cryptRopen: Industrialized Pseudonymization of Variables Across Datasets

Deterministic, salted-hash pseudonymization of one or many datasets, driven by a single Excel mask that lists, per input file, which columns to hash and which to drop.

Where to start

Three entry points cover the typical needs:

• crypt_vector() — hash one vector.

• crypt_data() — pseudonymize one data frame already in memory; the in-session workflow.

• crypt_r() — pseudonymize N files described by an Excel mask, in parallel, non-blocking; the industrialized workflow. Inspect / wait / finalize the returned job with cryptR_status(), cryptR_wait(), cryptR_collect(), cryptR_results(), and summary().

Use get_correspondence_tables() to retrieve the (original, hashed) mapping stored after crypt_data() / crypt_r() runs.

Learn more

• vignette("cryptRopen") — getting started.

• vignette("crypt_r-workflow") — industrialized walkthrough.

• NEWS.md — change history.

See also

Useful links:

• https://danielrak.github.io/cryptRopen/

• https://github.com/danielrak/cryptRopen

• Report bugs at https://github.com/danielrak/cryptRopen/issues

Author

Maintainer: Daniel Rakotomalala rakdanielh@gmail.com