#!/usr/bin/env python
# coding: utf-8

# ### Fennec analysis

# In[1]:


import ujson as json
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import plotly.plotly as py

from moztelemetry import get_pings, get_pings_properties, get_one_ping_per_client

get_ipython().run_line_magic('pylab', 'inline')


# In[79]:


pings = get_pings(sc, app="Fennec", channel="release", submission_date="20150609", fraction=1)


# In[80]:


subset = get_pings_properties(pings, ["clientID", "info/manufacturer", "info/device", "info/version"])


# In[81]:


subset.filter(lambda x: x.get("clientID", None) is not None).count() # ClientID isn't present in release


# In[82]:


samsung = pd.DataFrame(subset.filter(lambda x: x["info/manufacturer"] == "samsung" and x["info/device"] == "GT-I9500").collect())
len(samsung)


# In[83]:


histogram = samsung.groupby("info/version").apply(lambda x: len(x))


# In[91]:


(100*histogram/histogram.sum()).plot(kind="bar")
pylab.ylabel("submissions %")


# In[ ]: