16 Commits

Author SHA1 Message Date
cd489b19be haskell: do-notation desugar + stub IO monad (return/>>=/>>) (+14 tests, 382/382)
Some checks are pending
Test, Build, and Deploy / test-build-deploy (push) Waiting to run
2026-04-25 00:59:42 +00:00
04a25d17d0 haskell: seq + deepseq via lazy-builtin flag (+9 tests, 368/368)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-25 00:28:19 +00:00
cc5315a5e6 haskell: lazy : + ranges + Prelude (repeat/iterate/fibs/take, +25 tests, 359/359)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 23:58:21 +00:00
0e53e88b02 haskell: thunks + force, app args become lazy (+6 tests, 333/333)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 23:22:21 +00:00
fba92c2b69 haskell: strict evaluator + 38 eval tests, Phase 2 complete (329/329)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 22:49:12 +00:00
1aa06237f1 haskell: value-level pattern matcher (+31 tests, 281/281)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 22:15:13 +00:00
e9c8f803b5 haskell: runtime constructor registry (+24 tests, 250/250)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 21:45:51 +00:00
ef81fffb6f haskell: desugar guards/where/list-comp → core AST (+15 tests, 226/226)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 21:16:53 +00:00
cab7ca883f haskell: operator sections + list comprehensions, Phase 1 parser complete (+22 tests, 211/211)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 20:47:51 +00:00
bf0d72fd2f haskell: module header + imports (+16 tests, 189/189)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 20:08:30 +00:00
defbe0a612 haskell: guards + where clauses (+11 tests, 173/173)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 19:37:52 +00:00
869b0b552d haskell: top-level decls (fn-clause, type-sig, data, type, newtype, fixity) + type parser (+24 tests, 162/162)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 19:06:38 +00:00
58dbbc5d8b haskell: full patterns — as/lazy/negative/infix + lambda & let pat LHS (+18 tests, 138/138)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 18:34:47 +00:00
36234f0132 haskell: case/do + minimal patterns (+19 tests, 119/119)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 18:00:58 +00:00
6ccef45ce4 haskell: expression parser + precedence climbing (+42 tests, 100/100)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 17:31:38 +00:00
c07ff90f6b haskell: layout rule per §10.3 (+15 tests, 58/58)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Has been cancelled
2026-04-24 17:05:35 +00:00
37 changed files with 7403 additions and 3166 deletions

249
lib/haskell/desugar.sx Normal file
View File

@@ -0,0 +1,249 @@
;; Desugar the Haskell surface AST into a smaller core AST.
;;
;; Eliminates the three surface-only shapes produced by the parser:
;; :where BODY DECLS → :let DECLS BODY
;; :guarded GUARDS → :if C1 E1 (:if C2 E2 … (:app error …))
;; :list-comp EXPR QUALS → concatMap-based expression (§3.11)
;;
;; Everything else (:app, :op, :lambda, :let, :case, :do, :tuple,
;; :list, :range, :if, :neg, :sect-left / :sect-right, plus all
;; leaf forms and pattern / type nodes) is passed through after
;; recursing into children.
(define
hk-guards-to-if
(fn
(guards)
(cond
((empty? guards)
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards")))
(:else
(let
((g (first guards)))
(list
:if
(hk-desugar (nth g 1))
(hk-desugar (nth g 2))
(hk-guards-to-if (rest guards))))))))
;; do-notation desugaring (Haskell 98 §3.14):
;; do { e } = e
;; do { e ; ss } = e >> do { ss }
;; do { p <- e ; ss } = e >>= \p -> do { ss }
;; do { let decls ; ss } = let decls in do { ss }
(define
hk-desugar-do
(fn
(stmts)
(cond
((empty? stmts) (raise "empty do block"))
((empty? (rest stmts))
(let ((s (first stmts)))
(cond
((= (first s) "do-expr") (hk-desugar (nth s 1)))
(:else
(raise "do block must end with an expression")))))
(:else
(let
((s (first stmts)) (rest-stmts (rest stmts)))
(let
((rest-do (hk-desugar-do rest-stmts)))
(cond
((= (first s) "do-expr")
(list
:app
(list
:app
(list :var ">>")
(hk-desugar (nth s 1)))
rest-do))
((= (first s) "do-bind")
(list
:app
(list
:app
(list :var ">>=")
(hk-desugar (nth s 2)))
(list :lambda (list (nth s 1)) rest-do)))
((= (first s) "do-let")
(list
:let
(map hk-desugar (nth s 1))
rest-do))
(:else (raise "unknown do-stmt tag")))))))))
;; List-comprehension desugaring (Haskell 98 §3.11):
;; [e | ] = [e]
;; [e | b, Q ] = if b then [e | Q] else []
;; [e | p <- l, Q ] = concatMap (\p -> [e | Q]) l
;; [e | let ds, Q ] = let ds in [e | Q]
(define
hk-lc-desugar
(fn
(e quals)
(cond
((empty? quals) (list :list (list e)))
(:else
(let
((q (first quals)))
(let
((qtag (first q)))
(cond
((= qtag "q-guard")
(list
:if
(hk-desugar (nth q 1))
(hk-lc-desugar e (rest quals))
(list :list (list))))
((= qtag "q-gen")
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (nth q 1))
(hk-lc-desugar e (rest quals))))
(hk-desugar (nth q 2))))
((= qtag "q-let")
(list
:let
(map hk-desugar (nth q 1))
(hk-lc-desugar e (rest quals))))
(:else
(raise
(str
"hk-lc-desugar: unknown qualifier tag "
qtag))))))))))
(define
hk-desugar
(fn
(node)
(cond
((not (list? node)) node)
((empty? node) node)
(:else
(let
((tag (first node)))
(cond
;; Transformations
((= tag "where")
(list
:let
(map hk-desugar (nth node 2))
(hk-desugar (nth node 1))))
((= tag "guarded") (hk-guards-to-if (nth node 1)))
((= tag "list-comp")
(hk-lc-desugar
(hk-desugar (nth node 1))
(nth node 2)))
;; Expression nodes
((= tag "app")
(list
:app
(hk-desugar (nth node 1))
(hk-desugar (nth node 2))))
((= tag "op")
(list
:op
(nth node 1)
(hk-desugar (nth node 2))
(hk-desugar (nth node 3))))
((= tag "neg") (list :neg (hk-desugar (nth node 1))))
((= tag "if")
(list
:if
(hk-desugar (nth node 1))
(hk-desugar (nth node 2))
(hk-desugar (nth node 3))))
((= tag "tuple")
(list :tuple (map hk-desugar (nth node 1))))
((= tag "list")
(list :list (map hk-desugar (nth node 1))))
((= tag "range")
(list
:range
(hk-desugar (nth node 1))
(hk-desugar (nth node 2))))
((= tag "range-step")
(list
:range-step
(hk-desugar (nth node 1))
(hk-desugar (nth node 2))
(hk-desugar (nth node 3))))
((= tag "lambda")
(list
:lambda
(nth node 1)
(hk-desugar (nth node 2))))
((= tag "let")
(list
:let
(map hk-desugar (nth node 1))
(hk-desugar (nth node 2))))
((= tag "case")
(list
:case
(hk-desugar (nth node 1))
(map hk-desugar (nth node 2))))
((= tag "alt")
(list :alt (nth node 1) (hk-desugar (nth node 2))))
((= tag "do") (hk-desugar-do (nth node 1)))
((= tag "sect-left")
(list
:sect-left
(nth node 1)
(hk-desugar (nth node 2))))
((= tag "sect-right")
(list
:sect-right
(nth node 1)
(hk-desugar (nth node 2))))
;; Top-level
((= tag "program")
(list :program (map hk-desugar (nth node 1))))
((= tag "module")
(list
:module
(nth node 1)
(nth node 2)
(nth node 3)
(map hk-desugar (nth node 4))))
;; Decls carrying a body
((= tag "fun-clause")
(list
:fun-clause
(nth node 1)
(nth node 2)
(hk-desugar (nth node 3))))
((= tag "pat-bind")
(list
:pat-bind
(nth node 1)
(hk-desugar (nth node 2))))
((= tag "bind")
(list
:bind
(nth node 1)
(hk-desugar (nth node 2))))
;; Everything else: leaf literals, vars, cons, patterns,
;; types, imports, type-sigs, data / newtype / fixity, …
(:else node)))))))
;; Convenience — tokenize + layout + parse + desugar.
(define
hk-core
(fn (src) (hk-desugar (hk-parse-top src))))
(define
hk-core-expr
(fn (src) (hk-desugar (hk-parse src))))

