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

# In[7]:


import pandas as pd

git_log = pd.read_csv(
    "../../../spring-petclinic/git_knowledge.log",
    names=['additions', 'deletions', 'filename', 'author'],
    sep='\t')
git_log['author'] = git_log['author'].ffill()
git_log = git_log.dropna()
git_log.head()


# In[30]:


git_log[git_log['filename'] == ".classpath"]


# In[40]:


git_log[git_log['filename'] == ".bowerrc"]


# In[53]:


get_ipython().run_line_magic('pinfo', 'pd.DataFrame.var')


# In[90]:


git_log['aid'] = pd.factorize(git_log.author)[0]
gs = git_log.groupby('filename')['author']
# pd.DataFrame(gs.ndim() / gs.count())
gs.


# In[70]:


git_log[git_log['filename'] == ".bowerrc"]


# In[79]:


git_log[git_log['filename'] == ".editorconfig"]


# In[42]:


git_log[git_log['filename'] == ".gitignore"]


# In[38]:


gb = git_log.groupby(['filename', 'author']).count()
gb.groupby(gb.index.get_level_values(0)).var()


# In[18]:


ownership = git_log.groupby(['filename', 'author']).count()
ownership.groupby(
    [ownership.index.get_level_values(0)]).sum()