pysmo

Source Code: https://github.com/pysmo/pysmo

The addition of type annotations to Python marked a significant step forward in user experience (e.g. intelligent autocompletion in modern editors) and type safety (by catching errors before code is executed). With types thus becoming more meaningful, it is worth revisiting what a type should mean to a seismologist.

Traditionally a lot of data are stored together (e.g. in a seismogram file of some format containing a time series and its metadata), only for most of that data to be ignored during processing. While it makes sense to store the data together, it is better to split them up into far simpler types when writing code for processing.

Pysmo offers simple data types for seismologists to write code with. Instead of working with one big class containing all kinds of data, psymo uses separate, narrowly defined protocol classes that are not tied to any particular file format or existing class.

Code written with pysmo types is easy to understand and maintain. Most importantly, it can often be reused in different projects, thus reducing the duplication of effort.

Pysmo itself is designed to be modular and easy to expand without interfering with the existing code base, making it straightforward to incorporate user contributions.