Changelog

Source: NEWS.md

readabilityes 0.1.0 (2025-08-11)

What’s new

• Sentence segmentation

  • segment_sentences(): protects abbreviations (e.g., Dr., etc.), initials (J. R. R.), decimals, and ellipses; supports extra_abbr, keep_delim, normalize_spacing.
  • count_sentences(): counts sentences per element, with na_as_zero and drop_empty.

• Tokenization and counts

  • tokenize_words(): options lowercase, keep_accents, strip_punct, keep_hyphens, remove_numbers, strip_symbols, flatten.
  • count_words(): token count per element (defensive flag normalization with isTRUE()).

• Syllabification (ES)

  • syll_split(), syll_count(), syll_hyphenate(): deterministic core with diphthongs/triphthongs and C+l/r clusters; optional cache via syll_cache_info() / syll_cache_clear().

• Average metrics

  • avg_syllables_per_word(), letters_per_word(), chars_per_word().

• Readability indexes

  • szigriszt_pazos(), inflesz() + inflesz_category(), gutierrez_de_polini().

Changes and fixes

• Segmentation regex: replaced variable-length lookbehind with a PCRE-compatible pattern to avoid lookbehind assertion is not fixed length.
• count_words() and derived functions: defensive coercion of logical flags (isTRUE()).
• Documentation updated: new strip_symbols and flatten arguments in tokenize_words().
• Expanded testthat suite and README examples in ES/EN.

Compatibility

• No known breaking changes in this initial release.