gitHistoryGrapher.py

import numpy as np
import pandas as pd
import inspect
import os
import datetime as dt
import pylab
import subprocess
import tempfile

import plotly.plotly as py
import plotly.graph_objs as go
import urllib.request
import numpy as np
import plotly
import json

import boto3


import re

plotly.tools.set_credentials_file(username='*****', api_key='******')

try:
    from itertools import izip
except ImportError:
    izip = zip

columns = ['instance_type', 'pricing']
region = 'us-east-1'



def execute(cmd):
    return subprocess.check_output(cmd.split())

frame = inspect.getfile(inspect.currentframe())
path = os.path.dirname(os.path.abspath(frame))
os.chdir("/Users/NathanTrujillo/Documents/gits/ec2instances.info")

checkout = "git checkout {}"

# get git history for file
subprocess.check_output(r"git log --all --pretty=format:'%h,%cd' -- www/instances.json > instanceCommitHistory.csv".split())
df = pd.read_csv('instanceCommitHistory.csv', delimiter=',')
df.columns = ['hash', 'date']
# make string date be date type
df['date'] = pd.to_datetime(df['date'])
history = {}
hashes = []
# now iterate through all hashes, and store the data in memory
for index, row in df.iterrows():
    print(row['hash'])
    date = pd.to_datetime(row['date'])
    hashes.append(date)
    execute(checkout.format(row['hash']))
    data = pd.read_json('www/instances.json')
    data = data[columns]
    data['pricing'] = data['pricing'].to_frame()
    #data['pricing'] = data['pricing'][region]
    history[date] = data

mega = pd.concat(history)
history[hashes[0]]
# extract region
slot = history[hashes[0]]['pricing'].to_frame()
slot['pricing'] = slot['pricing'].to_frame()
slot
#print(slot['pricing'][0][region])
mega.to_json('history.json')