Word Wrap

Level: Intermediate 30–60 min

Concepts: Strings

Solutions: C# | TypeScript | Python

---

Create a program that wraps text at a specified column width.

Requirements

1. Implement text wrapping rules:
   • Break text at word boundaries when possible
   • Break words only if they exceed the column width
   • Preserve whitespace between words
   • Handle multiple spaces and tabs
2. Handle edge cases:
   • Empty input
   • Input shorter than column width
   • Words longer than column width
   • Multiple consecutive spaces
3. Return appropriate results:
   • Wrapped text as a string
   • Error message for invalid inputs

Test Cases

Input	Column Width	Expected Output	Notes
”Hello”	5	”Hello”	No wrapping needed
”Hello World”	5	”Hello\nWorld”	Break at word boundary
”Hello World”	7	”Hello\nWorld”	Break at word boundary
”Hello World”	20	”Hello World”	No wrapping needed
”Supercalifragilisticexpialidocious”	10	”Supercalif\nragilisti\ncexpiali\ndocious”	Break long word
”Hello World”	5	”Hello\nWorld”	Handle multiple spaces

Edge Cases to Consider

• Empty string
• Null input
• Zero or negative column width
• Very long words
• Multiple consecutive spaces or tabs
• Text with line breaks

Tips

• Start with simple cases (no wrapping needed)
• Add word boundary breaking next
• Then handle long words
• Finally add whitespace handling
• Consider using regular expressions for word boundaries

Reference Walkthrough

Reference implementations in C#, TypeScript, and Python live at tddbuddy-reference-katas/word-wrap. This is an F1 kata — twelve scenarios covering empty and whitespace-only input, single-word lines, greedy word-boundary packing, oversize-word hard splitting, oversize-remainder joining with the next word, and whitespace collapsing — shared across all three languages, each a single pure function wrap(text, width) returning a \n-joined string.

• C# (.NET 8, xUnit, FluentAssertions 6.12.0) — walkthrough
• TypeScript (Node 20, Vitest 1.6, TS 5 strict) — walkthrough
• Python (3.11, pytest) — walkthrough

This kata ships in Agent Full-Bake mode at high gear: the rules land as one commit per language, with a brief walkthrough noting there are no builders because the inputs and outputs are the domain. A greedy line-packer tokenizes on runs of whitespace, accumulates words until the next word would push the line past width, splits any oversize word hard at the width-th character before letting its remainder begin the next line (where following words may still join it if they fit), and joins the resulting lines with \n — matching the classic Kent Beck kata shape that returns a single string rather than a list. See the repo’s Gears section for when high gear is the right call.