775
lib/haskell/eval.sx Normal file
View File

@@ -0,0 +1,775 @@
;; Haskell strict evaluator (Phase 2).
;;
;; Consumes the post-desugar core AST and produces SX values. Strict
;; throughout — laziness and thunks are Phase 3.
;;
;; Value representation:
;; numbers / strings / chars → raw SX values
;; constructor values → tagged lists (con-name first)
;; functions: closure / multifun → {:type "fn" :kind … …}
;; constructor partials → {:type "con-partial" …}
;; built-ins → {:type "builtin" …}
;;
;; Multi-clause top-level definitions are bundled into a single
;; multifun keyed by name; arguments are gathered through currying
;; until arity is reached, then each clause's pattern list is matched
;; in order. Recursive let bindings work because the binding env is
;; built mutably so closures captured during evaluation see the
;; eventual full env.
(define
hk-dict-copy
(fn
(d)
(let ((nd (dict)))
(for-each
(fn (k) (dict-set! nd k (get d k)))
(keys d))
nd)))
;; ── Thunks (Phase 3 — laziness) ─────────────────────────────
;; A thunk wraps an unevaluated AST plus the env in which it was
;; created. The first call to `hk-force` evaluates the body, replaces
;; the body with the cached value, and flips `forced`. Subsequent
;; forces return the cached value directly.
(define
hk-mk-thunk
(fn
(body env)
{:type "thunk" :body body :env env :forced false :value nil}))
(define
hk-is-thunk?
(fn (v) (and (dict? v) (= (get v "type") "thunk"))))
(define
hk-force
(fn
(v)
(cond
((hk-is-thunk? v)
(cond
((get v "forced") (get v "value"))
(:else
(let
((res (hk-force (hk-eval (get v "body") (get v "env")))))
(dict-set! v "forced" true)
(dict-set! v "value" res)
res))))
(:else v))))
;; Recursive force — used at the test/output boundary so test
;; expectations can compare against fully-evaluated structures.
(define
hk-deep-force
(fn
(v)
(let ((fv (hk-force v)))
(cond
((not (list? fv)) fv)
((empty? fv) fv)
(:else (map hk-deep-force fv))))))
;; ── Function value constructors ──────────────────────────────
(define
hk-mk-closure
(fn
(params body env)
{:type "fn" :kind "closure" :params params :body body :env env}))
(define
hk-mk-multifun
(fn
(arity clauses env)
{:type "fn" :kind "multi" :arity arity :clauses clauses :env env :collected (list)}))
(define
hk-mk-builtin
(fn
(name fn arity)
{:type "builtin" :name name :fn fn :arity arity :lazy false :collected (list)}))
;; A lazy built-in receives its collected args as raw thunks (or
;; values, if those happened to be eager) — the implementation is
;; responsible for forcing exactly what it needs. Used for `seq`
;; and `deepseq`, which are non-strict in their second argument.
(define
hk-mk-lazy-builtin
(fn
(name fn arity)
{:type "builtin" :name name :fn fn :arity arity :lazy true :collected (list)}))
;; ── Apply a function value to one argument ──────────────────
(define
hk-apply
(fn
(f arg)
(let ((f (hk-force f)))
(cond
((not (dict? f))
(raise (str "apply: not a function value: " f)))
((= (get f "type") "fn")
(cond
((= (get f "kind") "closure") (hk-apply-closure f arg))
((= (get f "kind") "multi") (hk-apply-multi f arg))
(:else (raise "apply: unknown fn kind"))))
((= (get f "type") "con-partial") (hk-apply-con-partial f arg))
((= (get f "type") "builtin") (hk-apply-builtin f arg))
(:else (raise "apply: not a function dict"))))))
(define
hk-apply-closure
(fn
(cl arg)
(let
((params (get cl "params"))
(body (get cl "body"))
(env (get cl "env")))
(cond
((empty? params) (raise "apply-closure: no params"))
(:else
(let
((p1 (first params)) (rest-p (rest params)))
(let
((env-after (hk-match p1 arg env)))
(cond
((nil? env-after)
(raise "pattern match failure in lambda"))
((empty? rest-p) (hk-eval body env-after))
(:else
(hk-mk-closure rest-p body env-after))))))))))
(define
hk-apply-multi
(fn
(mf arg)
(let
((arity (get mf "arity"))
(clauses (get mf "clauses"))
(env (get mf "env"))
(collected (append (get mf "collected") (list arg))))
(cond
((< (len collected) arity)
(assoc mf "collected" collected))
(:else (hk-dispatch-multi clauses collected env))))))
(define
hk-dispatch-multi
(fn
(clauses args env)
(cond
((empty? clauses)
(raise "non-exhaustive patterns in function definition"))
(:else
(let
((c (first clauses)))
(let
((pats (first c)) (body (first (rest c))))
(let
((env-after (hk-match-args pats args env)))
(cond
((nil? env-after)
(hk-dispatch-multi (rest clauses) args env))
(:else (hk-eval body env-after))))))))))
(define
hk-match-args
(fn
(pats args env)
(cond
((empty? pats) env)
(:else
(let
((res (hk-match (first pats) (first args) env)))
(cond
((nil? res) nil)
(:else
(hk-match-args (rest pats) (rest args) res))))))))
(define
hk-apply-con-partial
(fn
(cp arg)
(let
((name (get cp "name"))
(arity (get cp "arity"))
(args (append (get cp "args") (list arg))))
(cond
((= (len args) arity) (hk-mk-con name args))
(:else (assoc cp "args" args))))))
(define
hk-apply-builtin
(fn
(b arg)
(let
((arity (get b "arity"))
(collected (append (get b "collected") (list arg))))
(cond
((< (len collected) arity)
(assoc b "collected" collected))
(:else
;; Strict built-ins force every collected arg before
;; calling. Lazy ones (`seq`, `deepseq`) receive the raw
;; thunks so they can choose what to force.
(cond
((get b "lazy") (apply (get b "fn") collected))
(:else
(apply
(get b "fn")
(map hk-force collected)))))))))
;; ── Bool helpers (Bool values are tagged conses) ────────────
(define
hk-truthy?
(fn
(v)
(and (list? v) (not (empty? v)) (= (first v) "True"))))
(define hk-true (hk-mk-con "True" (list)))
(define hk-false (hk-mk-con "False" (list)))
(define hk-of-bool (fn (b) (if b hk-true hk-false)))
;; ── Core eval ───────────────────────────────────────────────
(define
hk-eval
(fn
(node env)
(cond
((not (list? node)) (raise (str "eval: not a list: " node)))
((empty? node) (raise "eval: empty list node"))
(:else
(let
((tag (first node)))
(cond
((= tag "int") (nth node 1))
((= tag "float") (nth node 1))
((= tag "string") (nth node 1))
((= tag "char") (nth node 1))
((= tag "var") (hk-eval-var (nth node 1) env))
((= tag "con") (hk-eval-con-ref (nth node 1)))
((= tag "neg")
(- 0 (hk-force (hk-eval (nth node 1) env))))
((= tag "if") (hk-eval-if node env))
((= tag "let") (hk-eval-let (nth node 1) (nth node 2) env))
((= tag "lambda")
(hk-mk-closure (nth node 1) (nth node 2) env))
((= tag "app")
(hk-apply
(hk-eval (nth node 1) env)
(hk-mk-thunk (nth node 2) env)))
((= tag "op")
(hk-eval-op
(nth node 1)
(nth node 2)
(nth node 3)
env))
((= tag "case")
(hk-eval-case (nth node 1) (nth node 2) env))
((= tag "tuple")
(hk-mk-tuple
(map (fn (e) (hk-eval e env)) (nth node 1))))
((= tag "list")
(hk-mk-list
(map (fn (e) (hk-eval e env)) (nth node 1))))
((= tag "range")
(let
((from (hk-force (hk-eval (nth node 1) env)))
(to (hk-force (hk-eval (nth node 2) env))))
(hk-build-range from to 1)))
((= tag "range-step")
(let
((from (hk-force (hk-eval (nth node 1) env)))
(nxt (hk-force (hk-eval (nth node 2) env)))
(to (hk-force (hk-eval (nth node 3) env))))
(hk-build-range from to (- nxt from))))
((= tag "range-from")
;; [from..] = iterate (+ 1) from — uses the Prelude.
(hk-eval
(list
:app
(list
:app
(list :var "iterate")
(list
:sect-right
"+"
(list :int 1)))
(nth node 1))
env))
((= tag "sect-left")
(hk-eval-sect-left (nth node 1) (nth node 2) env))
((= tag "sect-right")
(hk-eval-sect-right (nth node 1) (nth node 2) env))
(:else
(raise (str "eval: unknown node tag '" tag "'")))))))))
(define
hk-eval-var
(fn
(name env)
(cond
((has-key? env name) (get env name))
((hk-is-con? name) (hk-eval-con-ref name))
(:else (raise (str "unbound variable: " name))))))
(define
hk-eval-con-ref
(fn
(name)
(let ((arity (hk-con-arity name)))
(cond
((nil? arity) (raise (str "unknown constructor: " name)))
((= arity 0) (hk-mk-con name (list)))
(:else
{:type "con-partial" :name name :arity arity :args (list)})))))
(define
hk-eval-if
(fn
(node env)
(let ((cv (hk-force (hk-eval (nth node 1) env))))
(cond
((hk-truthy? cv) (hk-eval (nth node 2) env))
((and (list? cv) (= (first cv) "False"))
(hk-eval (nth node 3) env))
((= cv true) (hk-eval (nth node 2) env))
((= cv false) (hk-eval (nth node 3) env))
(:else (raise "if: condition is not Bool"))))))
(define
hk-extend-env-with-match!
(fn
(env match-env)
(for-each
(fn (k) (dict-set! env k (get match-env k)))
(keys match-env))))
(define
hk-eval-let-bind!
(fn
(b env)
(let ((tag (first b)))
(cond
((= tag "fun-clause")
(let
((name (nth b 1))
(pats (nth b 2))
(body (nth b 3)))
(cond
((empty? pats)
(dict-set! env name (hk-eval body env)))
(:else
(dict-set! env name (hk-mk-closure pats body env))))))
((or (= tag "bind") (= tag "pat-bind"))
(let ((pat (nth b 1)) (body (nth b 2)))
(let ((val (hk-eval body env)))
(let ((res (hk-match pat val env)))
(cond
((nil? res)
(raise "let: pattern bind failure"))
(:else
(hk-extend-env-with-match! env res)))))))
(:else nil)))))
(define
hk-eval-let
(fn
(binds body env)
(let ((new-env (hk-dict-copy env)))
;; Pre-seed names for fn-clauses so closures see themselves
;; (mutual recursion across the whole binding group).
(for-each
(fn (b)
(cond
((= (first b) "fun-clause")
(dict-set! new-env (nth b 1) nil))
((and
(= (first b) "bind")
(list? (nth b 1))
(= (first (nth b 1)) "p-var"))
(dict-set! new-env (nth (nth b 1) 1) nil))
(:else nil)))
binds)
(for-each (fn (b) (hk-eval-let-bind! b new-env)) binds)
(hk-eval body new-env))))
(define
hk-eval-case
(fn
(scrut alts env)
(let ((sv (hk-force (hk-eval scrut env))))
(hk-try-alts alts sv env))))
(define
hk-try-alts
(fn
(alts val env)
(cond
((empty? alts) (raise "case: non-exhaustive patterns"))
(:else
(let
((alt (first alts)))
(let
((pat (nth alt 1)) (body (nth alt 2)))
(let
((res (hk-match pat val env)))
(cond
((nil? res) (hk-try-alts (rest alts) val env))
(:else (hk-eval body res))))))))))
(define
hk-eval-op
(fn
(op left right env)
(cond
;; Cons is non-strict in both args: build a cons cell whose
;; head and tail are deferred. This is what makes `repeat x =
;; x : repeat x` and `fibs = 0 : 1 : zipWith (+) fibs (tail
;; fibs)` terminate.
((= op ":")
(hk-mk-cons
(hk-mk-thunk left env)
(hk-mk-thunk right env)))
(:else
(let
((lv (hk-force (hk-eval left env)))
(rv (hk-force (hk-eval right env))))
(hk-binop op lv rv))))))
(define
hk-list-append
(fn
(a b)
(cond
((and (list? a) (= (first a) "[]")) b)
((and (list? a) (= (first a) ":"))
(hk-mk-cons (nth a 1) (hk-list-append (nth a 2) b)))
(:else (raise "++: not a list")))))
;; Eager finite-range spine — handles [from..to] and [from,next..to].
;; Step direction is governed by the sign of `step`; when step > 0 we
;; stop at to; when step < 0 we stop at to going down.
(define
hk-build-range
(fn
(from to step)
(cond
((and (> step 0) (> from to)) (hk-mk-nil))
((and (< step 0) (< from to)) (hk-mk-nil))
((= step 0) (hk-mk-nil))
(:else
(hk-mk-cons from (hk-build-range (+ from step) to step))))))
(define
hk-binop
(fn
(op lv rv)
(cond
((= op "+") (+ lv rv))
((= op "-") (- lv rv))
((= op "*") (* lv rv))
((= op "/") (/ lv rv))
((= op "==") (hk-of-bool (= lv rv)))
((= op "/=") (hk-of-bool (not (= lv rv))))
((= op "<") (hk-of-bool (< lv rv)))
((= op "<=") (hk-of-bool (<= lv rv)))
((= op ">") (hk-of-bool (> lv rv)))
((= op ">=") (hk-of-bool (>= lv rv)))
((= op "&&") (hk-of-bool (and (hk-truthy? lv) (hk-truthy? rv))))
((= op "||") (hk-of-bool (or (hk-truthy? lv) (hk-truthy? rv))))
((= op ":") (hk-mk-cons lv rv))
((= op "++") (hk-list-append lv rv))
(:else (raise (str "unknown operator: " op))))))
(define
hk-eval-sect-left
(fn
(op e env)
;; (e op) = \x -> e op x — bind e once, defer the operator call.
(let ((ev (hk-eval e env)))
(let ((cenv (hk-dict-copy env)))
(dict-set! cenv "__hk-sect-l" ev)
(hk-mk-closure
(list (list :p-var "__hk-sect-x"))
(list
:op
op
(list :var "__hk-sect-l")
(list :var "__hk-sect-x"))
cenv)))))
(define
hk-eval-sect-right
(fn
(op e env)
(let ((ev (hk-eval e env)))
(let ((cenv (hk-dict-copy env)))
(dict-set! cenv "__hk-sect-r" ev)
(hk-mk-closure
(list (list :p-var "__hk-sect-x"))
(list
:op
op
(list :var "__hk-sect-x")
(list :var "__hk-sect-r"))
cenv)))))
;; ── Top-level program evaluation ────────────────────────────
;; Operator-as-value built-ins — let `(+)`, `(*)`, etc. work as
;; first-class functions for `zipWith (+)` and friends. Strict in
;; both args (built-ins are forced via hk-apply-builtin).
(define
hk-make-binop-builtin
(fn
(name op-name)
(hk-mk-builtin
name
(fn (a b) (hk-binop op-name a b))
2)))
;; Inline Prelude source — loaded into the initial env so simple
;; programs can use `head`, `take`, `repeat`, etc. without each
;; user file redefining them. The Prelude itself uses lazy `:` for
;; the recursive list-building functions.
(define
hk-prelude-src
"head (x:_) = x
tail (_:xs) = xs
fst (a, _) = a
snd (_, b) = b
take 0 _ = []
take _ [] = []
take n (x:xs) = x : take (n - 1) xs
drop 0 xs = xs
drop _ [] = []
drop n (_:xs) = drop (n - 1) xs
repeat x = x : repeat x
iterate f x = x : iterate f (f x)
length [] = 0
length (_:xs) = 1 + length xs
map _ [] = []
map f (x:xs) = f x : map f xs
filter _ [] = []
filter p (x:xs) = if p x then x : filter p xs else filter p xs
zipWith _ [] _ = []
zipWith _ _ [] = []
zipWith f (x:xs) (y:ys) = f x y : zipWith f xs ys
fibs = 0 : 1 : zipWith plus fibs (tail fibs)
plus a b = a + b
")
(define
hk-load-into!
(fn
(env src)
(let ((ast (hk-core src)))
(hk-register-program! ast)
(let
((decls
(cond
((= (first ast) "program") (nth ast 1))
((= (first ast) "module") (nth ast 4))
(:else (list)))))
(hk-bind-decls! env decls)))))
(define
hk-init-env
(fn
()
(let ((env (dict)))
(dict-set! env "otherwise" hk-true)
(dict-set!
env
"error"
(hk-mk-builtin
"error"
(fn (msg) (raise (str "*** Exception: " msg)))
1))
(dict-set!
env
"not"
(hk-mk-builtin
"not"
(fn (b) (hk-of-bool (not (hk-truthy? b))))
1))
(dict-set!
env
"id"
(hk-mk-builtin "id" (fn (x) x) 1))
;; `seq a b` — strict in `a`, lazy in `b`. Forces `a` to WHNF
;; and returns `b` unchanged (still a thunk if it was one).
(dict-set!
env
"seq"
(hk-mk-lazy-builtin
"seq"
(fn (a b) (do (hk-force a) b))
2))
;; `deepseq a b` — like seq but forces `a` to normal form.
(dict-set!
env
"deepseq"
(hk-mk-lazy-builtin
"deepseq"
(fn (a b) (do (hk-deep-force a) b))
2))
;; ── Stub IO monad ─────────────────────────────────────
;; IO actions are tagged values `("IO" payload)`; `>>=` and
;; `>>` chain them. Lazy in the action arguments so do-blocks
;; can be deeply structured without forcing the whole chain
;; up front.
(dict-set!
env
"return"
(hk-mk-lazy-builtin
"return"
(fn (x) (list "IO" x))
1))
(dict-set!
env
">>="
(hk-mk-lazy-builtin
">>="
(fn (m f)
(let ((io-val (hk-force m)))
(cond
((and
(list? io-val)
(= (first io-val) "IO"))
(hk-apply (hk-force f) (nth io-val 1)))
(:else
(raise "(>>=): left side is not an IO action")))))
2))
(dict-set!
env
">>"
(hk-mk-lazy-builtin
">>"
(fn (m n)
(let ((io-val (hk-force m)))
(cond
((and
(list? io-val)
(= (first io-val) "IO"))
(hk-force n))
(:else
(raise "(>>): left side is not an IO action")))))
2))
;; Operators as first-class values
(dict-set! env "+" (hk-make-binop-builtin "+" "+"))
(dict-set! env "-" (hk-make-binop-builtin "-" "-"))
(dict-set! env "*" (hk-make-binop-builtin "*" "*"))
(dict-set! env "/" (hk-make-binop-builtin "/" "/"))
(dict-set! env "==" (hk-make-binop-builtin "==" "=="))
(dict-set! env "/=" (hk-make-binop-builtin "/=" "/="))
(dict-set! env "<" (hk-make-binop-builtin "<" "<"))
(dict-set! env "<=" (hk-make-binop-builtin "<=" "<="))
(dict-set! env ">" (hk-make-binop-builtin ">" ">"))
(dict-set! env ">=" (hk-make-binop-builtin ">=" ">="))
(dict-set! env "&&" (hk-make-binop-builtin "&&" "&&"))
(dict-set! env "||" (hk-make-binop-builtin "||" "||"))
(dict-set! env "++" (hk-make-binop-builtin "++" "++"))
(hk-load-into! env hk-prelude-src)
env)))
(define
hk-bind-decls!
(fn
(env decls)
(let ((groups (dict)) (pat-binds (list)))
;; Pass 1: collect fun-clause groups by name; collect pat-binds
;; in source order. Pre-seed env so any name can already be
;; looked up by closures built in pass 2.
(for-each
(fn (d)
(cond
((= (first d) "fun-clause")
(let
((name (nth d 1)))
(dict-set!
groups
name
(append
(if
(has-key? groups name)
(get groups name)
(list))
(list (list (nth d 2) (nth d 3)))))
(when
(not (has-key? env name))
(dict-set! env name nil))))
((or (= (first d) "bind") (= (first d) "pat-bind"))
(append! pat-binds d))
(:else nil)))
decls)
;; Pass 2: install multifuns for arity > 0; mark 0-arity for
;; pass 3. The mutable env means recursive references work.
(let ((zero-arity (list)))
(for-each
(fn (name)
(let ((clauses (get groups name)))
(let ((arity (len (first (first clauses)))))
(cond
((> arity 0)
(dict-set!
env
name
(hk-mk-multifun arity clauses env)))
(:else (append! zero-arity name))))))
(keys groups))
;; Pass 3: evaluate 0-arity bodies and pat-binds.
(for-each
(fn (name)
(let ((clauses (get groups name)))
(dict-set!
env
name
(hk-eval (first (rest (first clauses))) env))))
zero-arity)
(for-each
(fn (d)
(let ((pat (nth d 1)) (body (nth d 2)))
(let ((val (hk-eval body env)))
(let ((res (hk-match pat val env)))
(cond
((nil? res)
(raise "top-level pattern bind failure"))
(:else (hk-extend-env-with-match! env res)))))))
pat-binds))
env)))
(define
hk-eval-program
(fn
(ast)
(cond
((nil? ast) (raise "eval-program: nil ast"))
((not (list? ast)) (raise "eval-program: not a list"))
(:else
(do
(hk-register-program! ast)
(let ((env (hk-init-env)))
(let
((decls
(cond
((= (first ast) "program") (nth ast 1))
((= (first ast) "module") (nth ast 4))
(:else (raise "eval-program: bad shape")))))
(hk-bind-decls! env decls))))))))
;; ── Source-level convenience ────────────────────────────────
(define
hk-run
(fn
(src)
(let ((env (hk-eval-program (hk-core src))))
(cond
((has-key? env "main") (get env "main"))
(:else env)))))
(define
hk-eval-expr-source
(fn
(src)
(hk-deep-force (hk-eval (hk-core-expr src) (hk-init-env)))))

329
lib/haskell/layout.sx Normal file
View File

@@ -0,0 +1,329 @@
;; Haskell 98 layout algorithm (§10.3).
;;
;; Consumes the raw token stream produced by hk-tokenize and inserts
;; virtual braces / semicolons (types vlbrace / vrbrace / vsemi) based
;; on indentation. Newline tokens are consumed and stripped.
;;
;; (hk-layout (hk-tokenize src)) → tokens-with-virtual-layout
;; ── Pre-pass ──────────────────────────────────────────────────────
;;
;; Walks the raw token list and emits an augmented stream containing
;; two fresh pseudo-tokens:
;;
;; {:type "layout-open" :col N :keyword K}
;; At stream start (K = "<module>") unless the first real token is
;; `module` or `{`. Also immediately after every `let` / `where` /
;; `do` / `of` whose following token is NOT `{`. N is the column
;; of the token that follows.
;;
;; {:type "layout-indent" :col N}
;; Before any token whose line is strictly greater than the line
;; of the previously emitted real token, EXCEPT when that token
;; is already preceded by a layout-open (Haskell 98 §10.3 note 3).
;;
;; Raw newline tokens are dropped.
(define
hk-layout-keyword?
(fn
(tok)
(and
(= (get tok "type") "reserved")
(or
(= (get tok "value") "let")
(= (get tok "value") "where")
(= (get tok "value") "do")
(= (get tok "value") "of")))))
(define
hk-layout-pre
(fn
(tokens)
(let
((result (list))
(n (len tokens))
(i 0)
(prev-line -1)
(first-real-emitted false)
(suppress-next-indent false))
(define
hk-next-real-idx
(fn
(start)
(let
((j start))
(define
hk-nri-loop
(fn
()
(when
(and
(< j n)
(= (get (nth tokens j) "type") "newline"))
(do (set! j (+ j 1)) (hk-nri-loop)))))
(hk-nri-loop)
j)))
(define
hk-pre-step
(fn
()
(when
(< i n)
(let
((tok (nth tokens i)) (ty (get tok "type")))
(cond
((= ty "newline") (do (set! i (+ i 1)) (hk-pre-step)))
(:else
(do
(when
(not first-real-emitted)
(do
(set! first-real-emitted true)
(when
(not
(or
(and
(= ty "reserved")
(= (get tok "value") "module"))
(= ty "lbrace")))
(do
(append!
result
{:type "layout-open"
:col (get tok "col")
:keyword "<module>"
:line (get tok "line")})
(set! suppress-next-indent true)))))
(when
(and
(>= prev-line 0)
(> (get tok "line") prev-line)
(not suppress-next-indent))
(append!
result
{:type "layout-indent"
:col (get tok "col")
:line (get tok "line")}))
(set! suppress-next-indent false)
(set! prev-line (get tok "line"))
(append! result tok)
(when
(hk-layout-keyword? tok)
(let
((j (hk-next-real-idx (+ i 1))))
(cond
((>= j n)
(do
(append!
result
{:type "layout-open"
:col 0
:keyword (get tok "value")
:line (get tok "line")})
(set! suppress-next-indent true)))
((= (get (nth tokens j) "type") "lbrace") nil)
(:else
(do
(append!
result
{:type "layout-open"
:col (get (nth tokens j) "col")
:keyword (get tok "value")
:line (get tok "line")})
(set! suppress-next-indent true))))))
(set! i (+ i 1))
(hk-pre-step))))))))
(hk-pre-step)
result)))
;; ── Main pass: L algorithm ────────────────────────────────────────
;;
;; Stack is a list; the head is the top of stack. Each entry is
;; either the keyword :explicit (pushed by an explicit `{`) or a dict
;; {:col N :keyword K} pushed by a layout-open marker.
;;
;; Rules (following Haskell 98 §10.3):
;;
;; layout-open(n) vs stack:
;; empty or explicit top → push n; emit {
;; n > top-col → push n; emit {
;; otherwise → emit { }; retry as indent(n)
;;
;; layout-indent(n) vs stack:
;; empty or explicit top → drop
;; n == top-col → emit ;
;; n < top-col → emit }; pop; recurse
;; n > top-col → drop
;;
;; lbrace → push :explicit; emit {
;; rbrace → pop if :explicit; emit }
;; `in` with implicit let on top → emit }; pop; emit in
;; any other token → emit
;;
;; EOF: emit } for every remaining implicit context.
(define
hk-layout-L
(fn
(pre-toks)
(let
((result (list))
(stack (list))
(n (len pre-toks))
(i 0))
(define hk-emit (fn (t) (append! result t)))
(define
hk-indent-at
(fn
(col line)
(cond
((or (empty? stack) (= (first stack) :explicit)) nil)
(:else
(let
((top-col (get (first stack) "col")))
(cond
((= col top-col)
(hk-emit
{:type "vsemi" :value ";" :line line :col col}))
((< col top-col)
(do
(hk-emit
{:type "vrbrace" :value "}" :line line :col col})
(set! stack (rest stack))
(hk-indent-at col line)))
(:else nil)))))))
(define
hk-open-at
(fn
(col keyword line)
(cond
((and
(> col 0)
(or
(empty? stack)
(= (first stack) :explicit)
(> col (get (first stack) "col"))))
(do
(hk-emit
{:type "vlbrace" :value "{" :line line :col col})
(set! stack (cons {:col col :keyword keyword} stack))))
(:else
(do
(hk-emit
{:type "vlbrace" :value "{" :line line :col col})
(hk-emit
{:type "vrbrace" :value "}" :line line :col col})
(hk-indent-at col line))))))
(define
hk-close-eof
(fn
()
(when
(and
(not (empty? stack))
(not (= (first stack) :explicit)))
(do
(hk-emit {:type "vrbrace" :value "}" :line 0 :col 0})
(set! stack (rest stack))
(hk-close-eof)))))
;; Peek past further layout-indent / layout-open markers to find
;; the next real token's value when its type is `reserved`.
;; Returns nil if no such token.
(define
hk-peek-next-reserved
(fn
(start)
(let ((j (+ start 1)) (found nil) (done false))
(define
hk-pnr-loop
(fn
()
(when
(and (not done) (< j n))
(let
((t (nth pre-toks j)) (ty (get t "type")))
(cond
((or
(= ty "layout-indent")
(= ty "layout-open"))
(do (set! j (+ j 1)) (hk-pnr-loop)))
((= ty "reserved")
(do (set! found (get t "value")) (set! done true)))
(:else (set! done true)))))))
(hk-pnr-loop)
found)))
(define
hk-layout-step
(fn
()
(when
(< i n)
(let
((tok (nth pre-toks i)) (ty (get tok "type")))
(cond
((= ty "eof")
(do
(hk-close-eof)
(hk-emit tok)
(set! i (+ i 1))
(hk-layout-step)))
((= ty "layout-open")
(do
(hk-open-at
(get tok "col")
(get tok "keyword")
(get tok "line"))
(set! i (+ i 1))
(hk-layout-step)))
((= ty "layout-indent")
(cond
((= (hk-peek-next-reserved i) "in")
(do (set! i (+ i 1)) (hk-layout-step)))
(:else
(do
(hk-indent-at (get tok "col") (get tok "line"))
(set! i (+ i 1))
(hk-layout-step)))))
((= ty "lbrace")
(do
(set! stack (cons :explicit stack))
(hk-emit tok)
(set! i (+ i 1))
(hk-layout-step)))
((= ty "rbrace")
(do
(when
(and
(not (empty? stack))
(= (first stack) :explicit))
(set! stack (rest stack)))
(hk-emit tok)
(set! i (+ i 1))
(hk-layout-step)))
((and
(= ty "reserved")
(= (get tok "value") "in")
(not (empty? stack))
(not (= (first stack) :explicit))
(= (get (first stack) "keyword") "let"))
(do
(hk-emit
{:type "vrbrace"
:value "}"
:line (get tok "line")
:col (get tok "col")})
(set! stack (rest stack))
(hk-emit tok)
(set! i (+ i 1))
(hk-layout-step)))
(:else
(do
(hk-emit tok)
(set! i (+ i 1))
(hk-layout-step))))))))
(hk-layout-step)
(hk-close-eof)
result)))
(define hk-layout (fn (tokens) (hk-layout-L (hk-layout-pre tokens))))

201
lib/haskell/match.sx Normal file
View File

@@ -0,0 +1,201 @@
;; Value-level pattern matching.
;;
;; Constructor values are tagged lists whose first element is the
;; constructor name (a string). Tuples use the special tag "Tuple".
;; Lists use the spine of `:` cons and `[]` nil.
;;
;; Just 5 → ("Just" 5)
;; Nothing → ("Nothing")
;; (1, 2) → ("Tuple" 1 2)
;; [1, 2] → (":" 1 (":" 2 ("[]")))
;; () → ("()")
;;
;; Primitive values (numbers, strings, chars) are stored raw.
;;
;; The matcher takes a pattern AST node, a value, and an environment
;; dict; it returns an extended dict on success, or `nil` on failure.
;; ── Value builders ──────────────────────────────────────────
(define
hk-mk-con
(fn
(cname args)
(let ((result (list cname)))
(for-each (fn (a) (append! result a)) args)
result)))
(define
hk-mk-tuple
(fn
(items)
(let ((result (list "Tuple")))
(for-each (fn (x) (append! result x)) items)
result)))
(define hk-mk-nil (fn () (list "[]")))
(define hk-mk-cons (fn (h t) (list ":" h t)))
(define
hk-mk-list
(fn
(items)
(cond
((empty? items) (hk-mk-nil))
(:else
(hk-mk-cons (first items) (hk-mk-list (rest items)))))))
;; ── Predicates / accessors on constructor values ───────────
(define
hk-is-con-val?
(fn
(v)
(and
(list? v)
(not (empty? v))
(string? (first v)))))
(define hk-val-con-name (fn (v) (first v)))
(define hk-val-con-args (fn (v) (rest v)))
;; ── The matcher ────────────────────────────────────────────
;;
;; Pattern match forces the scrutinee to WHNF before inspecting it
;; — except for `p-wild`, `p-var`, and `p-lazy`, which never need
;; to look at the value. Args of constructor / tuple / list values
;; remain thunked (they're forced only when their own pattern needs
;; to inspect them, recursively).
(define
hk-match
(fn
(pat val env)
(cond
((not (list? pat)) nil)
((empty? pat) nil)
(:else
(let
((tag (first pat)))
(cond
((= tag "p-wild") env)
((= tag "p-var") (assoc env (nth pat 1) val))
((= tag "p-lazy") (hk-match (nth pat 1) val env))
((= tag "p-as")
(let
((res (hk-match (nth pat 2) val env)))
(cond
((nil? res) nil)
(:else (assoc res (nth pat 1) val)))))
(:else
(let ((fv (hk-force val)))
(cond
((= tag "p-int")
(if
(and (number? fv) (= fv (nth pat 1)))
env
nil))
((= tag "p-float")
(if
(and (number? fv) (= fv (nth pat 1)))
env
nil))
((= tag "p-string")
(if
(and (string? fv) (= fv (nth pat 1)))
env
nil))
((= tag "p-char")
(if
(and (string? fv) (= fv (nth pat 1)))
env
nil))
((= tag "p-con")
(let
((pat-name (nth pat 1)) (pat-args (nth pat 2)))
(cond
((not (hk-is-con-val? fv)) nil)
((not (= (hk-val-con-name fv) pat-name)) nil)
(:else
(let
((val-args (hk-val-con-args fv)))
(cond
((not (= (len pat-args) (len val-args)))
nil)
(:else
(hk-match-all
pat-args
val-args
env))))))))
((= tag "p-tuple")
(let
((items (nth pat 1)))
(cond
((not (hk-is-con-val? fv)) nil)
((not (= (hk-val-con-name fv) "Tuple")) nil)
((not (= (len (hk-val-con-args fv)) (len items)))
nil)
(:else
(hk-match-all
items
(hk-val-con-args fv)
env)))))
((= tag "p-list")
(hk-match-list-pat (nth pat 1) fv env))
(:else nil))))))))))
(define
hk-match-all
(fn
(pats vals env)
(cond
((empty? pats) env)
(:else
(let
((res (hk-match (first pats) (first vals) env)))
(cond
((nil? res) nil)
(:else
(hk-match-all (rest pats) (rest vals) res))))))))
(define
hk-match-list-pat
(fn
(items val env)
(let ((fv (hk-force val)))
(cond
((empty? items)
(if
(and
(hk-is-con-val? fv)
(= (hk-val-con-name fv) "[]"))
env
nil))
(:else
(cond
((not (hk-is-con-val? fv)) nil)
((not (= (hk-val-con-name fv) ":")) nil)
(:else
(let
((args (hk-val-con-args fv)))
(let
((h (first args)) (t (first (rest args))))
(let
((res (hk-match (first items) h env)))
(cond
((nil? res) nil)
(:else
(hk-match-list-pat
(rest items)
t
res)))))))))))))
;; ── Convenience: parse a pattern from source for tests ─────
;; (Uses the parser's case-alt entry — `case _ of pat -> 0` —
;; to extract a pattern AST.)
(define
hk-parse-pat-source
(fn
(src)
(let
((expr (hk-parse (str "case 0 of " src " -> 0"))))
(nth (nth (nth expr 2) 0) 1))))

1994
lib/haskell/parser.sx Normal file

File diff suppressed because it is too large Load Diff

130
lib/haskell/runtime.sx Normal file
View File

@@ -0,0 +1,130 @@
;; Haskell runtime: constructor registry.
;;
;; A mutable dict keyed by constructor name (e.g. "Just", "[]") with
;; entries of shape {:arity N :type TYPE-NAME-STRING}.
;; Populated by ingesting `data` / `newtype` decls from parsed ASTs.
;; Pre-registers a small set of constructors tied to Haskell syntactic
;; forms (Bool, list, unit) — every nontrivial program depends on
;; these, and the parser/desugar pipeline emits them as (:var "True")
;; etc. without a corresponding `data` decl.
(define hk-constructors (dict))
(define
hk-register-con!
(fn
(cname arity type-name)
(dict-set!
hk-constructors
cname
{:arity arity :type type-name})))
(define hk-is-con? (fn (name) (has-key? hk-constructors name)))
(define
hk-con-arity
(fn
(name)
(if
(has-key? hk-constructors name)
(get (get hk-constructors name) "arity")
nil)))
(define
hk-con-type
(fn
(name)
(if
(has-key? hk-constructors name)
(get (get hk-constructors name) "type")
nil)))
(define hk-con-names (fn () (keys hk-constructors)))
;; ── Registration from AST ────────────────────────────────────
;; (:data NAME TVARS ((:con-def CNAME FIELDS) …))
(define
hk-register-data!
(fn
(data-node)
(let
((type-name (nth data-node 1))
(cons-list (nth data-node 3)))
(for-each
(fn
(cd)
(hk-register-con!
(nth cd 1)
(len (nth cd 2))
type-name))
cons-list))))
;; (:newtype NAME TVARS CNAME FIELD)
(define
hk-register-newtype!
(fn
(nt-node)
(hk-register-con!
(nth nt-node 3)
1
(nth nt-node 1))))
;; Walk a decls list, registering every `data` / `newtype` decl.
(define
hk-register-decls!
(fn
(decls)
(for-each
(fn
(d)
(cond
((and
(list? d)
(not (empty? d))
(= (first d) "data"))
(hk-register-data! d))
((and
(list? d)
(not (empty? d))
(= (first d) "newtype"))
(hk-register-newtype! d))
(:else nil)))
decls)))
(define
hk-register-program!
(fn
(ast)
(cond
((nil? ast) nil)
((not (list? ast)) nil)
((empty? ast) nil)
((= (first ast) "program")
(hk-register-decls! (nth ast 1)))
((= (first ast) "module")
(hk-register-decls! (nth ast 4)))
(:else nil))))
;; Convenience: source → AST → desugar → register.
(define
hk-load-source!
(fn (src) (hk-register-program! (hk-core src))))
;; ── Built-in constructors pre-registered ─────────────────────
;; Bool — used implicitly by `if`, comparison operators.
(hk-register-con! "True" 0 "Bool")
(hk-register-con! "False" 0 "Bool")
;; List — used by list literals, range syntax, and cons operator.
(hk-register-con! "[]" 0 "List")
(hk-register-con! ":" 2 "List")
;; Unit — produced by empty parens `()`.
(hk-register-con! "()" 0 "Unit")
;; Standard Prelude types — pre-registered so expression-level
;; programs can use them without a `data` decl.
(hk-register-con! "Nothing" 0 "Maybe")
(hk-register-con! "Just" 1 "Maybe")
(hk-register-con! "Left" 1 "Either")
(hk-register-con! "Right" 1 "Either")
(hk-register-con! "LT" 0 "Ordering")
(hk-register-con! "EQ" 0 "Ordering")
(hk-register-con! "GT" 0 "Ordering")

View File

@@ -46,6 +46,13 @@ for FILE in "${FILES[@]}"; do
cat > "$TMPFILE" <<EPOCHS
(epoch 1)
(load "lib/haskell/tokenizer.sx")
(load "lib/haskell/layout.sx")
(load "lib/haskell/parser.sx")
(load "lib/haskell/desugar.sx")
(load "lib/haskell/runtime.sx")
(load "lib/haskell/match.sx")
(load "lib/haskell/eval.sx")
(load "lib/haskell/testlib.sx")
(epoch 2)
(load "$FILE")
(epoch 3)
@@ -81,6 +88,13 @@ EPOCHS
cat > "$TMPFILE2" <<EPOCHS
(epoch 1)
(load "lib/haskell/tokenizer.sx")
(load "lib/haskell/layout.sx")
(load "lib/haskell/parser.sx")
(load "lib/haskell/desugar.sx")
(load "lib/haskell/runtime.sx")
(load "lib/haskell/match.sx")
(load "lib/haskell/eval.sx")
(load "lib/haskell/testlib.sx")
(epoch 2)
(load "$FILE")
(epoch 3)

58
lib/haskell/testlib.sx Normal file
View File

@@ -0,0 +1,58 @@
;; Shared test harness for Haskell-on-SX tests.
;; Each test file expects hk-test / hk-deep=? / counters to already be bound.
(define
hk-deep=?
(fn
(a b)
(cond
((= a b) true)
((and (dict? a) (dict? b))
(let
((ak (keys a)) (bk (keys b)))
(if
(not (= (len ak) (len bk)))
false
(every?
(fn
(k)
(and (has-key? b k) (hk-deep=? (get a k) (get b k))))
ak))))
((and (list? a) (list? b))
(if
(not (= (len a) (len b)))
false
(let
((i 0) (ok true))
(define
hk-de-loop
(fn
()
(when
(and ok (< i (len a)))
(do
(when
(not (hk-deep=? (nth a i) (nth b i)))
(set! ok false))
(set! i (+ i 1))
(hk-de-loop)))))
(hk-de-loop)
ok)))
(:else false))))
(define hk-test-pass 0)
(define hk-test-fail 0)
(define hk-test-fails (list))
(define
hk-test
(fn
(name actual expected)
(if
(hk-deep=? actual expected)
(set! hk-test-pass (+ hk-test-pass 1))
(do
(set! hk-test-fail (+ hk-test-fail 1))
(append!
hk-test-fails
{:actual actual :expected expected :name name})))))

View File

@@ -0,0 +1,305 @@
;; Desugar tests — surface AST → core AST.
;; :guarded → nested :if
;; :where → :let
;; :list-comp → concatMap-based tree
(define
hk-prog
(fn (&rest decls) (list :program decls)))
;; ── Guards → if ──
(hk-test
"two-way guarded rhs"
(hk-desugar (hk-parse-top "abs x | x < 0 = - x\n | otherwise = x"))
(hk-prog
(list
:fun-clause
"abs"
(list (list :p-var "x"))
(list
:if
(list :op "<" (list :var "x") (list :int 0))
(list :neg (list :var "x"))
(list
:if
(list :var "otherwise")
(list :var "x")
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards")))))))
(hk-test
"three-way guarded rhs"
(hk-desugar
(hk-parse-top "sign n | n > 0 = 1\n | n < 0 = -1\n | otherwise = 0"))
(hk-prog
(list
:fun-clause
"sign"
(list (list :p-var "n"))
(list
:if
(list :op ">" (list :var "n") (list :int 0))
(list :int 1)
(list
:if
(list :op "<" (list :var "n") (list :int 0))
(list :neg (list :int 1))
(list
:if
(list :var "otherwise")
(list :int 0)
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards"))))))))
(hk-test
"case-alt guards desugared too"
(hk-desugar
(hk-parse "case x of\n Just y | y > 0 -> y\n | otherwise -> 0\n Nothing -> -1"))
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list
:if
(list :op ">" (list :var "y") (list :int 0))
(list :var "y")
(list
:if
(list :var "otherwise")
(list :int 0)
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards")))))
(list
:alt
(list :p-con "Nothing" (list))
(list :neg (list :int 1))))))
;; ── Where → let ──
(hk-test
"where with single binding"
(hk-desugar (hk-parse-top "f x = y\n where y = x + 1"))
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:let
(list
(list
:fun-clause
"y"
(list)
(list :op "+" (list :var "x") (list :int 1))))
(list :var "y")))))
(hk-test
"where with two bindings"
(hk-desugar
(hk-parse-top "f x = y + z\n where y = x + 1\n z = x - 1"))
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:let
(list
(list
:fun-clause
"y"
(list)
(list :op "+" (list :var "x") (list :int 1)))
(list
:fun-clause
"z"
(list)
(list :op "-" (list :var "x") (list :int 1))))
(list :op "+" (list :var "y") (list :var "z"))))))
(hk-test
"guards + where — guarded body inside let"
(hk-desugar
(hk-parse-top "f x | x > 0 = y\n | otherwise = 0\n where y = 99"))
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:let
(list (list :fun-clause "y" (list) (list :int 99)))
(list
:if
(list :op ">" (list :var "x") (list :int 0))
(list :var "y")
(list
:if
(list :var "otherwise")
(list :int 0)
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards"))))))))
;; ── List comprehensions → concatMap / if / let ──
(hk-test
"list-comp: single generator"
(hk-core-expr "[x | x <- xs]")
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (list :p-var "x"))
(list :list (list (list :var "x")))))
(list :var "xs")))
(hk-test
"list-comp: generator then guard"
(hk-core-expr "[x * 2 | x <- xs, x > 0]")
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (list :p-var "x"))
(list
:if
(list :op ">" (list :var "x") (list :int 0))
(list
:list
(list (list :op "*" (list :var "x") (list :int 2))))
(list :list (list)))))
(list :var "xs")))
(hk-test
"list-comp: generator then let"
(hk-core-expr "[y | x <- xs, let y = x + 1]")
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (list :p-var "x"))
(list
:let
(list
(list
:bind
(list :p-var "y")
(list :op "+" (list :var "x") (list :int 1))))
(list :list (list (list :var "y"))))))
(list :var "xs")))
(hk-test
"list-comp: two generators (nested concatMap)"
(hk-core-expr "[(x, y) | x <- xs, y <- ys]")
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (list :p-var "x"))
(list
:app
(list
:app
(list :var "concatMap")
(list
:lambda
(list (list :p-var "y"))
(list
:list
(list
(list
:tuple
(list (list :var "x") (list :var "y")))))))
(list :var "ys"))))
(list :var "xs")))
;; ── Pass-through cases ──
(hk-test
"plain int literal unchanged"
(hk-core-expr "42")
(list :int 42))
(hk-test
"lambda + if passes through"
(hk-core-expr "\\x -> if x > 0 then x else - x")
(list
:lambda
(list (list :p-var "x"))
(list
:if
(list :op ">" (list :var "x") (list :int 0))
(list :var "x")
(list :neg (list :var "x")))))
(hk-test
"simple fun-clause (no guards/where) passes through"
(hk-desugar (hk-parse-top "id x = x"))
(hk-prog
(list
:fun-clause
"id"
(list (list :p-var "x"))
(list :var "x"))))
(hk-test
"data decl passes through"
(hk-desugar (hk-parse-top "data Maybe a = Nothing | Just a"))
(hk-prog
(list
:data
"Maybe"
(list "a")
(list
(list :con-def "Nothing" (list))
(list :con-def "Just" (list (list :t-var "a")))))))
(hk-test
"module header passes through, body desugared"
(hk-desugar
(hk-parse-top "module M where\nf x | x > 0 = 1\n | otherwise = 0"))
(list
:module
"M"
nil
(list)
(list
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:if
(list :op ">" (list :var "x") (list :int 0))
(list :int 1)
(list
:if
(list :var "otherwise")
(list :int 0)
(list
:app
(list :var "error")
(list :string "Non-exhaustive guards"))))))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

