df_string 0.4.1

A package designed to extend Dart strings by offering additional functionality.

Changelog

0.4.1

• Released @ 10/2026 (UTC)
• fix: case conversions (toSnakeCase, toKebabCase, toDotCase, toPathCase, toCamelCase, toPascalCase, toCapitalizedKebabCase and their upper/lower variants) keep letters, accent marks and digits of every script instead of deleting every non-ASCII character: café_au_lait → café_au_lait (was caf_au_lait), 日本語 → 日本語 (was empty), überCool → über_cool. Output for ASCII input is byte-identical to 0.4.0 (pinned by a golden table and a fuzz test against the 0.4.0 implementation). Emoji, symbols and punctuation are still delimiters.
• fix: capitalize, withFirstLetterAsUpperCase and withFirstLetterAsLowerCase convert a first letter outside the Basic Multilingual Plane instead of leaving it unchanged.
• fix: replaceLast with a String pattern replaces the true last occurrence even when it overlaps an earlier one: 'aaa'.replaceLast('aa', 'X') is 'aX' (was 'Xa'). RegExp patterns still replace the last non-overlapping match.
• fix: truncToLength counts runes and never splits a surrogate pair: '😀😀'.truncToLength(1) is '😀' (was half an emoji). Behavioural change: for text with characters outside the BMP (e.g. emoji), length now counts each such character once rather than as two UTF-16 code units. Output for BMP-only text is unchanged.

0.4.0

• Released @ 8/2026 (UTC)
• fix: case conversions (toSnakeCase, toKebabCase, toDotCase, toCamelCase, toPascalCase, toPathCase, and their variants) no longer split at letter↔digit boundaries — digits now stay attached to the letter run they touch, matching the Rails underscore convention. phoneE164 → phone_e164 (was phone_e_164), line1 → line1 (was line_1), version1 → version1 (was version_1). A digit immediately before an uppercase letter is still a word boundary (foo1Bar → foo1_bar), so camel/snake round-trips stay stable. This keeps generated wire keys aligned with database columns that embed digits. Behavioural change: consumers that relied on the old digit-splitting output will see different results.

0.3.0

• Released @ 5/2026 (UTC)
• fix: truncToLength now appends ellipsis only when truncation actually occurs, returns short inputs unchanged, preserves leading whitespace (only trailing whitespace exposed by the cut is trimmed), and throws RangeError on negative length
• fix: forceObjToString is guaranteed not to throw even when both toString and hashCode misbehave; falls back to '<unrepresentable object>' in that case

0.2.10

• Released @ 12/2025 (UTC)
• Add an UNCHANGED StringCaseType

0.2.9

• Released @ 6/2025 (UTC)
• Update dependencies

0.2.8

• Released @ 6/2025 (UTC)
• Update dependencies

0.2.7

• Released @ 6/2025 (UTC)
• Refactor and clean

0.2.6

• Released @ 6/2025 (UTC)
• Update dependencies

0.2.5

• Released @ 6/2025 (UTC)
• Make string type conversions more robust

0.2.4

• Released @ 12/2024 (UTC)
• refactor: Improve code structure

0.2.3

• Released @ 12/2024 (UTC)
• chore: Update CI/CD script

0.2.2

• Released @ 12/2024 (UTC)
• docs: Update readme and comments

0.2.1

• Released @ 12/2024 (UTC)
• chore: Update topics in pubspec.yaml, docs and comments

0.2.0

• Released @ 12/2024 (UTC)
• chore: Update docs and format code

0.1.2

• Released @ 12/2024 (UTC)
• feat: Add extension NullIfEmptyOnStringX

0.1.1

• Released @ 12/2024 (UTC)
• chore: Update workflow scripts

0.1.0

• Released @ 12/2024 (UTC)
• Initial commit