rimhistory Files

New features

• Use data from a series of save files to produce time-series datasets and visualizations
• Each dataset is extracted from a series of save files and stored in pandas DataFrames internally
• The in-game time is extracted from each save file and added to the data model for use as a time dimension
• Null handling logic passes a None value into the pandas DataFrame when data is missing, such as missing plant age, pawn ambient temperature, or pawn first name

Improvements

• Reduce memory usage by deleting the XML root element as soon as possible after instantiating a Save object
• Bump pandas to 1.4.1
• Bump pytest to 7.0.1
• Configure flake8 errors to halt the GitHub Actions build process
• Clean up code style issues raised by flake8

Notes

Example chart generated from 167 save files:

Performance for large file sets:
During testing, the latest version was able to process 167 sequential save files in 124 seconds. This allows users with a large number of saves to load them all without too long of a wait. This could be sped up further by refactoring the SaveSeries class to initialize its Save objects asynchronously.