117
lib/haskell/tests/do-io.sx Normal file
View File

@@ -0,0 +1,117 @@
;; do-notation + stub IO monad. Desugaring is per Haskell 98 §3.14:
;; do { e ; ss } = e >> do { ss }
;; do { p <- e ; ss } = e >>= \p -> do { ss }
;; do { let ds ; ss } = let ds in do { ss }
;; do { e } = e
;; The IO type is just `("IO" payload)` for now — no real side
;; effects yet. `return`, `>>=`, `>>` are built-ins.
(define
hk-prog-val
(fn
(src name)
(hk-deep-force (get (hk-eval-program (hk-core src)) name))))
;; ── Single-statement do ──
(hk-test
"do with a single expression"
(hk-eval-expr-source "do { return 5 }")
(list "IO" 5))
(hk-test
"return wraps any expression"
(hk-eval-expr-source "return (1 + 2 * 3)")
(list "IO" 7))
;; ── Bind threads results ──
(hk-test
"single bind"
(hk-eval-expr-source
"do { x <- return 5 ; return (x + 1) }")
(list "IO" 6))
(hk-test
"two binds"
(hk-eval-expr-source
"do\n x <- return 5\n y <- return 7\n return (x + y)")
(list "IO" 12))
(hk-test
"three binds — accumulating"
(hk-eval-expr-source
"do\n a <- return 1\n b <- return 2\n c <- return 3\n return (a + b + c)")
(list "IO" 6))
;; ── Mixing >> and >>= ──
(hk-test
">> sequencing — last wins"
(hk-eval-expr-source
"do\n return 1\n return 2\n return 3")
(list "IO" 3))
(hk-test
">> then >>= — last bind wins"
(hk-eval-expr-source
"do\n return 99\n x <- return 5\n return x")
(list "IO" 5))
;; ── do-let ──
(hk-test
"do-let single binding"
(hk-eval-expr-source
"do\n let x = 3\n return (x * 2)")
(list "IO" 6))
(hk-test
"do-let multi-bind, used after"
(hk-eval-expr-source
"do\n let x = 4\n y = 5\n return (x * y)")
(list "IO" 20))
(hk-test
"do-let interleaved with bind"
(hk-eval-expr-source
"do\n x <- return 10\n let y = x + 1\n return (x * y)")
(list "IO" 110))
;; ── Bind + pattern ──
(hk-test
"bind to constructor pattern"
(hk-eval-expr-source
"do\n Just x <- return (Just 7)\n return (x + 100)")
(list "IO" 107))
(hk-test
"bind to tuple pattern"
(hk-eval-expr-source
"do\n (a, b) <- return (3, 4)\n return (a * b)")
(list "IO" 12))
;; ── User-defined IO functions ──
(hk-test
"do inside top-level fun"
(hk-prog-val
"addM x y = do\n a <- return x\n b <- return y\n return (a + b)\nresult = addM 5 6"
"result")
(list "IO" 11))
(hk-test
"nested do"
(hk-eval-expr-source
"do\n x <- do { y <- return 3 ; return (y + 1) }\n return (x * 2)")
(list "IO" 8))
;; ── (>>=) and (>>) used directly as functions ──
(hk-test
">>= used directly"
(hk-eval-expr-source
"(return 4) >>= (\\x -> return (x + 100))")
(list "IO" 104))
(hk-test
">> used directly"
(hk-eval-expr-source
"(return 1) >> (return 2)")
(list "IO" 2))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

278
lib/haskell/tests/eval.sx Normal file
View File

@@ -0,0 +1,278 @@
;; Strict evaluator tests. Each test parses, desugars, and evaluates
;; either an expression (hk-eval-expr-source) or a full program
;; (hk-eval-program → look up a named value).
(define
hk-prog-val
(fn
(src name)
(hk-deep-force (get (hk-eval-program (hk-core src)) name))))
;; ── Literals ──
(hk-test "int literal" (hk-eval-expr-source "42") 42)
(hk-test "float literal" (hk-eval-expr-source "3.14") 3.14)
(hk-test "string literal" (hk-eval-expr-source "\"hi\"") "hi")
(hk-test "char literal" (hk-eval-expr-source "'a'") "a")
(hk-test "negative literal" (hk-eval-expr-source "- 5") -5)
;; ── Arithmetic ──
(hk-test "addition" (hk-eval-expr-source "1 + 2") 3)
(hk-test
"precedence"
(hk-eval-expr-source "1 + 2 * 3")
7)
(hk-test
"parens override precedence"
(hk-eval-expr-source "(1 + 2) * 3")
9)
(hk-test
"subtraction left-assoc"
(hk-eval-expr-source "10 - 3 - 2")
5)
;; ── Comparison + Bool ──
(hk-test
"less than is True"
(hk-eval-expr-source "3 < 5")
(list "True"))
(hk-test
"equality is False"
(hk-eval-expr-source "1 == 2")
(list "False"))
(hk-test
"&& shortcuts"
(hk-eval-expr-source "(1 == 1) && (2 == 2)")
(list "True"))
;; ── if / otherwise ──
(hk-test
"if True"
(hk-eval-expr-source "if True then 1 else 2")
1)
(hk-test
"if comparison branch"
(hk-eval-expr-source "if 5 > 3 then \"yes\" else \"no\"")
"yes")
(hk-test "otherwise is True" (hk-eval-expr-source "otherwise") (list "True"))
;; ── let ──
(hk-test
"let single binding"
(hk-eval-expr-source "let x = 5 in x + 1")
6)
(hk-test
"let two bindings"
(hk-eval-expr-source "let x = 1; y = 2 in x + y")
3)
(hk-test
"let recursive: factorial 5"
(hk-eval-expr-source
"let f n = if n == 0 then 1 else n * f (n - 1) in f 5")
120)
;; ── Lambdas ──
(hk-test
"lambda apply"
(hk-eval-expr-source "(\\x -> x + 1) 5")
6)
(hk-test
"lambda multi-arg"
(hk-eval-expr-source "(\\x y -> x * y) 3 4")
12)
(hk-test
"lambda with constructor pattern"
(hk-eval-expr-source "(\\(Just x) -> x + 1) (Just 7)")
8)
;; ── Constructors ──
(hk-test
"0-arity constructor"
(hk-eval-expr-source "Nothing")
(list "Nothing"))
(hk-test
"1-arity constructor applied"
(hk-eval-expr-source "Just 5")
(list "Just" 5))
(hk-test
"True / False as bools"
(hk-eval-expr-source "True")
(list "True"))
;; ── case ──
(hk-test
"case Just"
(hk-eval-expr-source
"case Just 7 of Just x -> x ; Nothing -> 0")
7)
(hk-test
"case Nothing"
(hk-eval-expr-source
"case Nothing of Just x -> x ; Nothing -> 99")
99)
(hk-test
"case literal pattern"
(hk-eval-expr-source
"case 0 of 0 -> \"zero\" ; n -> \"other\"")
"zero")
(hk-test
"case tuple"
(hk-eval-expr-source
"case (1, 2) of (a, b) -> a + b")
3)
(hk-test
"case wildcard fallback"
(hk-eval-expr-source
"case 5 of 0 -> \"z\" ; _ -> \"nz\"")
"nz")
;; ── List literals + cons ──
(hk-test
"list literal as cons spine"
(hk-eval-expr-source "[1, 2, 3]")
(list ":" 1 (list ":" 2 (list ":" 3 (list "[]")))))
(hk-test
"empty list literal"
(hk-eval-expr-source "[]")
(list "[]"))
(hk-test
"cons via :"
(hk-eval-expr-source "1 : []")
(list ":" 1 (list "[]")))
(hk-test
"++ concatenates lists"
(hk-eval-expr-source "[1, 2] ++ [3]")
(list ":" 1 (list ":" 2 (list ":" 3 (list "[]")))))
;; ── Tuples ──
(hk-test
"2-tuple"
(hk-eval-expr-source "(1, 2)")
(list "Tuple" 1 2))
(hk-test
"3-tuple"
(hk-eval-expr-source "(\"a\", 5, True)")
(list "Tuple" "a" 5 (list "True")))
;; ── Sections ──
(hk-test
"right section (+ 1) applied"
(hk-eval-expr-source "(+ 1) 5")
6)
(hk-test
"left section (10 -) applied"
(hk-eval-expr-source "(10 -) 4")
6)
;; ── Multi-clause top-level functions ──
(hk-test
"multi-clause: factorial"
(hk-prog-val
"fact 0 = 1\nfact n = n * fact (n - 1)\nresult = fact 6"
"result")
720)
(hk-test
"multi-clause: list length via cons pattern"
(hk-prog-val
"len [] = 0\nlen (x:xs) = 1 + len xs\nresult = len [10, 20, 30, 40]"
"result")
4)
(hk-test
"multi-clause: Maybe handler"
(hk-prog-val
"fromMaybe d Nothing = d\nfromMaybe _ (Just x) = x\nresult = fromMaybe 0 (Just 9)"
"result")
9)
(hk-test
"multi-clause: Maybe with default"
(hk-prog-val
"fromMaybe d Nothing = d\nfromMaybe _ (Just x) = x\nresult = fromMaybe 0 Nothing"
"result")
0)
;; ── User-defined data and matching ──
(hk-test
"custom data with pattern match"
(hk-prog-val
"data Color = Red | Green | Blue\nname Red = \"red\"\nname Green = \"green\"\nname Blue = \"blue\"\nresult = name Green"
"result")
"green")
(hk-test
"custom binary tree height"
(hk-prog-val
"data Tree = Leaf | Node Tree Tree\nh Leaf = 0\nh (Node l r) = 1 + max (h l) (h r)\nmax a b = if a > b then a else b\nresult = h (Node (Node Leaf Leaf) Leaf)"
"result")
2)
;; ── Currying ──
(hk-test
"partial application"
(hk-prog-val
"add x y = x + y\nadd5 = add 5\nresult = add5 7"
"result")
12)
;; ── Higher-order ──
(hk-test
"higher-order: function as arg"
(hk-prog-val
"twice f x = f (f x)\ninc x = x + 1\nresult = twice inc 10"
"result")
12)
;; ── Error built-in ──
(hk-test
"error short-circuits via if"
(hk-eval-expr-source
"if True then 1 else error \"unreachable\"")
1)
;; ── Laziness: app args evaluate only when forced ──
(hk-test
"second arg never forced"
(hk-eval-expr-source
"(\\x y -> x) 1 (error \"never\")")
1)
(hk-test
"first arg never forced"
(hk-eval-expr-source
"(\\x y -> y) (error \"never\") 99")
99)
(hk-test
"constructor argument is lazy under wildcard pattern"
(hk-eval-expr-source
"case Just (error \"deeply\") of Just _ -> 7 ; Nothing -> 0")
7)
(hk-test
"lazy: const drops its second argument"
(hk-prog-val
"const x y = x\nresult = const 5 (error \"boom\")"
"result")
5)
(hk-test
"lazy: head ignores tail"
(hk-prog-val
"myHead (x:_) = x\nresult = myHead (1 : (error \"tail\") : [])"
"result")
1)
(hk-test
"lazy: Just on undefined evaluates only on force"
(hk-prog-val
"wrapped = Just (error \"oh no\")\nresult = case wrapped of Just _ -> True ; Nothing -> False"
"result")
(list "True"))
;; ── not / id built-ins ──
(hk-test "not True" (hk-eval-expr-source "not True") (list "False"))
(hk-test "not False" (hk-eval-expr-source "not False") (list "True"))
(hk-test "id" (hk-eval-expr-source "id 42") 42)
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,137 @@
;; Infinite structures + Prelude tests. The lazy `:` operator builds
;; cons cells with thunked head/tail so recursive list-defining
;; functions terminate when only a finite prefix is consumed.
(define
hk-prog-val
(fn
(src name)
(hk-deep-force (get (hk-eval-program (hk-core src)) name))))
(define hk-as-list
(fn (xs)
(cond
((and (list? xs) (= (first xs) "[]")) (list))
((and (list? xs) (= (first xs) ":"))
(cons (nth xs 1) (hk-as-list (nth xs 2))))
(:else xs))))
(define
hk-eval-list
(fn (src) (hk-as-list (hk-eval-expr-source src))))
;; ── Prelude basics ──
(hk-test "head of literal" (hk-eval-expr-source "head [1, 2, 3]") 1)
(hk-test
"tail of literal"
(hk-eval-list "tail [1, 2, 3]")
(list 2 3))
(hk-test "length" (hk-eval-expr-source "length [10, 20, 30, 40]") 4)
(hk-test "length empty" (hk-eval-expr-source "length []") 0)
(hk-test
"map with section"
(hk-eval-list "map (+ 1) [1, 2, 3]")
(list 2 3 4))
(hk-test
"filter"
(hk-eval-list "filter (\\x -> x > 2) [1, 2, 3, 4, 5]")
(list 3 4 5))
(hk-test
"drop"
(hk-eval-list "drop 2 [10, 20, 30, 40]")
(list 30 40))
(hk-test "fst" (hk-eval-expr-source "fst (7, 9)") 7)
(hk-test "snd" (hk-eval-expr-source "snd (7, 9)") 9)
(hk-test
"zipWith"
(hk-eval-list "zipWith plus [1, 2, 3] [10, 20, 30]")
(list 11 22 33))
;; ── Infinite structures ──
(hk-test
"take from repeat"
(hk-eval-list "take 5 (repeat 7)")
(list 7 7 7 7 7))
(hk-test
"take 0 from repeat returns empty"
(hk-eval-list "take 0 (repeat 7)")
(list))
(hk-test
"take from iterate"
(hk-eval-list "take 5 (iterate (\\x -> x + 1) 0)")
(list 0 1 2 3 4))
(hk-test
"iterate with multiplication"
(hk-eval-list "take 4 (iterate (\\x -> x * 2) 1)")
(list 1 2 4 8))
(hk-test
"head of repeat"
(hk-eval-expr-source "head (repeat 99)")
99)
;; ── Fibonacci stream ──
(hk-test
"first 10 Fibonacci numbers"
(hk-eval-list "take 10 fibs")
(list 0 1 1 2 3 5 8 13 21 34))
(hk-test
"fib at position 8"
(hk-eval-expr-source "head (drop 8 fibs)")
21)
;; ── Building infinite structures in user code ──
(hk-test
"user-defined infinite ones"
(hk-prog-val
"ones = 1 : ones\nresult = take 6 ones"
"result")
(list ":" 1 (list ":" 1 (list ":" 1 (list ":" 1 (list ":" 1 (list ":" 1 (list "[]"))))))))
(hk-test
"user-defined nats"
(hk-prog-val
"nats = naturalsFrom 1\nnaturalsFrom n = n : naturalsFrom (n + 1)\nresult = take 5 nats"
"result")
(list ":" 1 (list ":" 2 (list ":" 3 (list ":" 4 (list ":" 5 (list "[]")))))))
;; ── Range syntax ──
(hk-test
"finite range [1..5]"
(hk-eval-list "[1..5]")
(list 1 2 3 4 5))
(hk-test
"empty range when from > to"
(hk-eval-list "[10..3]")
(list))
(hk-test
"stepped range"
(hk-eval-list "[1, 3..10]")
(list 1 3 5 7 9))
(hk-test
"open range — head"
(hk-eval-expr-source "head [1..]")
1)
(hk-test
"open range — drop then head"
(hk-eval-expr-source "head (drop 99 [1..])")
100)
(hk-test
"open range — take 5"
(hk-eval-list "take 5 [10..]")
(list 10 11 12 13 14))
;; ── Composing Prelude functions ──
(hk-test
"map then filter"
(hk-eval-list
"filter (\\x -> x > 5) (map (\\x -> x * 2) [1, 2, 3, 4])")
(list 6 8))
(hk-test
"sum-via-foldless"
(hk-prog-val
"mySum [] = 0\nmySum (x:xs) = x + mySum xs\nresult = mySum (take 5 (iterate (\\x -> x + 1) 1))"
"result")
15)
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

245
lib/haskell/tests/layout.sx Normal file
View File

@@ -0,0 +1,245 @@
;; Haskell layout-rule tests. hk-tokenizer + hk-layout produce a
;; virtual-brace-annotated stream; these tests cover the algorithm
;; from Haskell 98 §10.3 plus the pragmatic let/in single-line rule.
;; Convenience — tokenize, run layout, strip eof, keep :type/:value.
(define
hk-lay
(fn
(src)
(map
(fn (tok) {:value (get tok "value") :type (get tok "type")})
(filter
(fn (tok) (not (= (get tok "type") "eof")))
(hk-layout (hk-tokenize src))))))
;; ── 1. Basics ──
(hk-test
"empty input produces empty module { }"
(hk-lay "")
(list
{:value "{" :type "vlbrace"}
{:value "}" :type "vrbrace"}))
(hk-test
"single token → module open+close"
(hk-lay "foo")
(list
{:value "{" :type "vlbrace"}
{:value "foo" :type "varid"}
{:value "}" :type "vrbrace"}))
(hk-test
"two top-level decls get vsemi between"
(hk-lay "foo = 1\nbar = 2")
(list
{:value "{" :type "vlbrace"}
{:value "foo" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value ";" :type "vsemi"}
{:value "bar" :type "varid"}
{:value "=" :type "reservedop"}
{:value 2 :type "integer"}
{:value "}" :type "vrbrace"}))
;; ── 2. Layout keywords — do / let / where / of ──
(hk-test
"do block with two stmts"
(hk-lay "f = do\n x\n y")
(list
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "=" :type "reservedop"}
{:value "do" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value ";" :type "vsemi"}
{:value "y" :type "varid"}
{:value "}" :type "vrbrace"}
{:value "}" :type "vrbrace"}))
(hk-test
"single-line let ... in"
(hk-lay "let x = 1 in x")
(list
{:value "{" :type "vlbrace"}
{:value "let" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value "}" :type "vrbrace"}
{:value "in" :type "reserved"}
{:value "x" :type "varid"}
{:value "}" :type "vrbrace"}))
(hk-test
"where block with two bindings"
(hk-lay "f = g\n where\n g = 1\n h = 2")
(list
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "=" :type "reservedop"}
{:value "g" :type "varid"}
{:value "where" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "g" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value ";" :type "vsemi"}
{:value "h" :type "varid"}
{:value "=" :type "reservedop"}
{:value 2 :type "integer"}
{:value "}" :type "vrbrace"}
{:value "}" :type "vrbrace"}))
(hk-test
"case … of with arms"
(hk-lay "f x = case x of\n Just y -> y\n Nothing -> 0")
(list
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "x" :type "varid"}
{:value "=" :type "reservedop"}
{:value "case" :type "reserved"}
{:value "x" :type "varid"}
{:value "of" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "Just" :type "conid"}
{:value "y" :type "varid"}
{:value "->" :type "reservedop"}
{:value "y" :type "varid"}
{:value ";" :type "vsemi"}
{:value "Nothing" :type "conid"}
{:value "->" :type "reservedop"}
{:value 0 :type "integer"}
{:value "}" :type "vrbrace"}
{:value "}" :type "vrbrace"}))
;; ── 3. Explicit braces disable layout ──
(hk-test
"explicit braces — no implicit vlbrace/vsemi/vrbrace inside"
(hk-lay "do { x ; y }")
(list
{:value "{" :type "vlbrace"}
{:value "do" :type "reserved"}
{:value "{" :type "lbrace"}
{:value "x" :type "varid"}
{:value ";" :type "semi"}
{:value "y" :type "varid"}
{:value "}" :type "rbrace"}
{:value "}" :type "vrbrace"}))
;; ── 4. Dedent closes nested blocks ──
(hk-test
"dedent back to module level closes do block"
(hk-lay "f = do\n x\n y\ng = 2")
(list
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "=" :type "reservedop"}
{:value "do" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value ";" :type "vsemi"}
{:value "y" :type "varid"}
{:value "}" :type "vrbrace"}
{:value ";" :type "vsemi"}
{:value "g" :type "varid"}
{:value "=" :type "reservedop"}
{:value 2 :type "integer"}
{:value "}" :type "vrbrace"}))
(hk-test
"dedent closes inner let, emits vsemi at outer do level"
(hk-lay "main = do\n let x = 1\n print x")
(list
{:value "{" :type "vlbrace"}
{:value "main" :type "varid"}
{:value "=" :type "reservedop"}
{:value "do" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "let" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value "}" :type "vrbrace"}
{:value ";" :type "vsemi"}
{:value "print" :type "varid"}
{:value "x" :type "varid"}
{:value "}" :type "vrbrace"}
{:value "}" :type "vrbrace"}))
;; ── 5. Module header skips outer implicit open ──
(hk-test
"module M where — only where opens a block"
(hk-lay "module M where\n f = 1")
(list
{:value "module" :type "reserved"}
{:value "M" :type "conid"}
{:value "where" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value "}" :type "vrbrace"}))
;; ── 6. Newlines are stripped ──
(hk-test
"newline tokens do not appear in output"
(let
((toks (hk-layout (hk-tokenize "foo\nbar"))))
(every?
(fn (t) (not (= (get t "type") "newline")))
toks))
true)
;; ── 7. Continuation — deeper indent does NOT emit vsemi ──
(hk-test
"line continuation (deeper indent) just merges"
(hk-lay "foo = 1 +\n 2")
(list
{:value "{" :type "vlbrace"}
{:value "foo" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value "+" :type "varsym"}
{:value 2 :type "integer"}
{:value "}" :type "vrbrace"}))
;; ── 8. Stack closing at EOF ──
(hk-test
"EOF inside nested do closes all implicit blocks"
(let
((toks (hk-lay "main = do\n do\n x")))
(let
((n (len toks)))
(list
(get (nth toks (- n 1)) "type")
(get (nth toks (- n 2)) "type")
(get (nth toks (- n 3)) "type"))))
(list "vrbrace" "vrbrace" "vrbrace"))
;; ── 9. Qualified-newline: x at deeper col than stack top does nothing ──
(hk-test
"mixed where + do"
(hk-lay "f = do\n x\n where\n x = 1")
(list
{:value "{" :type "vlbrace"}
{:value "f" :type "varid"}
{:value "=" :type "reservedop"}
{:value "do" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value "}" :type "vrbrace"}
{:value "where" :type "reserved"}
{:value "{" :type "vlbrace"}
{:value "x" :type "varid"}
{:value "=" :type "reservedop"}
{:value 1 :type "integer"}
{:value "}" :type "vrbrace"}
{:value "}" :type "vrbrace"}))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

