flash-installed-and-versions

flash-installed-and-versions.ipynb

flash-installed-and-versions.py

# coding: utf-8

# In[1]:

get_ipython().magic(u'pylab inline')


# In[2]:

sc.defaultParallelism


# In[19]:

q = "SELECT subsession_start_date, active_plugins FROM longitudinal"


# In[28]:

frame = sqlContext.sql(q)


# In[21]:

frame.schema


# In[22]:

frame.rdd.first()


# In[ ]:

import itertools

def map_flash_version(row):
    for ssd, plugins in itertools.izip_longest(row.subsession_start_date, row.active_plugins or [], fillvalue=[]):
        if not ssd.startswith('2016-10'):
            continue
        for plugin in plugins:
            if plugin.name == 'Shockwave Flash':
                return [plugin.version]
        return [None]
    return []
values = frame.flatMap(map_flash_version).countByValue()


# In[ ]: