/incisive

A simple library for CSV files

Primary LanguagePythonMIT LicenseMIT

incisive

incisive is a tiny library for handling CSV in Python. It's a wrapper for the csv module.

Usage

  • You can read a csv file like this:

    >>> iris = read_csv("data/iris.csv")
>>> iris.next()
{'petal_length': 1.4,
 'petal_width': 0.2,
 'sepal_length': 5.1,
 'sepal_width': 3.5,
 'species': 'setosa'}

read_csv returns a generator. Note that incisive, by default, tries to guess the type of the columns.

  • Writing a CSV file can be done in two ways:

  1. with a list of dictionaries:

    >>> data = [{'name': 'Lancelot', 'actor': 'John Cleese', 'color': 'blue'},
           {'name': 'Galahad', 'actor': 'Michael Palin', 'color': 'yellow'}]

>>> write_csv('bridge.csv', ('name', 'actor', 'color'), data=data)

(the keys are the field names of the CSV file.)

  1. or with just a list of rows:

    >>> rows = [('Lancelot', 'John Cleese', 'blue'),
        ('Galahad', 'Michael Palin', 'yellow')]

>>> write_csv('bridge.csv', ('name', 'actor', 'color'), rows=rows)

Note that this second method requires that your field names correspond exactly to the elements of your rows.

Features

  • Simple API
  • Guess the types when it's possible
  • Accept functions to specify the types