Word Frequency

Define a word before counting: ask how to handle case, punctuation, apostrophes, hyphens, Unicode, stop words, and ties. Normalize according to that contract, tokenize with an appropriate parser or delimiter rule, and increment a hash-map count for each token. If only the map is required, runtime is O(n) in input characters and O(u) space for u unique words. To produce a ranked list, sort entries by count descending and lexical order for deterministic ties, O(u log u), or use a size-k heap for top k. Avoid assuming a shell command's whitespace behavior is a complete tokenizer. Test repeated punctuation, mixed case, empty input, non-ASCII text, and equal-frequency terms. Explain whether the result counts raw tokens or normalized words.