nx_didyoumean.nx

nx_didyoumean.nx -- SEARCH RUNG F4: THE DID-YOU-MEAN LINE, BUILT FROM THE INDEX'S OWN TERM DICTIONARY (2026-09-18). WHAT IT DECIDES. For each eligible word of a query, one question: is the word RARE in the index while a one-edit neighbour of it is COMMON? When it is, the commonest such neighbour is offered in the word's place ("Did you mean ...?"), every other byte of the query kept. It ranks, filters and rewrites nothing: the typed query is still the one searched, and the line only offers the alternative. THE CANDIDATES ARE GENERATED, NOT SEARCHED FOR. The seg's zero-hit corrector (dss_correct_raw) walks the sorted run of dictionary terms sharing the word's first byte and runs a bounded Levenshtein per entry -- the right shape for a two-edit repair, and the reason it only runs when a query finds nothing: that run is a large slice of a multi-segment dictionary. This lib asks the converse question. It writes out every string exactly ONE Damerau edit from the word -- deletion, adjacent transposition, substitution, insertion over the tokenizer's letter class -- and asks the dictionary for each one's document frequency by EXACT lookup (ss_term_dcount, the sampled-window path the ranker's idf already reads). The work is a function of the word's LENGTH and never of the dictionary's size, which is what lets it run on queries that DO find results. EVERY BOUND IS DERIVED FROM THE RESOURCE IT DESCRIBES: - THE EDIT BOUND comes from the word's length: an edit is admitted only while it leaves a strict majority of the typed characters untouched (2 x edits < length). A two-letter word is never corrected; a word of three or more letters may take the one edit this lib generates. Two-edit repairs stay where they already live, in the zero-result fallback (dss_correct), unchanged. - RARE and COMMON come from the index's own document-frequency scale, the one the ranker weighs terms by (nx_intlog idf_q10). The widest idf the index can give any term is idf(N, 1). A word is RARE when its idf exceeds half that span; a candidate is COMMON when its idf does not. That is the midpoint of the log-df scale -- df below, or at or above, sqrt(N + 1) -- found without a square root and drawn from the LIVE document count N, so it moves with the corpus. dym_mid names the value by bisecting the predicate itself, so the announced number and the decision agree. - THE FIRST BYTE IS KEPT: the zero-hit corrector's documented choice, kept for the same determinism and because an edit at position 0 is where a short rare word most often lands on an unrelated common one. - THE WORK IS BOUNDED BY THE QUERY: at most `maxwords` words (the caller passes the ranker's own term cap), each costing at most dym_lookup_bound(length) exact lookups, every one of them counted in the stats the caller publishes. DOCUMENTED IMPRECISION -- the price of document frequency alone, with no query log and no context: - a rare real word one edit from a far commoner real word draws a suggestion. The typed word's results are still served, so this error costs one line of text and never a result. - a misspelling that is itself common (df at or above the midpoint) is taken at its word. - a slip in the FIRST letter is out of scope (uantum is not offered quantum), as it is for the incumbent corrector. license_tier: ORIGINAL No hw writes (Rule 26). Read-only on the index.