│ ├── empty
│ │ └── ...
│ ├── gradual-reset-time
│ │ └── ...
│ ├── malformed-input
│ │ └── ...
│ ├── never-reset-time
│ │ └── ...
│ ├── one-tweet
│ │ └── ...
│ ├── test-2-tweets-all-distinct
│ │ └── ...
│ └── time-without-edge
│ └── ...
├── README.md
├── run.sh
├── src
│ ├── challenge
│ │ ├── challibs.py
│ │ └── tweetprocess.py
│ └── challenge-runner.py
├── tweet_input
│ └── tweets.txt -> ../data-gen/tweets.txt
└── tweet_output
└── output.txt