v0.2.0

0.2.0 (2026-01-09)

Added

• CLI option -m/--max-repeat and env var RANDSTR_MAX_REPEAT to control the maximum repetition cap used by * and +.
• Library parameter randstr-max-repeat for configuring the */+ repetition cap programmatically.
• Character class enhancements:
  • Negated classes like [^abc] (complement within ASCII printable characters).
  • Common escapes inside [...]: \d \D \w \W \s \S, plus literal escapes like \] \- \\.

Changed

• tokenize-pattern was refactored to avoid repeated length calls, improving performance on longer patterns.
• parse-quantifier contracts were tightened to reflect its actual return shapes (n, (normal ...), (normal-range ...)).

Fixed

• Group alternation parsing now only splits on top-level | (doesn't break nested groups or | inside character classes).
• Unicode filtering iteration off-by-one in property generation.

Full Changelog: v0.1.1...v0.2.0