Commit graph

3 commits

Author SHA1 Message Date
Yogthos
e75771084e fix: merge nil/empty + conj semantics to match Clojure
core-merge returned {} for (merge) and errored on nil args. Rewrite to follow
Clojure: (when (some identity maps) (reduce conj (or (first maps) {}) (rest maps))).

- (merge) and (merge nil nil) -> nil
- nil args elsewhere are no-ops: (merge {} nil) -> {}, (merge nil {:a 1}) -> {:a 1}
- later args follow conj semantics: a map merges its entries, a [k v]
  vector/map-entry adds that entry ((merge {} [:foo 1]) -> {:foo 1},
  (merge {} (first {:a 1})) -> {:a 1})
- collection keys preserved (first arg's phm type kept; entries via core-assoc
  which promotes struct->phm for collection keys)

spec: 9 new merge cases in maps-spec. merge.cljc 13/18/11 -> 29/11/2.
clojure-test-suite pass 1688->1704. jpm test green.

Remaining merge.cljc fails are the lenient-where-Clojure-throws class (atomic
args, >2-element vectors). Fixes jolt-dxx.
2026-06-05 09:22:27 -04:00
Yogthos
0db7eb6ac8 fix: value-semantics for collection keys/elements; set literals evaluate
Close jolt-do7. Maps/sets keyed by a collection (a map, vector, ...) now compare
by value instead of Janet identity:
- PHM/PHS hash and compare keys through an injected canonicalizer (collection
  keys -> value-hashable struct/tuple); keys are still stored as-is
- map literals and core-assoc promote to a phm when a key is a collection
- frequencies/group-by use a phm base so collection elements/keys dedup by value
- set equality is value-based (from earlier)

Real bugs found and fixed along the way:
- set literals #{(inc 1)} did not evaluate their elements (stored raw forms!)
- the REPL printer rendered phm maps as {} (they hit the deftype branch); now a
  phm branch prints entries

Added spec cases (maps/collection-keys, sets/literals & value elements).
conformance 218/218, jpm test green.
2026-06-05 01:02:00 -04:00
Yogthos
b353d625f1 test(spec): collections contract (sequences, vectors, lists, maps, sets)
Add behavioral spec suites for the collection API (~180 cases). Writing them
surfaced and fixed real bugs:
- (count nil) errored; now 0 (Clojure semantics)
- (repeat x) and (repeatedly f) — the infinite 1-arg arities — were unsupported;
  now return lazy infinite seqs
- set equality used deep= (representation-sensitive), so a set of map literals
  could differ from an equal set built differently; now value-based (each
  element value-equal to some element of the other)

Known limitation filed: phm-typed maps (from hash-map) used as set elements /
map keys hash by identity (map *literals* are structs and work).

jpm test green.
2026-06-05 00:08:37 -04:00