256
lib/haskell/tests/match.sx Normal file
View File

@@ -0,0 +1,256 @@
;; Pattern-matcher tests. The matcher takes (pat val env) and returns
;; an extended env dict on success, or `nil` on failure. Constructor
;; values are tagged lists (con-name first); tuples use the "Tuple"
;; tag; lists use chained `:` cons with `[]` nil.
;; ── Atomic patterns ──
(hk-test
"wildcard always matches"
(hk-match (list :p-wild) 42 (dict))
(dict))
(hk-test
"var binds value"
(hk-match (list :p-var "x") 42 (dict))
{:x 42})
(hk-test
"var preserves prior env"
(hk-match (list :p-var "y") 7 {:x 1})
{:x 1 :y 7})
(hk-test
"int literal matches equal"
(hk-match (list :p-int 5) 5 (dict))
(dict))
(hk-test
"int literal fails on mismatch"
(hk-match (list :p-int 5) 6 (dict))
nil)
(hk-test
"negative int literal matches"
(hk-match (list :p-int -3) -3 (dict))
(dict))
(hk-test
"string literal matches"
(hk-match (list :p-string "hi") "hi" (dict))
(dict))
(hk-test
"string literal fails"
(hk-match (list :p-string "hi") "bye" (dict))
nil)
(hk-test
"char literal matches"
(hk-match (list :p-char "a") "a" (dict))
(dict))
;; ── Constructor patterns ──
(hk-test
"0-arity con matches"
(hk-match
(list :p-con "Nothing" (list))
(hk-mk-con "Nothing" (list))
(dict))
(dict))
(hk-test
"1-arity con matches and binds"
(hk-match
(list :p-con "Just" (list (list :p-var "y")))
(hk-mk-con "Just" (list 9))
(dict))
{:y 9})
(hk-test
"con name mismatch fails"
(hk-match
(list :p-con "Just" (list (list :p-var "y")))
(hk-mk-con "Nothing" (list))
(dict))
nil)
(hk-test
"con arity mismatch fails"
(hk-match
(list :p-con "Pair" (list (list :p-var "a") (list :p-var "b")))
(hk-mk-con "Pair" (list 1))
(dict))
nil)
(hk-test
"nested con: Just (Just x)"
(hk-match
(list
:p-con
"Just"
(list
(list
:p-con
"Just"
(list (list :p-var "x")))))
(hk-mk-con "Just" (list (hk-mk-con "Just" (list 42))))
(dict))
{:x 42})
;; ── Tuple patterns ──
(hk-test
"2-tuple matches and binds"
(hk-match
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))
(hk-mk-tuple (list 10 20))
(dict))
{:a 10 :b 20})
(hk-test
"tuple arity mismatch fails"
(hk-match
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))
(hk-mk-tuple (list 10 20 30))
(dict))
nil)
;; ── List patterns ──
(hk-test
"[] pattern matches empty list"
(hk-match (list :p-list (list)) (hk-mk-nil) (dict))
(dict))
(hk-test
"[] pattern fails on non-empty"
(hk-match (list :p-list (list)) (hk-mk-list (list 1)) (dict))
nil)
(hk-test
"[a] pattern matches singleton"
(hk-match
(list :p-list (list (list :p-var "a")))
(hk-mk-list (list 7))
(dict))
{:a 7})
(hk-test
"[a, b] pattern matches pair-list and binds"
(hk-match
(list
:p-list
(list (list :p-var "a") (list :p-var "b")))
(hk-mk-list (list 1 2))
(dict))
{:a 1 :b 2})
(hk-test
"[a, b] fails on too-long list"
(hk-match
(list
:p-list
(list (list :p-var "a") (list :p-var "b")))
(hk-mk-list (list 1 2 3))
(dict))
nil)
;; Cons-style infix pattern (which the parser produces as :p-con ":")
(hk-test
"cons (h:t) on non-empty list"
(hk-match
(list
:p-con
":"
(list (list :p-var "h") (list :p-var "t")))
(hk-mk-list (list 1 2 3))
(dict))
{:h 1 :t (list ":" 2 (list ":" 3 (list "[]")))})
(hk-test
"cons fails on empty list"
(hk-match
(list
:p-con
":"
(list (list :p-var "h") (list :p-var "t")))
(hk-mk-nil)
(dict))
nil)
;; ── as patterns ──
(hk-test
"as binds whole + sub-pattern"
(hk-match
(list
:p-as
"all"
(list :p-con "Just" (list (list :p-var "x"))))
(hk-mk-con "Just" (list 99))
(dict))
{:all (list "Just" 99) :x 99})
(hk-test
"as on wildcard binds whole"
(hk-match
(list :p-as "v" (list :p-wild))
"anything"
(dict))
{:v "anything"})
(hk-test
"as fails when sub-pattern fails"
(hk-match
(list
:p-as
"n"
(list :p-con "Just" (list (list :p-var "x"))))
(hk-mk-con "Nothing" (list))
(dict))
nil)
;; ── lazy ~ pattern (eager equivalent for now) ──
(hk-test
"lazy pattern eager-matches its inner"
(hk-match
(list :p-lazy (list :p-var "y"))
42
(dict))
{:y 42})
;; ── Source-driven: parse a real Haskell pattern, match a value ──
(hk-test
"parsed pattern: Just x against Just 5"
(hk-match
(hk-parse-pat-source "Just x")
(hk-mk-con "Just" (list 5))
(dict))
{:x 5})
(hk-test
"parsed pattern: x : xs against [10, 20, 30]"
(hk-match
(hk-parse-pat-source "x : xs")
(hk-mk-list (list 10 20 30))
(dict))
{:x 10 :xs (list ":" 20 (list ":" 30 (list "[]")))})
(hk-test
"parsed pattern: (a, b) against (1, 2)"
(hk-match
(hk-parse-pat-source "(a, b)")
(hk-mk-tuple (list 1 2))
(dict))
{:a 1 :b 2})
(hk-test
"parsed pattern: n@(Just x) against Just 7"
(hk-match
(hk-parse-pat-source "n@(Just x)")
(hk-mk-con "Just" (list 7))
(dict))
{:n (list "Just" 7) :x 7})
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -3,60 +3,8 @@
;; Lightweight runner: each test checks actual vs expected with
;; structural (deep) equality and accumulates pass/fail counters.
;; Final value of this file is a summary dict with :pass :fail :fails.
(define
hk-deep=?
(fn
(a b)
(cond
((= a b) true)
((and (dict? a) (dict? b))
(let
((ak (keys a)) (bk (keys b)))
(if
(not (= (len ak) (len bk)))
false
(every?
(fn
(k)
(and (has-key? b k) (hk-deep=? (get a k) (get b k))))
ak))))
((and (list? a) (list? b))
(if
(not (= (len a) (len b)))
false
(let
((i 0) (ok true))
(define
hk-de-loop
(fn
()
(when
(and ok (< i (len a)))
(do
(when
(not (hk-deep=? (nth a i) (nth b i)))
(set! ok false))
(set! i (+ i 1))
(hk-de-loop)))))
(hk-de-loop)
ok)))
(:else false))))
(define hk-test-pass 0)
(define hk-test-fail 0)
(define hk-test-fails (list))
(define
hk-test
(fn
(name actual expected)
(if
(hk-deep=? actual expected)
(set! hk-test-pass (+ hk-test-pass 1))
(do
(set! hk-test-fail (+ hk-test-fail 1))
(append! hk-test-fails {:actual actual :expected expected :name name})))))
;; The hk-test / hk-deep=? helpers live in lib/haskell/testlib.sx
;; and are preloaded by lib/haskell/test.sh.
;; Convenience: tokenize and drop newline + eof tokens so tests focus
;; on meaningful content. Returns list of {:type :value} pairs.

