Returns the text range of the word at the given offset. Characters not part of a word, such as spaces, symbols, and punctuation, have word breaks on both sides. In such cases, this method will return a text range that contains the given text position.
Word boundaries are defined more precisely in Unicode Standard Annex #29 www.unicode.org/reports/tr29/#Word_Boundaries.
Source
TextRange getWordBoundary(TextPosition position) {
assert(!_needsLayout);
List<int> indices = _paragraph.getWordBoundary(position.offset);
return new TextRange(start: indices[0], end: indices[1]);
}