New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Schema evolution #817

Closed

martindurant opened this issue Nov 2, 2022 · 0 comments · Fixed by #818

Member

martindurant commented Nov 2, 2022

We can enable schema evolution in fastparquet as follows:

take the schema definition from the last file in a dataset; or otherwise allow upcasting while scanning a dataset
any missing fields in some data file will become nulls
allowed transitions will need conversion (just like the current conversion code does - would amount to an astype)
allow setting the schema of ParquetFile (in init, or a special method) to define the dtypes it will try to fill.

The driver for this idea is the ability to adapt to Iceberg datasets.

martindurant mentioned this issue

Merged

martindurant closed this as completed in #818

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment