Release 0.3.5

What's Changed

• Fix last special printable characters by @qubixes in #13

Full Changelog: v0.3.4...v0.3.5

Release 0.3.4

Fixes a bug that broke deserialization of the characters '~' and '\r'.

What's Changed

• Add contact information and logo by @qubixes in #10
• Add two more special characters to list by @qubixes in #11
• Make pylint happy by @qubixes in #12

Full Changelog: v0.3.3...v0.3.4

Release 0.3.3

Fix deprecation warnings with polars.

Full Changelog: v0.3.2...v0.3.3

Release 0.3.2

Small bugfix release related to parsing regexes.

What's Changed

• Add cheatsheet by @qubixes in #9

Full Changelog: v0.3.1...v0.3.2

Release 0.3.1

What's Changed

• Improve readibility of formulas by @qubixes in #5
• Improve regex parsing and output by @qubixes in #6
• Fix issue where long strings would result in error by @qubixes in #7

Full Changelog: v0.3.0...v0.3.1

Release 0.3.0

What's Changed

The regexmodel now is able to parse and estimate regex class ranges such as [A-F], [g-s], [4-7]. This should also result in small improvements in performance.

• Allow regex ranges [A-C], etc by @qubixes in #4

Full Changelog: v0.2.1...v0.3.0

Release 0.2.1

What's Changed

• Add not-fitted error and fix count_thres issues by @qubixes in #3

Full Changelog: v0.2.0...v0.2.1

Release 0.2.0

This is basically a rewrite of the functionality, while keeping most of the interface the same. It is expected that the new code structure will stay and that the old one will not come back.

New highlights:

• model.regex will give the regex that the model has found.
• Extended regex language: ([a]|[b]) is now technically supported.
• There is now an "accurate" and "fast" method to determine how fast the regex model will fit. The "fast" method is multiple times faster than the previous implementation. The "accurate" method is slightly slower than the old method in some cases.

Release 0.1.0

First release.