pythonic-df

Making a native python DataFrame library from scratch, just for fun!

Usage

Installation

You can install this package via pip:

pip install https://github.com/guillesd/pythonic-df.git

Creating a DataFrame

The most basic way of doing this is by defining Columns and passing on a List of this to the DataFrame constructor, i.e.:

from pythonic_df.data_structures import Column, DataFrame

col_a = Column(
    name = "Numeros",
    data_type = int,
    values = [1, 2, 4]
)

col_b = Column(
    name = "Letras",
    data_type = str,
    values = ["a", "b", "d"]
)

col_c = Column(
    name = "Boleano",
    data_type = bool,
    values = [True, False, True]
)

df = DataFrame([col_a, col_b, col_c])

Of course, this is not something that you usually would do. The most convenient way of creating a DataFrame would be from a dictionary...

from pythonic_df.utils import dataframe_from_dict
pretty_print(dataframe_from_dict({'col_a': [1, 2], 'col_b': [2, 4]}))

... or by reading a file in, let's say, .csv format:

from pythonic_df.io.csv import read_csv

df = read_csv(
    file_path="./data/example.csv",
    delimiter=",",
    has_headers=True,
    data_types={"col_a": int, "col_b": str}
)

Displaying your DataFrame

By default, the DataFrame will display as an instance of a python dataclass. Meaning if you just type print(df) the output would be something like:

[out]: DataFrame(columns=[Column(name='col_a', data_type=<class 'int'>, values=['1', '2', '3'], length=3), Column(name='col_b', data_type=<class 'str'>, values=['a', 'b', 'c'], length=3)], index=[0, 1, 2])

This is obviously not a nice way of visualizing your structured data. The better approach would be to use a special printer provided by pythonic-df, like so...

from pythonic_df.printer import pretty_print
pretty_print(df)

...which will print out the following:

[out]:

      index    col_a  col_b
    -------  -------  -------
          0        1  a
          1        2  b
          2        3  c

DataFrame operations

This module intends to provide an interface to operate with DataFrame structures. Examples of this operations would be:

• Append
• Join
• Group by
• Windowing
• User defined functions
• etc.

An example of how to interact with this module to append two DataFrames:

from pythonic_df.operations.dataframe_operations import append

df1 = dataframe_from_dict({'col_a': [1, 2], 'col_b': [2, 4]})
df2 = dataframe_from_dict({'col_a': [7, 8], 'col_b': [9, 11]})
appended_df = append(df1, df2)
pretty_print(appended_df)

[out]:

      index    col_a    col_b
    -------  -------  -------
          0        1        2
          1        2        4
          2        7        9
          3        8       11

You can also join two DataFrames, like so:

from pythonic_df.operations.dataframe_operations import join

df1 = dataframe_from_dict({'col_a': [1, 2, 1], 'col_b': [2, 4, 2]})
df2 = dataframe_from_dict({'col_a': [1, 8], 'col_c': [9, 11], 'col_d': [True, False]})

pretty_print(
    join(
        df1=df1, 
        df2=df2, 
        key1='col_a', 
        key2='col_a',
        type='inner'
    )
)

[out]:

  index    col_a    col_b    col_a_2    col_c  col_d
-------  -------  -------  ---------  -------  -------
      0        1        2          1        9  True
      1        1        2          1        9  True