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

# # DataFrame Conversions
# 
# This notebook demonstrates how to [convert](https://nexus-forge.readthedocs.io/en/latest/interaction.html#converting) a [Resource](https://nexus-forge.readthedocs.io/en/latest/interaction.html#resource) to pandas DataFrame and vice-versa.

# In[1]:


from kgforge.core import KnowledgeGraphForge


# A configuration file is needed in order to create a KnowledgeGraphForge session. A configuration can be generated using the notebook [00-Initialization.ipynb](00%20-%20Initialization.ipynb).

# In[2]:


forge = KnowledgeGraphForge("../../configurations/forge.yml")


# ## Imports

# In[3]:


import pandas as pd
import numpy as np


# In[4]:


from kgforge.core import Resource


# ## List of Resources to DataFrame

# In[5]:


address = Resource(type="PostalAddress", country="Switzerland", locality="Geneva")


# In[6]:


jane = Resource(type="Person", name="Jane Doe", address=address, email="(missing)")


# In[7]:


john = Resource(type="Person", name="John Smith", email="john.smith@epfl.ch")


# In[8]:


persons = [jane, john]


# In[9]:


forge.register(persons)


# In[10]:


forge.as_json(jane)


# In[11]:


forge.as_json(john)


# In[12]:


john._store_metadata


# In[13]:


forge.as_dataframe(persons)


# It is possible to specify what values (here '(missing)') should be replaced by `NaN` using the `na` parameter.

# In[14]:


forge.as_dataframe(persons, na="(missing)")


# It is possible to specify a string to use in the column names to show nested values, the default is dot `.`.

# In[15]:


forge.as_dataframe(persons, nesting="__")


# The `expanded` parameter will show fields and values according to the JSON-LD context.

# In[16]:


forge.as_dataframe(persons, expanded=True)


# In[17]:


forge.as_dataframe(persons, store_metadata=True)


# ## DataFrame to list of Resources

# In[20]:


data = pd.DataFrame([
    {
        "type": "Person",
        "address.type": "PostalAddress",
        "address.country": "Switzerland",
        "address.locality": "Geneva",
        "email": "(missing)",
        "name": "Jane Doe",
    },
    {
        "type": "Person",
        "address.type": np.nan,
        "address.country": np.nan,
        "address.locality": np.nan,
        "email": "john.smith@epfl.ch",
        "name": "John Smith",
    }
])


# In[21]:


data


# In[22]:


resources = forge.from_dataframe(data)


# In[23]:


address = Resource(type="PostalAddress", country="Switzerland", locality="Geneva")


# In[24]:


jane = Resource(type="Person", name="Jane Doe", address=address, email="(missing)")


# In[25]:


john = Resource(type="Person", name="John Smith", email="john.smith@epfl.ch")


# In[26]:


persons = [jane, john]


# In[27]:


resources == persons


# In[28]:


resources_na = forge.from_dataframe(data, na="(missing)")


# In[29]:


print(resources[0])


# In[30]:


print(resources_na[0])


# In[31]:


resources_nesting = forge.from_dataframe(data, nesting=".")


# In[32]:


print(resources_nesting[0])


# In[33]:


data = pd.DataFrame([
    {
        "type": "Person",
        "address_type": "PostalAddress",
        "address_country": "Switzerland",
        "address_locality": "Geneva",
        "email": "(missing)",
        "name": "Jane Doe",
    },
    {
        "type": "Person",
        "address_type": np.nan,
        "address_country": np.nan,
        "address_locality": np.nan,
        "email": "john.smith@epfl.ch",
        "name": "John Smith",
    }
])


# In[34]:


resources_nesting = forge.from_dataframe(data, nesting="_")


# In[35]:


print(resources_nesting[0])