Happy Holidays! The first stable release of get-reader, version 1.0.0 is now available. * https://pypi.org/project/get-reader/ * https://github.com/shawnbrown/get_reader This module provides a `get_reader()` function that returns reader objects similar to those returned by `csv.reader()`. This package: * reduces common boilerplate code for handling files and reading records * reads data from CSV, pandas, SQL connections, MS Excel, DBF, and squint * provides a single interface across Python versions (including seamless Unicode-aware CSV support for Python 2) * is easy to incorporate into your own projects: * has no hard dependencies * runs on Python 2.6, 2.7, 3.2 through 3.8, PyPy, PyPy3, and Jython * is freely available under the Apache License, version 2 * can be easily vendored directly into your codebase if you don't want to include it as a dependency Some examples:
from get_reader import get_reader
# CSV file. reader = get_reader('myfile.csv')
# Database connection. connection = ... reader = get_reader(connection, 'SELECT col1, col2 FROM mytable;')
# Pandas DataFrame. df = pd.DataFrame([...]) reader = get_reader(df)
# Excel file. reader = get_reader('myfile.xlsx', worksheet='Sheet2')
The internal file object is automatically closed when the iterator is exhausted, when the object is deleted, or when it is explicitly closed. An example using a context manager:
with get_reader('myfile.csv') as reader: for row in reader: print(', '.join(row))
The internal file object is also closed when exiting a with block even if the for-loop doesn't finish exhausting the reader. Install: The get-reader module has no hard dependencies; is tested on Python 2.6, 2.7, 3.2 through 3.8, PyPy, PyPy3, and Jython; and is freely available under the Apache License, version 2. You can install get_reader using pip: pip install get-reader To install optional support for MS Excel and DBF files (dBase, Foxpro, etc.), use the following: pip install get-reader[excel,dbf] Python 2 Support Statement: While official support for Python 2 ends on January 1, 2020, this project will continue to support older versions as long as the existing ecosystem provides the ability to run automated tests on those older versions.