Python Q & A

 

How to convert Python 2 code to Python 3?

Converting Python 2 code to Python 3 is a critical step for maintaining the longevity and compatibility of your code, especially since Python 2 reached its end of life on January 1, 2020. Transitioning between the two versions can be non-trivial due to several differences, but there are tools and best practices to streamline the process.

 

  1. Preparation:

Before diving into conversion, ensure your Python 2 codebase has a good suite of unit tests. This ensures that you can verify the correctness of the converted code. If you’re lacking tests, now is the time to write them.

 

  1. `2to3` Tool:

Python offers a tool named `2to3` that can automatically translate Python 2 code to Python 3. It can be run on single files or entire directories, and it can make most of the code translations needed:

```bash

$ 2to3 -w your_project_directory/

```

 

The `-w` flag tells `2to3` to write the changes back to the files. It’s a good idea to have version control in place before using it, so you can review the changes it makes.

 

  1. Manual Adjustments:

While `2to3` is effective, there will likely be some changes it cannot handle. Some notable manual adjustments include:

– Unicode: In Python 2, strings can be either Unicode or bytes, while in Python 3, they are distinct types. Ensure you handle strings properly, especially when reading from or writing to files or working with libraries that expect byte strings. 

– Division: In Python 2, `/` performs integer division if both operands are integers. In Python 3, it performs floating-point division. If you want integer division in Python 3, use `//`.

– Exception syntax: Python 3 no longer supports the `,` syntax for exceptions, e.g., `except Exception, e:`. This should be changed to `except Exception as e:`.

 

  1. Libraries and Dependencies:

Check that third-party libraries your project relies on are compatible with Python 3 or if there are newer versions available that are.

The journey from Python 2 to Python 3 requires both automated tools and some manual intervention. Given the widespread adoption of Python 3 and the cessation of support for Python 2, it’s a worthy investment to update your codebase. Always keep your code under version control and rely on tests to catch potential migration issues.

Previously at
Flag Argentina
Brazil
time icon
GMT-3
Senior Software Engineer with 7+ yrs Python experience. Improved Kafka-S3 ingestion, GCP Pub/Sub metrics. Proficient in Flask, FastAPI, AWS, GCP, Kafka, Git