tryParse static method

TextWritingDirection? tryParse(
  1. String? value
)

Parses 'ltr' / 'rtl' (case-insensitive, whitespace-trimmed) into a TextWritingDirection; returns null for anything else — never throws.

Matching rules, pinned so config/JSON round-trips stay stable:

  • Case-insensitive: 'LTR', 'Ltr', 'rTl' all resolve.
  • Only LEADING/TRAILING whitespace is stripped (via String.trim()), so inner-space tokens like 'l t r' are rejected — there is no substring or partial matching ('ltrx', 'xrtl', 'ltr;' → null).
  • String.trim() strips Unicode whitespace too (U+00A0 NBSP, thin space, ideographic space), so ' ltr ' parses; but it does NOT strip zero-width chars (U+200B, U+FEFF BOM), so those leading a token → null.
  • The Unicode direction MARKS themselves (LRM U+200E, RLM U+200F) are not the literal tokens and so → null.
  • No normalization or case-folding beyond ASCII lowercasing: full-width 'ｌｔｒ', accented 'ĺtr', and emoji never match.
  • Lowercasing is invariant/ASCII-only here (L/T/R), so the result is independent of the active locale — no Turkish-I dotless-i pitfall.

Total function: returns null (never throws) for null, empty, whitespace-only, numeric strings, control characters, and arbitrarily long junk — safe to call on any untrusted input.

Example:

TextDirectionParseUtils.tryParse('  LTR ');  // TextWritingDirection.ltr
TextDirectionParseUtils.tryParse('auto');    // null
TextDirectionParseUtils.tryParse(null);      // null

Audited: 2026-06-12 11:26 EDT

Implementation

static TextWritingDirection? tryParse(String? value) =>
    // Total by construction: a null input maps straight to null, and the
    // switch's exhaustive default turns every unmatched token into null,
    // so no path can throw regardless of the string's content.
    switch (value == null ? null : _trim(value).toLowerCase()) {
      'rtl' => TextWritingDirection.rtl,
      'ltr' => TextWritingDirection.ltr,
      _ => null,
    };