cube query

version = 1.0.3

introduction

This is a simple way to query data cubes on spark.

The only thing you should do is to create a cube.ini file under conf.

Supports parquet, csv, and json.

query

Choose your target data by clicking on buttons under "table name". (see feature.1)

Key in SQL to query tables, but please note that now we only support one table query at a time.

plot

Select column for x-axis and y-axis. You may group by x-column then apply aggregate functions on y-columns; Note that you may choose multi y-columns.

You may experiment on various plot types that we offer.

By choosing a cohort column, it would first group by cohort column, then generate a graph for each cohort. Remember: you can choose only a set of x-column and y-column for better comparison between cohorts.

quick start

    $ git clone https://github.com/KuanTingLin/cube_query.git
    $ cd cube_query/project
    $ python app.py

Now go to http://localhost:5000

set host and port

    $ python app.py --host 127.0.0.1 -p 5000

check arguments

    $ python app.py -h

test new branch3