View File

@@ -0,0 +1,278 @@
;; case-of and do-notation parser tests.
;; Covers the minimal patterns needed to make these meaningful: var,
;; wildcard, literal, constructor (with and without args), tuple, list.
;; ── Patterns (in case arms) ──
(hk-test
"wildcard pat"
(hk-parse "case x of _ -> 0")
(list
:case
(list :var "x")
(list (list :alt (list :p-wild) (list :int 0)))))
(hk-test
"var pat"
(hk-parse "case x of y -> y")
(list
:case
(list :var "x")
(list
(list :alt (list :p-var "y") (list :var "y")))))
(hk-test
"0-arity constructor pat"
(hk-parse "case x of\n Nothing -> 0\n Just y -> y")
(list
:case
(list :var "x")
(list
(list :alt (list :p-con "Nothing" (list)) (list :int 0))
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list :var "y")))))
(hk-test
"int literal pat"
(hk-parse "case n of\n 0 -> 1\n _ -> n")
(list
:case
(list :var "n")
(list
(list :alt (list :p-int 0) (list :int 1))
(list :alt (list :p-wild) (list :var "n")))))
(hk-test
"string literal pat"
(hk-parse "case s of\n \"hi\" -> 1\n _ -> 0")
(list
:case
(list :var "s")
(list
(list :alt (list :p-string "hi") (list :int 1))
(list :alt (list :p-wild) (list :int 0)))))
(hk-test
"tuple pat"
(hk-parse "case p of (a, b) -> a")
(list
:case
(list :var "p")
(list
(list
:alt
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))
(list :var "a")))))
(hk-test
"list pat"
(hk-parse "case xs of\n [] -> 0\n [a] -> a")
(list
:case
(list :var "xs")
(list
(list :alt (list :p-list (list)) (list :int 0))
(list
:alt
(list :p-list (list (list :p-var "a")))
(list :var "a")))))
(hk-test
"nested constructor pat"
(hk-parse "case x of\n Just (a, b) -> a\n _ -> 0")
(list
:case
(list :var "x")
(list
(list
:alt
(list
:p-con
"Just"
(list
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))))
(list :var "a"))
(list :alt (list :p-wild) (list :int 0)))))
(hk-test
"constructor with multiple var args"
(hk-parse "case t of Pair a b -> a")
(list
:case
(list :var "t")
(list
(list
:alt
(list
:p-con
"Pair"
(list (list :p-var "a") (list :p-var "b")))
(list :var "a")))))
;; ── case-of shapes ──
(hk-test
"case with explicit braces"
(hk-parse "case x of { Just y -> y ; Nothing -> 0 }")
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list :var "y"))
(list :alt (list :p-con "Nothing" (list)) (list :int 0)))))
(hk-test
"case scrutinee is a full expression"
(hk-parse "case f x + 1 of\n y -> y")
(list
:case
(list
:op
"+"
(list :app (list :var "f") (list :var "x"))
(list :int 1))
(list (list :alt (list :p-var "y") (list :var "y")))))
(hk-test
"case arm body is full expression"
(hk-parse "case x of\n Just y -> y + 1")
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list :op "+" (list :var "y") (list :int 1))))))
;; ── do blocks ──
(hk-test
"do with two expressions"
(hk-parse "do\n putStrLn \"hi\"\n return 0")
(list
:do
(list
(list
:do-expr
(list :app (list :var "putStrLn") (list :string "hi")))
(list
:do-expr
(list :app (list :var "return") (list :int 0))))))
(hk-test
"do with bind"
(hk-parse "do\n x <- getLine\n putStrLn x")
(list
:do
(list
(list :do-bind (list :p-var "x") (list :var "getLine"))
(list
:do-expr
(list :app (list :var "putStrLn") (list :var "x"))))))
(hk-test
"do with let"
(hk-parse "do\n let y = 5\n print y")
(list
:do
(list
(list
:do-let
(list (list :bind (list :p-var "y") (list :int 5))))
(list
:do-expr
(list :app (list :var "print") (list :var "y"))))))
(hk-test
"do with multiple let bindings"
(hk-parse "do\n let x = 1\n y = 2\n print (x + y)")
(list
:do
(list
(list
:do-let
(list
(list :bind (list :p-var "x") (list :int 1))
(list :bind (list :p-var "y") (list :int 2))))
(list
:do-expr
(list
:app
(list :var "print")
(list :op "+" (list :var "x") (list :var "y")))))))
(hk-test
"do with bind using constructor pat"
(hk-parse "do\n Just x <- getMaybe\n return x")
(list
:do
(list
(list
:do-bind
(list :p-con "Just" (list (list :p-var "x")))
(list :var "getMaybe"))
(list
:do-expr
(list :app (list :var "return") (list :var "x"))))))
(hk-test
"do with explicit braces"
(hk-parse "do { x <- a ; y <- b ; return (x + y) }")
(list
:do
(list
(list :do-bind (list :p-var "x") (list :var "a"))
(list :do-bind (list :p-var "y") (list :var "b"))
(list
:do-expr
(list
:app
(list :var "return")
(list :op "+" (list :var "x") (list :var "y")))))))
;; ── Mixing case/do inside expressions ──
(hk-test
"case inside let"
(hk-parse "let f = \\x -> case x of\n Just y -> y\n _ -> 0\nin f 5")
(list
:let
(list
(list
:bind
(list :p-var "f")
(list
:lambda
(list (list :p-var "x"))
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list :var "y"))
(list :alt (list :p-wild) (list :int 0)))))))
(list :app (list :var "f") (list :int 5))))
(hk-test
"lambda containing do"
(hk-parse "\\x -> do\n y <- x\n return y")
(list
:lambda
(list (list :p-var "x"))
(list
:do
(list
(list :do-bind (list :p-var "y") (list :var "x"))
(list
:do-expr
(list :app (list :var "return") (list :var "y")))))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,273 @@
;; Top-level declarations: function clauses, type signatures, data,
;; type, newtype, fixity. Driven by hk-parse-top which produces
;; a (:program DECLS) node.
(define
hk-prog
(fn
(&rest decls)
(list :program decls)))
;; ── Function clauses & pattern bindings ──
(hk-test
"simple fun-clause"
(hk-parse-top "f x = x + 1")
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list :op "+" (list :var "x") (list :int 1)))))
(hk-test
"nullary decl"
(hk-parse-top "answer = 42")
(hk-prog
(list :fun-clause "answer" (list) (list :int 42))))
(hk-test
"multi-clause fn (separate defs for each pattern)"
(hk-parse-top "fact 0 = 1\nfact n = n")
(hk-prog
(list :fun-clause "fact" (list (list :p-int 0)) (list :int 1))
(list
:fun-clause
"fact"
(list (list :p-var "n"))
(list :var "n"))))
(hk-test
"constructor pattern in fn args"
(hk-parse-top "fromJust (Just x) = x")
(hk-prog
(list
:fun-clause
"fromJust"
(list (list :p-con "Just" (list (list :p-var "x"))))
(list :var "x"))))
(hk-test
"pattern binding at top level"
(hk-parse-top "(a, b) = pair")
(hk-prog
(list
:pat-bind
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))
(list :var "pair"))))
;; ── Type signatures ──
(hk-test
"single-name sig"
(hk-parse-top "f :: Int -> Int")
(hk-prog
(list
:type-sig
(list "f")
(list :t-fun (list :t-con "Int") (list :t-con "Int")))))
(hk-test
"multi-name sig"
(hk-parse-top "f, g, h :: Int -> Bool")
(hk-prog
(list
:type-sig
(list "f" "g" "h")
(list :t-fun (list :t-con "Int") (list :t-con "Bool")))))
(hk-test
"sig with type application"
(hk-parse-top "f :: Maybe a -> a")
(hk-prog
(list
:type-sig
(list "f")
(list
:t-fun
(list :t-app (list :t-con "Maybe") (list :t-var "a"))
(list :t-var "a")))))
(hk-test
"sig with list type"
(hk-parse-top "len :: [a] -> Int")
(hk-prog
(list
:type-sig
(list "len")
(list
:t-fun
(list :t-list (list :t-var "a"))
(list :t-con "Int")))))
(hk-test
"sig with tuple and right-assoc ->"
(hk-parse-top "pair :: a -> b -> (a, b)")
(hk-prog
(list
:type-sig
(list "pair")
(list
:t-fun
(list :t-var "a")
(list
:t-fun
(list :t-var "b")
(list
:t-tuple
(list (list :t-var "a") (list :t-var "b"))))))))
(hk-test
"sig + implementation together"
(hk-parse-top "id :: a -> a\nid x = x")
(hk-prog
(list
:type-sig
(list "id")
(list :t-fun (list :t-var "a") (list :t-var "a")))
(list
:fun-clause
"id"
(list (list :p-var "x"))
(list :var "x"))))
;; ── data declarations ──
(hk-test
"data Maybe"
(hk-parse-top "data Maybe a = Nothing | Just a")
(hk-prog
(list
:data
"Maybe"
(list "a")
(list
(list :con-def "Nothing" (list))
(list :con-def "Just" (list (list :t-var "a")))))))
(hk-test
"data Either"
(hk-parse-top "data Either a b = Left a | Right b")
(hk-prog
(list
:data
"Either"
(list "a" "b")
(list
(list :con-def "Left" (list (list :t-var "a")))
(list :con-def "Right" (list (list :t-var "b")))))))
(hk-test
"data with no type parameters"
(hk-parse-top "data Bool = True | False")
(hk-prog
(list
:data
"Bool"
(list)
(list
(list :con-def "True" (list))
(list :con-def "False" (list))))))
(hk-test
"recursive data type"
(hk-parse-top "data Tree a = Leaf | Node (Tree a) a (Tree a)")
(hk-prog
(list
:data
"Tree"
(list "a")
(list
(list :con-def "Leaf" (list))
(list
:con-def
"Node"
(list
(list :t-app (list :t-con "Tree") (list :t-var "a"))
(list :t-var "a")
(list :t-app (list :t-con "Tree") (list :t-var "a"))))))))
;; ── type synonyms ──
(hk-test
"simple type synonym"
(hk-parse-top "type Name = String")
(hk-prog
(list :type-syn "Name" (list) (list :t-con "String"))))
(hk-test
"parameterised type synonym"
(hk-parse-top "type Pair a = (a, a)")
(hk-prog
(list
:type-syn
"Pair"
(list "a")
(list
:t-tuple
(list (list :t-var "a") (list :t-var "a"))))))
;; ── newtype ──
(hk-test
"newtype"
(hk-parse-top "newtype Age = Age Int")
(hk-prog (list :newtype "Age" (list) "Age" (list :t-con "Int"))))
(hk-test
"parameterised newtype"
(hk-parse-top "newtype Wrap a = Wrap a")
(hk-prog
(list :newtype "Wrap" (list "a") "Wrap" (list :t-var "a"))))
;; ── fixity declarations ──
(hk-test
"infixl with precedence"
(hk-parse-top "infixl 5 +:, -:")
(hk-prog (list :fixity "l" 5 (list "+:" "-:"))))
(hk-test
"infixr"
(hk-parse-top "infixr 9 .")
(hk-prog (list :fixity "r" 9 (list "."))))
(hk-test
"infix (non-assoc) default prec"
(hk-parse-top "infix ==")
(hk-prog (list :fixity "n" 9 (list "=="))))
(hk-test
"fixity with backtick operator name"
(hk-parse-top "infixl 7 `div`")
(hk-prog (list :fixity "l" 7 (list "div"))))
;; ── Several decls combined ──
(hk-test
"mixed: data + sig + fn + type"
(hk-parse-top "data Maybe a = Nothing | Just a\ntype Entry = Maybe Int\nf :: Entry -> Int\nf (Just x) = x\nf Nothing = 0")
(hk-prog
(list
:data
"Maybe"
(list "a")
(list
(list :con-def "Nothing" (list))
(list :con-def "Just" (list (list :t-var "a")))))
(list
:type-syn
"Entry"
(list)
(list :t-app (list :t-con "Maybe") (list :t-con "Int")))
(list
:type-sig
(list "f")
(list :t-fun (list :t-con "Entry") (list :t-con "Int")))
(list
:fun-clause
"f"
(list (list :p-con "Just" (list (list :p-var "x"))))
(list :var "x"))
(list
:fun-clause
"f"
(list (list :p-con "Nothing" (list)))
(list :int 0))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,258 @@
;; Haskell expression parser tests.
;; hk-parse tokenises, runs layout, then parses. Output is an AST
;; whose head is a keyword tag (evaluates to its string name).
;; ── 1. Literals ──
(hk-test "integer" (hk-parse "42") (list :int 42))
(hk-test "float" (hk-parse "3.14") (list :float 3.14))
(hk-test "string" (hk-parse "\"hi\"") (list :string "hi"))
(hk-test "char" (hk-parse "'a'") (list :char "a"))
;; ── 2. Variables and constructors ──
(hk-test "varid" (hk-parse "foo") (list :var "foo"))
(hk-test "conid" (hk-parse "Nothing") (list :con "Nothing"))
(hk-test "qvarid" (hk-parse "Data.Map.lookup") (list :var "Data.Map.lookup"))
(hk-test "qconid" (hk-parse "Data.Map") (list :con "Data.Map"))
;; ── 3. Parens / unit / tuple ──
(hk-test "parens strip" (hk-parse "(42)") (list :int 42))
(hk-test "unit" (hk-parse "()") (list :con "()"))
(hk-test
"2-tuple"
(hk-parse "(1, 2)")
(list :tuple (list (list :int 1) (list :int 2))))
(hk-test
"3-tuple"
(hk-parse "(x, y, z)")
(list
:tuple
(list (list :var "x") (list :var "y") (list :var "z"))))
;; ── 4. Lists ──
(hk-test "empty list" (hk-parse "[]") (list :list (list)))
(hk-test
"singleton list"
(hk-parse "[1]")
(list :list (list (list :int 1))))
(hk-test
"list of ints"
(hk-parse "[1, 2, 3]")
(list
:list
(list (list :int 1) (list :int 2) (list :int 3))))
(hk-test
"range"
(hk-parse "[1..10]")
(list :range (list :int 1) (list :int 10)))
(hk-test
"range with step"
(hk-parse "[1, 3..10]")
(list
:range-step
(list :int 1)
(list :int 3)
(list :int 10)))
;; ── 5. Application ──
(hk-test
"one-arg app"
(hk-parse "f x")
(list :app (list :var "f") (list :var "x")))
(hk-test
"multi-arg app is left-assoc"
(hk-parse "f x y z")
(list
:app
(list
:app
(list :app (list :var "f") (list :var "x"))
(list :var "y"))
(list :var "z")))
(hk-test
"app with con"
(hk-parse "Just 5")
(list :app (list :con "Just") (list :int 5)))
;; ── 6. Infix operators ──
(hk-test
"simple +"
(hk-parse "1 + 2")
(list :op "+" (list :int 1) (list :int 2)))
(hk-test
"precedence: * binds tighter than +"
(hk-parse "1 + 2 * 3")
(list
:op
"+"
(list :int 1)
(list :op "*" (list :int 2) (list :int 3))))
(hk-test
"- is left-assoc"
(hk-parse "10 - 3 - 2")
(list
:op
"-"
(list :op "-" (list :int 10) (list :int 3))
(list :int 2)))
(hk-test
": is right-assoc"
(hk-parse "a : b : c")
(list
:op
":"
(list :var "a")
(list :op ":" (list :var "b") (list :var "c"))))
(hk-test
"app binds tighter than op"
(hk-parse "f x + g y")
(list
:op
"+"
(list :app (list :var "f") (list :var "x"))
(list :app (list :var "g") (list :var "y"))))
(hk-test
"$ is lowest precedence, right-assoc"
(hk-parse "f $ g x")
(list
:op
"$"
(list :var "f")
(list :app (list :var "g") (list :var "x"))))
;; ── 7. Backticks (varid-as-operator) ──
(hk-test
"backtick operator"
(hk-parse "x `mod` 3")
(list :op "mod" (list :var "x") (list :int 3)))
;; ── 8. Unary negation ──
(hk-test
"unary -"
(hk-parse "- 5")
(list :neg (list :int 5)))
(hk-test
"unary - on application"
(hk-parse "- f x")
(list :neg (list :app (list :var "f") (list :var "x"))))
(hk-test
"- n + m → (- n) + m"
(hk-parse "- 1 + 2")
(list
:op
"+"
(list :neg (list :int 1))
(list :int 2)))
;; ── 9. Lambda ──
(hk-test
"lambda single param"
(hk-parse "\\x -> x")
(list :lambda (list (list :p-var "x")) (list :var "x")))
(hk-test
"lambda multi-param"
(hk-parse "\\x y -> x + y")
(list
:lambda
(list (list :p-var "x") (list :p-var "y"))
(list :op "+" (list :var "x") (list :var "y"))))
(hk-test
"lambda body is full expression"
(hk-parse "\\f -> f 1 + f 2")
(list
:lambda
(list (list :p-var "f"))
(list
:op
"+"
(list :app (list :var "f") (list :int 1))
(list :app (list :var "f") (list :int 2)))))
;; ── 10. if-then-else ──
(hk-test
"if basic"
(hk-parse "if x then 1 else 2")
(list :if (list :var "x") (list :int 1) (list :int 2)))
(hk-test
"if with infix cond"
(hk-parse "if x == 0 then y else z")
(list
:if
(list :op "==" (list :var "x") (list :int 0))
(list :var "y")
(list :var "z")))
;; ── 11. let-in ──
(hk-test
"let single binding"
(hk-parse "let x = 1 in x")
(list
:let
(list (list :bind (list :p-var "x") (list :int 1)))
(list :var "x")))
(hk-test
"let two bindings (multi-line)"
(hk-parse "let x = 1\n y = 2\nin x + y")
(list
:let
(list
(list :bind (list :p-var "x") (list :int 1))
(list :bind (list :p-var "y") (list :int 2)))
(list :op "+" (list :var "x") (list :var "y"))))
(hk-test
"let with explicit braces"
(hk-parse "let { x = 1 ; y = 2 } in x + y")
(list
:let
(list
(list :bind (list :p-var "x") (list :int 1))
(list :bind (list :p-var "y") (list :int 2)))
(list :op "+" (list :var "x") (list :var "y"))))
;; ── 12. Mixed / nesting ──
(hk-test
"nested application"
(hk-parse "f (g x) y")
(list
:app
(list
:app
(list :var "f")
(list :app (list :var "g") (list :var "x")))
(list :var "y")))
(hk-test
"lambda applied"
(hk-parse "(\\x -> x + 1) 5")
(list
:app
(list
:lambda
(list (list :p-var "x"))
(list :op "+" (list :var "x") (list :int 1)))
(list :int 5)))
(hk-test
"lambda + if"
(hk-parse "\\n -> if n == 0 then 1 else n")
(list
:lambda
(list (list :p-var "n"))
(list
:if
(list :op "==" (list :var "n") (list :int 0))
(list :int 1)
(list :var "n"))))
;; ── 13. Precedence corners ──
(hk-test
". is right-assoc (prec 9)"
(hk-parse "f . g . h")
(list
:op
"."
(list :var "f")
(list :op "." (list :var "g") (list :var "h"))))
(hk-test
"== is non-associative (single use)"
(hk-parse "x == y")
(list :op "==" (list :var "x") (list :var "y")))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,261 @@
;; Guards and where-clauses — on fun-clauses, case alts, and
;; let-bindings (which now also accept funclause-style LHS like
;; `let f x = e` or `let f x | g = e | g = e`).
(define
hk-prog
(fn (&rest decls) (list :program decls)))
;; ── Guarded fun-clauses ──
(hk-test
"simple guards (two branches)"
(hk-parse-top "abs x | x < 0 = - x\n | otherwise = x")
(hk-prog
(list
:fun-clause
"abs"
(list (list :p-var "x"))
(list
:guarded
(list
(list
:guard
(list :op "<" (list :var "x") (list :int 0))
(list :neg (list :var "x")))
(list :guard (list :var "otherwise") (list :var "x")))))))
(hk-test
"three-way guard"
(hk-parse-top "sign n | n > 0 = 1\n | n < 0 = -1\n | otherwise = 0")
(hk-prog
(list
:fun-clause
"sign"
(list (list :p-var "n"))
(list
:guarded
(list
(list
:guard
(list :op ">" (list :var "n") (list :int 0))
(list :int 1))
(list
:guard
(list :op "<" (list :var "n") (list :int 0))
(list :neg (list :int 1)))
(list
:guard
(list :var "otherwise")
(list :int 0)))))))
(hk-test
"mixed: one eq clause plus one guarded clause"
(hk-parse-top "sign 0 = 0\nsign n | n > 0 = 1\n | otherwise = -1")
(hk-prog
(list
:fun-clause
"sign"
(list (list :p-int 0))
(list :int 0))
(list
:fun-clause
"sign"
(list (list :p-var "n"))
(list
:guarded
(list
(list
:guard
(list :op ">" (list :var "n") (list :int 0))
(list :int 1))
(list
:guard
(list :var "otherwise")
(list :neg (list :int 1))))))))
;; ── where on fun-clauses ──
(hk-test
"where with one binding"
(hk-parse-top "f x = y + y\n where y = x + 1")
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:where
(list :op "+" (list :var "y") (list :var "y"))
(list
(list
:fun-clause
"y"
(list)
(list :op "+" (list :var "x") (list :int 1))))))))
(hk-test
"where with multiple bindings"
(hk-parse-top "f x = y * z\n where y = x + 1\n z = x - 1")
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:where
(list :op "*" (list :var "y") (list :var "z"))
(list
(list
:fun-clause
"y"
(list)
(list :op "+" (list :var "x") (list :int 1)))
(list
:fun-clause
"z"
(list)
(list :op "-" (list :var "x") (list :int 1))))))))
(hk-test
"guards + where"
(hk-parse-top "f x | x > 0 = y\n | otherwise = 0\n where y = 99")
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:where
(list
:guarded
(list
(list
:guard
(list :op ">" (list :var "x") (list :int 0))
(list :var "y"))
(list
:guard
(list :var "otherwise")
(list :int 0))))
(list
(list :fun-clause "y" (list) (list :int 99)))))))
;; ── Guards in case alts ──
(hk-test
"case alt with guards"
(hk-parse "case x of\n Just y | y > 0 -> y\n | otherwise -> 0\n Nothing -> 0")
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list
:guarded
(list
(list
:guard
(list :op ">" (list :var "y") (list :int 0))
(list :var "y"))
(list
:guard
(list :var "otherwise")
(list :int 0)))))
(list :alt (list :p-con "Nothing" (list)) (list :int 0)))))
(hk-test
"case alt with where"
(hk-parse "case x of\n Just y -> y + z where z = 5\n Nothing -> 0")
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-con "Just" (list (list :p-var "y")))
(list
:where
(list :op "+" (list :var "y") (list :var "z"))
(list
(list :fun-clause "z" (list) (list :int 5)))))
(list :alt (list :p-con "Nothing" (list)) (list :int 0)))))
;; ── let-bindings: funclause form, guards, where ──
(hk-test
"let with funclause shorthand"
(hk-parse "let f x = x + 1 in f 5")
(list
:let
(list
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list :op "+" (list :var "x") (list :int 1))))
(list :app (list :var "f") (list :int 5))))
(hk-test
"let with guards"
(hk-parse "let f x | x > 0 = x\n | otherwise = 0\nin f 3")
(list
:let
(list
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:guarded
(list
(list
:guard
(list :op ">" (list :var "x") (list :int 0))
(list :var "x"))
(list
:guard
(list :var "otherwise")
(list :int 0))))))
(list :app (list :var "f") (list :int 3))))
(hk-test
"let funclause + where"
(hk-parse "let f x = y where y = x + 1\nin f 7")
(list
:let
(list
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:where
(list :var "y")
(list
(list
:fun-clause
"y"
(list)
(list :op "+" (list :var "x") (list :int 1)))))))
(list :app (list :var "f") (list :int 7))))
;; ── Nested: where inside where (via recursive hk-parse-decl) ──
(hk-test
"where block can contain a type signature"
(hk-parse-top "f x = y\n where y :: Int\n y = x")
(hk-prog
(list
:fun-clause
"f"
(list (list :p-var "x"))
(list
:where
(list :var "y")
(list
(list :type-sig (list "y") (list :t-con "Int"))
(list
:fun-clause
"y"
(list)
(list :var "x")))))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,202 @@
;; Module header + imports. The parser switches from (:program DECLS)
;; to (:module NAME EXPORTS IMPORTS DECLS) as soon as a module header
;; or any `import` decl appears.
;; ── Module header ──
(hk-test
"simple module, no exports"
(hk-parse-top "module M where\n f = 1")
(list
:module
"M"
nil
(list)
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"module with dotted name"
(hk-parse-top "module Data.Map where\nf = 1")
(list
:module
"Data.Map"
nil
(list)
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"module with empty export list"
(hk-parse-top "module M () where\nf = 1")
(list
:module
"M"
(list)
(list)
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"module with exports (var, tycon-all, tycon-with)"
(hk-parse-top "module M (f, g, Maybe(..), List(Cons, Nil)) where\nf = 1\ng = 2")
(list
:module
"M"
(list
(list :ent-var "f")
(list :ent-var "g")
(list :ent-all "Maybe")
(list :ent-with "List" (list "Cons" "Nil")))
(list)
(list
(list :fun-clause "f" (list) (list :int 1))
(list :fun-clause "g" (list) (list :int 2)))))
(hk-test
"module export list including another module"
(hk-parse-top "module M (module Foo, f) where\nf = 1")
(list
:module
"M"
(list (list :ent-module "Foo") (list :ent-var "f"))
(list)
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"module export with operator"
(hk-parse-top "module M ((+:), f) where\nf = 1")
(list
:module
"M"
(list (list :ent-var "+:") (list :ent-var "f"))
(list)
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"empty module body"
(hk-parse-top "module M where")
(list :module "M" nil (list) (list)))
;; ── Imports ──
(hk-test
"plain import"
(hk-parse-top "import Foo")
(list
:module
nil
nil
(list (list :import false "Foo" nil nil))
(list)))
(hk-test
"qualified import"
(hk-parse-top "import qualified Data.Map")
(list
:module
nil
nil
(list (list :import true "Data.Map" nil nil))
(list)))
(hk-test
"import with alias"
(hk-parse-top "import Data.Map as M")
(list
:module
nil
nil
(list (list :import false "Data.Map" "M" nil))
(list)))
(hk-test
"import with explicit list"
(hk-parse-top "import Foo (bar, Baz(..), Quux(X, Y))")
(list
:module
nil
nil
(list
(list
:import
false
"Foo"
nil
(list
:spec-items
(list
(list :ent-var "bar")
(list :ent-all "Baz")
(list :ent-with "Quux" (list "X" "Y"))))))
(list)))
(hk-test
"import hiding"
(hk-parse-top "import Foo hiding (x, y)")
(list
:module
nil
nil
(list
(list
:import
false
"Foo"
nil
(list
:spec-hiding
(list (list :ent-var "x") (list :ent-var "y")))))
(list)))
(hk-test
"qualified + alias + hiding"
(hk-parse-top "import qualified Data.List as L hiding (sort)")
(list
:module
nil
nil
(list
(list
:import
true
"Data.List"
"L"
(list :spec-hiding (list (list :ent-var "sort")))))
(list)))
;; ── Combinations ──
(hk-test
"module with multiple imports and a decl"
(hk-parse-top "module M where\nimport Foo\nimport qualified Bar as B\nf = 1")
(list
:module
"M"
nil
(list
(list :import false "Foo" nil nil)
(list :import true "Bar" "B" nil))
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"headerless file with imports"
(hk-parse-top "import Foo\nimport Bar (baz)\nf = 1")
(list
:module
nil
nil
(list
(list :import false "Foo" nil nil)
(list
:import
false
"Bar"
nil
(list :spec-items (list (list :ent-var "baz")))))
(list (list :fun-clause "f" (list) (list :int 1)))))
(hk-test
"plain program (no header, no imports) still uses :program"
(hk-parse-top "f = 1\ng = 2")
(list
:program
(list
(list :fun-clause "f" (list) (list :int 1))
(list :fun-clause "g" (list) (list :int 2)))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,234 @@
;; Full-pattern parser tests: as-patterns, lazy ~, negative literals,
;; infix constructor patterns (`:`, any consym), lambda pattern args,
;; and let pattern-bindings.
;; ── as-patterns ──
(hk-test
"as pattern, wraps constructor"
(hk-parse "case x of n@(Just y) -> n")
(list
:case
(list :var "x")
(list
(list
:alt
(list
:p-as
"n"
(list :p-con "Just" (list (list :p-var "y"))))
(list :var "n")))))
(hk-test
"as pattern, wraps wildcard"
(hk-parse "case x of all@_ -> all")
(list
:case
(list :var "x")
(list
(list
:alt
(list :p-as "all" (list :p-wild))
(list :var "all")))))
(hk-test
"as in lambda"
(hk-parse "\\xs@(a : rest) -> xs")
(list
:lambda
(list
(list
:p-as
"xs"
(list
:p-con
":"
(list (list :p-var "a") (list :p-var "rest")))))
(list :var "xs")))
;; ── lazy patterns ──
(hk-test
"lazy var"
(hk-parse "case x of ~y -> y")
(list
:case
(list :var "x")
(list
(list :alt (list :p-lazy (list :p-var "y")) (list :var "y")))))
(hk-test
"lazy constructor"
(hk-parse "\\(~(Just x)) -> x")
(list
:lambda
(list
(list
:p-lazy
(list :p-con "Just" (list (list :p-var "x")))))
(list :var "x")))
;; ── negative literal patterns ──
(hk-test
"negative int pattern"
(hk-parse "case n of\n -1 -> 0\n _ -> n")
(list
:case
(list :var "n")
(list
(list :alt (list :p-int -1) (list :int 0))
(list :alt (list :p-wild) (list :var "n")))))
(hk-test
"negative float pattern"
(hk-parse "case x of -0.5 -> 1")
(list
:case
(list :var "x")
(list (list :alt (list :p-float -0.5) (list :int 1)))))
;; ── infix constructor patterns (`:` and any consym) ──
(hk-test
"cons pattern"
(hk-parse "case xs of x : rest -> x")
(list
:case
(list :var "xs")
(list
(list
:alt
(list
:p-con
":"
(list (list :p-var "x") (list :p-var "rest")))
(list :var "x")))))
(hk-test
"cons is right-associative in pats"
(hk-parse "case xs of a : b : rest -> rest")
(list
:case
(list :var "xs")
(list
(list
:alt
(list
:p-con
":"
(list
(list :p-var "a")
(list
:p-con
":"
(list (list :p-var "b") (list :p-var "rest")))))
(list :var "rest")))))
(hk-test
"consym pattern"
(hk-parse "case p of a :+: b -> a")
(list
:case
(list :var "p")
(list
(list
:alt
(list
:p-con
":+:"
(list (list :p-var "a") (list :p-var "b")))
(list :var "a")))))
;; ── lambda with pattern args ──
(hk-test
"lambda with constructor pattern"
(hk-parse "\\(Just x) -> x")
(list
:lambda
(list (list :p-con "Just" (list (list :p-var "x"))))
(list :var "x")))
(hk-test
"lambda with tuple pattern"
(hk-parse "\\(a, b) -> a + b")
(list
:lambda
(list
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b"))))
(list :op "+" (list :var "a") (list :var "b"))))
(hk-test
"lambda with wildcard"
(hk-parse "\\_ -> 42")
(list :lambda (list (list :p-wild)) (list :int 42)))
(hk-test
"lambda with mixed apats"
(hk-parse "\\x _ (Just y) -> y")
(list
:lambda
(list
(list :p-var "x")
(list :p-wild)
(list :p-con "Just" (list (list :p-var "y"))))
(list :var "y")))
;; ── let pattern-bindings ──
(hk-test
"let tuple pattern-binding"
(hk-parse "let (x, y) = pair in x + y")
(list
:let
(list
(list
:bind
(list
:p-tuple
(list (list :p-var "x") (list :p-var "y")))
(list :var "pair")))
(list :op "+" (list :var "x") (list :var "y"))))
(hk-test
"let constructor pattern-binding"
(hk-parse "let Just x = m in x")
(list
:let
(list
(list
:bind
(list :p-con "Just" (list (list :p-var "x")))
(list :var "m")))
(list :var "x")))
(hk-test
"let cons pattern-binding"
(hk-parse "let (x : rest) = xs in x")
(list
:let
(list
(list
:bind
(list
:p-con
":"
(list (list :p-var "x") (list :p-var "rest")))
(list :var "xs")))
(list :var "x")))
;; ── do with constructor-pattern binds ──
(hk-test
"do bind to tuple pattern"
(hk-parse "do\n (a, b) <- pairs\n return a")
(list
:do
(list
(list
:do-bind
(list
:p-tuple
(list (list :p-var "a") (list :p-var "b")))
(list :var "pairs"))
(list
:do-expr
(list :app (list :var "return") (list :var "a"))))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,191 @@
;; Operator sections and list comprehensions.
;; ── Operator references (unchanged expr shape) ──
(hk-test
"op as value (+)"
(hk-parse "(+)")
(list :var "+"))
(hk-test
"op as value (-)"
(hk-parse "(-)")
(list :var "-"))
(hk-test
"op as value (:)"
(hk-parse "(:)")
(list :var ":"))
(hk-test
"backtick op as value"
(hk-parse "(`div`)")
(list :var "div"))
;; ── Right sections (op expr) ──
(hk-test
"right section (+ 5)"
(hk-parse "(+ 5)")
(list :sect-right "+" (list :int 5)))
(hk-test
"right section (* x)"
(hk-parse "(* x)")
(list :sect-right "*" (list :var "x")))
(hk-test
"right section with backtick op"
(hk-parse "(`div` 2)")
(list :sect-right "div" (list :int 2)))
;; `-` is unary in expr position — (- 5) is negation, not a right section
(hk-test
"(- 5) is negation, not a section"
(hk-parse "(- 5)")
(list :neg (list :int 5)))
;; ── Left sections (expr op) ──
(hk-test
"left section (5 +)"
(hk-parse "(5 +)")
(list :sect-left "+" (list :int 5)))
(hk-test
"left section with backtick"
(hk-parse "(x `mod`)")
(list :sect-left "mod" (list :var "x")))
(hk-test
"left section with cons (x :)"
(hk-parse "(x :)")
(list :sect-left ":" (list :var "x")))
;; ── Mixed / nesting ──
(hk-test
"map (+ 1) xs"
(hk-parse "map (+ 1) xs")
(list
:app
(list
:app
(list :var "map")
(list :sect-right "+" (list :int 1)))
(list :var "xs")))
(hk-test
"filter (< 0) xs"
(hk-parse "filter (< 0) xs")
(list
:app
(list
:app
(list :var "filter")
(list :sect-right "<" (list :int 0)))
(list :var "xs")))
;; ── Plain parens and tuples still work ──
(hk-test
"plain parens unwrap"
(hk-parse "(1 + 2)")
(list :op "+" (list :int 1) (list :int 2)))
(hk-test
"tuple still parses"
(hk-parse "(a, b, c)")
(list
:tuple
(list (list :var "a") (list :var "b") (list :var "c"))))
;; ── List comprehensions ──
(hk-test
"simple list comprehension"
(hk-parse "[x | x <- xs]")
(list
:list-comp
(list :var "x")
(list
(list :q-gen (list :p-var "x") (list :var "xs")))))
(hk-test
"comprehension with filter"
(hk-parse "[x * 2 | x <- xs, x > 0]")
(list
:list-comp
(list :op "*" (list :var "x") (list :int 2))
(list
(list :q-gen (list :p-var "x") (list :var "xs"))
(list
:q-guard
(list :op ">" (list :var "x") (list :int 0))))))
(hk-test
"comprehension with let"
(hk-parse "[y | x <- xs, let y = x + 1]")
(list
:list-comp
(list :var "y")
(list
(list :q-gen (list :p-var "x") (list :var "xs"))
(list
:q-let
(list
(list
:bind
(list :p-var "y")
(list :op "+" (list :var "x") (list :int 1))))))))
(hk-test
"nested generators"
(hk-parse "[(x, y) | x <- xs, y <- ys]")
(list
:list-comp
(list :tuple (list (list :var "x") (list :var "y")))
(list
(list :q-gen (list :p-var "x") (list :var "xs"))
(list :q-gen (list :p-var "y") (list :var "ys")))))
(hk-test
"comprehension with constructor pattern"
(hk-parse "[v | Just v <- xs]")
(list
:list-comp
(list :var "v")
(list
(list
:q-gen
(list :p-con "Just" (list (list :p-var "v")))
(list :var "xs")))))
(hk-test
"comprehension with tuple pattern"
(hk-parse "[x + y | (x, y) <- pairs]")
(list
:list-comp
(list :op "+" (list :var "x") (list :var "y"))
(list
(list
:q-gen
(list
:p-tuple
(list (list :p-var "x") (list :p-var "y")))
(list :var "pairs")))))
(hk-test
"combination: generator, let, guard"
(hk-parse "[z | x <- xs, let z = x * 2, z > 10]")
(list
:list-comp
(list :var "z")
(list
(list :q-gen (list :p-var "x") (list :var "xs"))
(list
:q-let
(list
(list
:bind
(list :p-var "z")
(list :op "*" (list :var "x") (list :int 2)))))
(list
:q-guard
(list :op ">" (list :var "z") (list :int 10))))))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -0,0 +1,127 @@
;; Runtime constructor-registry tests. Built-ins are pre-registered
;; when lib/haskell/runtime.sx loads; user types are registered by
;; walking a parsed+desugared AST with hk-register-program! (or the
;; `hk-load-source!` convenience).
;; ── Pre-registered built-ins ──
(hk-test "True is a con" (hk-is-con? "True") true)
(hk-test "False is a con" (hk-is-con? "False") true)
(hk-test "[] is a con" (hk-is-con? "[]") true)
(hk-test ": (cons) is a con" (hk-is-con? ":") true)
(hk-test "() is a con" (hk-is-con? "()") true)
(hk-test "True arity 0" (hk-con-arity "True") 0)
(hk-test ": arity 2" (hk-con-arity ":") 2)
(hk-test "[] arity 0" (hk-con-arity "[]") 0)
(hk-test "True type Bool" (hk-con-type "True") "Bool")
(hk-test "False type Bool" (hk-con-type "False") "Bool")
(hk-test ": type List" (hk-con-type ":") "List")
(hk-test "() type Unit" (hk-con-type "()") "Unit")
;; ── Unknown names ──
(hk-test "is-con? false for varid" (hk-is-con? "foo") false)
(hk-test "arity nil for unknown" (hk-con-arity "NotACon") nil)
(hk-test "type nil for unknown" (hk-con-type "NotACon") nil)
;; ── data MyBool = Yes | No ──
(hk-test
"register simple data"
(do
(hk-load-source! "data MyBool = Yes | No")
(list
(hk-con-arity "Yes")
(hk-con-arity "No")
(hk-con-type "Yes")
(hk-con-type "No")))
(list 0 0 "MyBool" "MyBool"))
;; ── data Maybe a = Nothing | Just a ──
(hk-test
"register Maybe"
(do
(hk-load-source! "data Maybe a = Nothing | Just a")
(list
(hk-con-arity "Nothing")
(hk-con-arity "Just")
(hk-con-type "Nothing")
(hk-con-type "Just")))
(list 0 1 "Maybe" "Maybe"))
;; ── data Either a b = Left a | Right b ──
(hk-test
"register Either"
(do
(hk-load-source! "data Either a b = Left a | Right b")
(list
(hk-con-arity "Left")
(hk-con-arity "Right")
(hk-con-type "Left")
(hk-con-type "Right")))
(list 1 1 "Either" "Either"))
;; ── Recursive data ──
(hk-test
"register recursive Tree"
(do
(hk-load-source!
"data Tree a = Leaf | Node (Tree a) a (Tree a)")
(list
(hk-con-arity "Leaf")
(hk-con-arity "Node")
(hk-con-type "Leaf")
(hk-con-type "Node")))
(list 0 3 "Tree" "Tree"))
;; ── newtype ──
(hk-test
"register newtype"
(do
(hk-load-source! "newtype Age = MkAge Int")
(list
(hk-con-arity "MkAge")
(hk-con-type "MkAge")))
(list 1 "Age"))
;; ── Multiple data decls in one program ──
(hk-test
"multiple data decls"
(do
(hk-load-source!
"data Color = Red | Green | Blue\ndata Shape = Circle | Square\nf x = x")
(list
(hk-con-type "Red")
(hk-con-type "Green")
(hk-con-type "Blue")
(hk-con-type "Circle")
(hk-con-type "Square")))
(list "Color" "Color" "Color" "Shape" "Shape"))
;; ── Inside a module header ──
(hk-test
"register from module body"
(do
(hk-load-source!
"module M where\ndata Pair a = Pair a a")
(list
(hk-con-arity "Pair")
(hk-con-type "Pair")))
(list 2 "Pair"))
;; ── Non-data decls are ignored ──
(hk-test
"program with only fun-decl leaves registry unchanged for that name"
(do
(hk-load-source! "myFunctionNotACon x = x + 1")
(hk-is-con? "myFunctionNotACon"))
false)
;; ── Re-registering overwrites (last wins) ──
(hk-test
"re-registration overwrites the entry"
(do
(hk-load-source! "data Foo = Bar Int")
(hk-load-source! "data Foo = Bar Int Int")
(hk-con-arity "Bar"))
2)
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

85
lib/haskell/tests/seq.sx Normal file
View File

@@ -0,0 +1,85 @@
;; seq / deepseq tests. seq is strict in its first arg (forces to
;; WHNF) and returns the second arg unchanged. deepseq additionally
;; forces the first arg to normal form.
(define
hk-prog-val
(fn
(src name)
(hk-deep-force (get (hk-eval-program (hk-core src)) name))))
(define hk-as-list
(fn (xs)
(cond
((and (list? xs) (= (first xs) "[]")) (list))
((and (list? xs) (= (first xs) ":"))
(cons (nth xs 1) (hk-as-list (nth xs 2))))
(:else xs))))
(define
hk-eval-list
(fn (src) (hk-as-list (hk-eval-expr-source src))))
;; ── seq returns its second arg ──
(hk-test
"seq with primitive first arg"
(hk-eval-expr-source "seq 1 99")
99)
(hk-test
"seq forces first arg via let"
(hk-eval-expr-source "let x = 1 + 2 in seq x x")
3)
(hk-test
"seq second arg is whatever shape"
(hk-eval-expr-source "seq 0 \"hello\"")
"hello")
;; ── seq enables previously-lazy bottom to be forced ──
;; Without seq the let-binding `x = error …` is never forced;
;; with seq it must be forced because seq is strict in its first
;; argument. We don't run that error case here (it would terminate
;; the test), but we do verify the negative — that without seq,
;; the bottom bound is never demanded.
(hk-test
"lazy let — bottom never forced when unused"
(hk-eval-expr-source "let x = error \"never\" in 42")
42)
;; ── deepseq forces nested structure ──
(hk-test
"deepseq with finite list"
(hk-eval-expr-source "deepseq [1, 2, 3] 7")
7)
(hk-test
"deepseq with constructor value"
(hk-eval-expr-source "deepseq (Just 5) 11")
11)
(hk-test
"deepseq with tuple"
(hk-eval-expr-source "deepseq (1, 2) 13")
13)
;; ── seq + arithmetic ──
(hk-test
"seq used inside arithmetic doesn't poison the result"
(hk-eval-expr-source "(seq 1 5) + (seq 2 7)")
12)
;; ── seq in user code ──
(hk-test
"seq via fun-clause"
(hk-prog-val
"f x = seq x (x + 1)\nresult = f 10"
"result")
11)
(hk-test
"seq sequences list construction"
(hk-eval-list "[seq 1 10, seq 2 20]")
(list 10 20))
{:fails hk-test-fails :pass hk-test-pass :fail hk-test-fail}

View File

@@ -1,891 +0,0 @@
;; Smalltalk parser — produces an AST from the tokenizer's token stream.
;;
;; AST node shapes (dicts):
;; {:type "lit-int" :value N} integer
;; {:type "lit-float" :value F} float
;; {:type "lit-string" :value S} string
;; {:type "lit-char" :value C} character
;; {:type "lit-symbol" :value S} symbol literal (#foo)
;; {:type "lit-array" :elements (list ...)} literal array (#(1 2 #foo))
;; {:type "lit-byte-array" :elements (...)} byte array (#[1 2 3])
;; {:type "lit-nil" } / "lit-true" / "lit-false"
;; {:type "ident" :name "x"} variable reference
;; {:type "self"} / "super" / "thisContext" pseudo-variables
;; {:type "assign" :name "x" :expr E} x := E
;; {:type "return" :expr E} ^ E
;; {:type "send" :receiver R :selector S :args (list ...)}
;; {:type "cascade" :receiver R :messages (list {:selector :args} ...)}
;; {:type "block" :params (list "a") :temps (list "t") :body (list expr)}
;; {:type "seq" :exprs (list ...)} statement sequence
;; {:type "method" :selector S :params (list ...) :temps (list ...) :body (list ...) :pragmas (list ...)}
;;
;; A "chunk" / class-definition stream is parsed at a higher level (deferred).
;; ── Chunk-stream reader ────────────────────────────────────────────────
;; Pharo chunk format: chunks are separated by `!`. A doubled `!!` inside a
;; chunk represents a single literal `!`. Returns list of chunk strings with
;; surrounding whitespace trimmed.
(define
st-read-chunks
(fn
(src)
(let
((chunks (list))
(buf (list))
(pos 0)
(n (len src)))
(begin
(define
flush!
(fn
()
(let
((s (st-trim (join "" buf))))
(begin (append! chunks s) (set! buf (list))))))
(define
rc-loop
(fn
()
(when
(< pos n)
(let
((c (nth src pos)))
(cond
((= c "!")
(cond
((and (< (+ pos 1) n) (= (nth src (+ pos 1)) "!"))
(begin (append! buf "!") (set! pos (+ pos 2)) (rc-loop)))
(else
(begin (flush!) (set! pos (+ pos 1)) (rc-loop)))))
(else
(begin (append! buf c) (set! pos (+ pos 1)) (rc-loop))))))))
(rc-loop)
;; trailing text without a closing `!` — preserve as a chunk
(when (> (len buf) 0) (flush!))
chunks))))
(define
st-trim
(fn
(s)
(let
((n (len s)) (i 0) (j 0))
(begin
(set! j n)
(define
tl-loop
(fn
()
(when
(and (< i n) (st-trim-ws? (nth s i)))
(begin (set! i (+ i 1)) (tl-loop)))))
(tl-loop)
(define
tr-loop
(fn
()
(when
(and (> j i) (st-trim-ws? (nth s (- j 1))))
(begin (set! j (- j 1)) (tr-loop)))))
(tr-loop)
(slice s i j)))))
(define
st-trim-ws?
(fn (c) (or (= c " ") (= c "\t") (= c "\n") (= c "\r"))))
;; Parse a chunk stream. Walks chunks and applies the Pharo file-in
;; convention: a chunk that evaluates to "X methodsFor: 'cat'" or
;; "X class methodsFor: 'cat'" enters a methods batch — subsequent chunks
;; are method source until an empty chunk closes the batch.
;;
;; Returns list of entries:
;; {:kind "expr" :ast EXPR-AST}
;; {:kind "method" :class CLS :class-side? BOOL :category CAT :ast METHOD-AST}
;; {:kind "blank"} (empty chunks outside a methods batch)
;; {:kind "end-methods"} (empty chunk closing a methods batch)
(define
st-parse-chunks
(fn
(src)
(let
((chunks (st-read-chunks src))
(entries (list))
(mode "do-it")
(cls-name nil)
(class-side? false)
(category nil))
(begin
(for-each
(fn
(chunk)
(cond
((= chunk "")
(cond
((= mode "methods")
(begin
(append! entries {:kind "end-methods"})
(set! mode "do-it")
(set! cls-name nil)
(set! class-side? false)
(set! category nil)))
(else (append! entries {:kind "blank"}))))
((= mode "methods")
(append!
entries
{:kind "method"
:class cls-name
:class-side? class-side?
:category category
:ast (st-parse-method chunk)}))
(else
(let
((ast (st-parse-expr chunk)))
(begin
(append! entries {:kind "expr" :ast ast})
(let
((mf (st-detect-methods-for ast)))
(when
(not (= mf nil))
(begin
(set! mode "methods")
(set! cls-name (get mf :class))
(set! class-side? (get mf :class-side?))
(set! category (get mf :category))))))))))
chunks)
entries))))
;; Recognise `Foo methodsFor: 'cat'` (and related) as starting a methods batch.
;; Returns nil if the AST doesn't look like one of these forms.
(define
st-detect-methods-for
(fn
(ast)
(cond
((not (= (get ast :type) "send")) nil)
((not (st-is-methods-for-selector? (get ast :selector))) nil)
(else
(let
((recv (get ast :receiver)) (args (get ast :args)))
(let
((cat-arg (if (> (len args) 0) (nth args 0) nil)))
(let
((category
(cond
((= cat-arg nil) nil)
((= (get cat-arg :type) "lit-string") (get cat-arg :value))
((= (get cat-arg :type) "lit-symbol") (get cat-arg :value))
(else nil))))
(cond
((= (get recv :type) "ident")
{:class (get recv :name)
:class-side? false
:category category})
;; `Foo class methodsFor: 'cat'` — recv is a unary send `Foo class`
((and
(= (get recv :type) "send")
(= (get recv :selector) "class")
(= (get (get recv :receiver) :type) "ident"))
{:class (get (get recv :receiver) :name)
:class-side? true
:category category})
(else nil)))))))))
(define
st-is-methods-for-selector?
(fn
(sel)
(or
(= sel "methodsFor:")
(= sel "methodsFor:stamp:")
(= sel "category:"))))
(define st-tok-type (fn (t) (if (= t nil) "eof" (get t :type))))
(define st-tok-value (fn (t) (if (= t nil) nil (get t :value))))
;; Parse a *single* Smalltalk expression from source.
(define st-parse-expr (fn (src) (st-parse-with src "expr")))
;; Parse a sequence of statements separated by '.' Returns a {:type "seq"} node.
(define st-parse (fn (src) (st-parse-with src "seq")))
;; Parse a method body — `selector params | temps | body`.
;; Only the "method header + body" form (no chunk delimiters).
(define st-parse-method (fn (src) (st-parse-with src "method")))
(define
st-parse-with
(fn
(src mode)
(let
((tokens (st-tokenize src)) (idx 0) (tok-len 0))
(begin
(set! tok-len (len tokens))
(define peek-tok (fn () (nth tokens idx)))
(define
peek-tok-at
(fn (n) (if (< (+ idx n) tok-len) (nth tokens (+ idx n)) nil)))
(define advance-tok! (fn () (set! idx (+ idx 1))))
(define
at?
(fn
(type value)
(let
((t (peek-tok)))
(and
(= (st-tok-type t) type)
(or (= value nil) (= (st-tok-value t) value))))))
(define at-type? (fn (type) (= (st-tok-type (peek-tok)) type)))
(define
consume!
(fn
(type value)
(if
(at? type value)
(let ((t (peek-tok))) (begin (advance-tok!) t))
(error
(str
"st-parse: expected "
type
(if (= value nil) "" (str " '" value "'"))
" got "
(st-tok-type (peek-tok))
" '"
(st-tok-value (peek-tok))
"' at idx "
idx)))))
;; ── Primary: atoms, paren'd expr, blocks, literal arrays, byte arrays.
(define
parse-primary
(fn
()
(let
((t (peek-tok)))
(let
((ty (st-tok-type t)) (v (st-tok-value t)))
(cond
((= ty "number")
(begin
(advance-tok!)
(cond
((number? v) {:type (if (integer? v) "lit-int" "lit-float") :value v})
(else {:type "lit-int" :value v}))))
((= ty "string")
(begin (advance-tok!) {:type "lit-string" :value v}))
((= ty "char")
(begin (advance-tok!) {:type "lit-char" :value v}))
((= ty "symbol")
(begin (advance-tok!) {:type "lit-symbol" :value v}))
((= ty "array-open") (parse-literal-array))
((= ty "byte-array-open") (parse-byte-array))
((= ty "lparen")
(begin
(advance-tok!)
(let
((e (parse-expression)))
(begin (consume! "rparen" nil) e))))
((= ty "lbracket") (parse-block))
((= ty "ident")
(begin
(advance-tok!)
(cond
((= v "nil") {:type "lit-nil"})
((= v "true") {:type "lit-true"})
((= v "false") {:type "lit-false"})
((= v "self") {:type "self"})
((= v "super") {:type "super"})
((= v "thisContext") {:type "thisContext"})
(else {:type "ident" :name v}))))
((= ty "binary")
;; Negative numeric literal: '-' immediately before a number.
(cond
((and (= v "-") (= (st-tok-type (peek-tok-at 1)) "number"))
(let
((n (st-tok-value (peek-tok-at 1))))
(begin
(advance-tok!)
(advance-tok!)
(cond
((dict? n) {:type "lit-int" :value n})
((integer? n) {:type "lit-int" :value (- 0 n)})
(else {:type "lit-float" :value (- 0 n)})))))
(else
(error
(str "st-parse: unexpected binary '" v "' at idx " idx)))))
(else
(error
(str
"st-parse: unexpected "
ty
" '"
v
"' at idx "
idx))))))))
;; #(elem elem ...) — elements are atoms or nested parenthesised arrays.
(define
parse-literal-array
(fn
()
(let
((items (list)))
(begin
(consume! "array-open" nil)
(define
arr-loop
(fn
()
(cond
((at? "rparen" nil) (advance-tok!))
(else
(begin
(append! items (parse-array-element))
(arr-loop))))))
(arr-loop)
{:type "lit-array" :elements items}))))
;; #[1 2 3]
(define
parse-byte-array
(fn
()
(let
((items (list)))
(begin
(consume! "byte-array-open" nil)
(define
ba-loop
(fn
()
(cond
((at? "rbracket" nil) (advance-tok!))
(else
(let
((t (peek-tok)))
(cond
((= (st-tok-type t) "number")
(begin
(advance-tok!)
(append! items (st-tok-value t))
(ba-loop)))
(else
(error
(str
"st-parse: byte array expects number, got "
(st-tok-type t))))))))))
(ba-loop)
{:type "lit-byte-array" :elements items}))))
;; Inside a literal array: bare idents become symbols, nested (...) is a sub-array.
(define
parse-array-element
(fn
()
(let
((t (peek-tok)))
(let
((ty (st-tok-type t)) (v (st-tok-value t)))
(cond
((= ty "number") (begin (advance-tok!) {:type "lit-int" :value v}))
((= ty "string") (begin (advance-tok!) {:type "lit-string" :value v}))
((= ty "char") (begin (advance-tok!) {:type "lit-char" :value v}))
((= ty "symbol") (begin (advance-tok!) {:type "lit-symbol" :value v}))
((= ty "ident")
(begin
(advance-tok!)
(cond
((= v "nil") {:type "lit-nil"})
((= v "true") {:type "lit-true"})
((= v "false") {:type "lit-false"})
(else {:type "lit-symbol" :value v}))))
((= ty "keyword") (begin (advance-tok!) {:type "lit-symbol" :value v}))
((= ty "binary") (begin (advance-tok!) {:type "lit-symbol" :value v}))
((= ty "lparen")
(let ((items (list)))
(begin
(advance-tok!)
(define
sub-loop
(fn
()
(cond
((at? "rparen" nil) (advance-tok!))
(else
(begin (append! items (parse-array-element)) (sub-loop))))))
(sub-loop)
{:type "lit-array" :elements items})))
((= ty "array-open") (parse-literal-array))
((= ty "byte-array-open") (parse-byte-array))
(else
(error
(str "st-parse: bad literal-array element " ty " '" v "'"))))))))
;; [:a :b | | t1 t2 | body. body. ...]
(define
parse-block
(fn
()
(begin
(consume! "lbracket" nil)
(let
((params (list)) (temps (list)))
(begin
;; Block params
(define
p-loop
(fn
()
(when
(at? "colon" nil)
(begin
(advance-tok!)
(let
((t (consume! "ident" nil)))
(begin
(append! params (st-tok-value t))
(p-loop)))))))
(p-loop)
(when (> (len params) 0) (consume! "bar" nil))
;; Block temps: | t1 t2 |
(when
(and
(at? "bar" nil)
;; Not `|` followed immediately by binary content — the only
;; legitimate `|` inside a block here is the temp delimiter.
true)
(begin
(advance-tok!)
(define
t-loop
(fn
()
(when
(at? "ident" nil)
(let
((t (peek-tok)))
(begin
(advance-tok!)
(append! temps (st-tok-value t))
(t-loop))))))
(t-loop)
(consume! "bar" nil)))
;; Body: statements terminated by `.` or `]`
(let
((body (parse-statements "rbracket")))
(begin
(consume! "rbracket" nil)
{:type "block" :params params :temps temps :body body})))))))
;; Parse statements up to a closing token (rbracket or eof). Returns list.
(define
parse-statements
(fn
(terminator)
(let
((stmts (list)))
(begin
(define
s-loop
(fn
()
(cond
((at-type? terminator) nil)
((at-type? "eof") nil)
(else
(begin
(append! stmts (parse-statement))
;; consume optional period(s)
(define
dot-loop
(fn
()
(when
(at? "period" nil)
(begin (advance-tok!) (dot-loop)))))
(dot-loop)
(s-loop))))))
(s-loop)
stmts))))
;; Statement: ^expr | ident := expr | expr
(define
parse-statement
(fn
()
(cond
((at? "caret" nil)
(begin
(advance-tok!)
{:type "return" :expr (parse-expression)}))
((and (at-type? "ident") (= (st-tok-type (peek-tok-at 1)) "assign"))
(let
((name-tok (peek-tok)))
(begin
(advance-tok!)
(advance-tok!)
{:type "assign"
:name (st-tok-value name-tok)
:expr (parse-expression)})))
(else (parse-expression)))))
;; Top-level expression. Assignment (right-associative chain) sits at
;; the top; cascade is below.
(define
parse-expression
(fn
()
(cond
((and (at-type? "ident") (= (st-tok-type (peek-tok-at 1)) "assign"))
(let
((name-tok (peek-tok)))
(begin
(advance-tok!)
(advance-tok!)
{:type "assign"
:name (st-tok-value name-tok)
:expr (parse-expression)})))
(else (parse-cascade)))))
(define
parse-cascade
(fn
()
(let
((head (parse-keyword-message)))
(cond
((at? "semi" nil)
(let
((receiver (cascade-receiver head))
(first-msg (cascade-first-message head))
(msgs (list)))
(begin
(append! msgs first-msg)
(define
c-loop
(fn
()
(when
(at? "semi" nil)
(begin
(advance-tok!)
(append! msgs (parse-cascade-message))
(c-loop)))))
(c-loop)
{:type "cascade" :receiver receiver :messages msgs})))
(else head)))))
;; Extract the receiver from a head send so cascades share it.
(define
cascade-receiver
(fn
(head)
(cond
((= (get head :type) "send") (get head :receiver))
(else head))))
(define
cascade-first-message
(fn
(head)
(cond
((= (get head :type) "send")
{:selector (get head :selector) :args (get head :args)})
(else
;; Shouldn't happen — cascade requires at least one prior message.
(error "st-parse: cascade with no prior message")))))
;; Subsequent cascade message (after the `;`): unary | binary | keyword
(define
parse-cascade-message
(fn
()
(cond
((at-type? "ident")
(let ((t (peek-tok)))
(begin
(advance-tok!)
{:selector (st-tok-value t) :args (list)})))
((at-type? "binary")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(let
((arg (parse-unary-message)))
{:selector (st-tok-value t) :args (list arg)}))))
((at-type? "keyword")
(let
((sel-parts (list)) (args (list)))
(begin
(define
kw-loop
(fn
()
(when
(at-type? "keyword")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(append! sel-parts (st-tok-value t))
(append! args (parse-binary-message))
(kw-loop))))))
(kw-loop)
{:selector (join "" sel-parts) :args args})))
(else
(error
(str "st-parse: bad cascade message at idx " idx))))))
;; Keyword message: <binary> (kw <binary>)+
(define
parse-keyword-message
(fn
()
(let
((receiver (parse-binary-message)))
(cond
((at-type? "keyword")
(let
((sel-parts (list)) (args (list)))
(begin
(define
kw-loop
(fn
()
(when
(at-type? "keyword")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(append! sel-parts (st-tok-value t))
(append! args (parse-binary-message))
(kw-loop))))))
(kw-loop)
{:type "send"
:receiver receiver
:selector (join "" sel-parts)
:args args})))
(else receiver)))))
;; Binary message: <unary> (binop <unary>)*
(define
parse-binary-message
(fn
()
(let
((receiver (parse-unary-message)))
(begin
(define
b-loop
(fn
()
(when
(at-type? "binary")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(let
((arg (parse-unary-message)))
(set!
receiver
{:type "send"
:receiver receiver
:selector (st-tok-value t)
:args (list arg)}))
(b-loop))))))
(b-loop)
receiver))))
;; Unary message: <primary> ident* (ident NOT followed by ':')
(define
parse-unary-message
(fn
()
(let
((receiver (parse-primary)))
(begin
(define
u-loop
(fn
()
(when
(and
(at-type? "ident")
(let
((nxt (peek-tok-at 1)))
(not (= (st-tok-type nxt) "assign"))))
(let ((t (peek-tok)))
(begin
(advance-tok!)
(set!
receiver
{:type "send"
:receiver receiver
:selector (st-tok-value t)
:args (list)})
(u-loop))))))
(u-loop)
receiver))))
;; Parse a single pragma: `<keyword: literal (keyword: literal)* >`
;; Returns {:selector "primitive:" :args (list literal-asts)}.
(define
parse-pragma
(fn
()
(begin
(consume! "binary" "<")
(let
((sel-parts (list)) (args (list)))
(begin
(define
pr-loop
(fn
()
(when
(at-type? "keyword")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(append! sel-parts (st-tok-value t))
(append! args (parse-pragma-arg))
(pr-loop))))))
(pr-loop)
(consume! "binary" ">")
{:selector (join "" sel-parts) :args args})))))
;; Pragma arguments are literals only.
(define
parse-pragma-arg
(fn
()
(let
((t (peek-tok)))
(let
((ty (st-tok-type t)) (v (st-tok-value t)))
(cond
((= ty "number")
(begin
(advance-tok!)
{:type (if (integer? v) "lit-int" "lit-float") :value v}))
((= ty "string") (begin (advance-tok!) {:type "lit-string" :value v}))
((= ty "char") (begin (advance-tok!) {:type "lit-char" :value v}))
((= ty "symbol") (begin (advance-tok!) {:type "lit-symbol" :value v}))
((= ty "ident")
(begin
(advance-tok!)
(cond
((= v "nil") {:type "lit-nil"})
((= v "true") {:type "lit-true"})
((= v "false") {:type "lit-false"})
(else (error (str "st-parse: pragma arg must be literal, got ident " v))))))
((and (= ty "binary") (= v "-")
(= (st-tok-type (peek-tok-at 1)) "number"))
(let ((n (st-tok-value (peek-tok-at 1))))
(begin
(advance-tok!)
(advance-tok!)
{:type (if (integer? n) "lit-int" "lit-float")
:value (- 0 n)})))
(else
(error
(str "st-parse: pragma arg must be literal, got " ty))))))))
;; Method header: unary | binary arg | (kw arg)+
(define
parse-method
(fn
()
(let
((sel "")
(params (list))
(temps (list))
(pragmas (list))
(body (list)))
(begin
(cond
;; Unary header
((at-type? "ident")
(let ((t (peek-tok)))
(begin (advance-tok!) (set! sel (st-tok-value t)))))
;; Binary header: binop ident
((at-type? "binary")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(set! sel (st-tok-value t))
(let ((p (consume! "ident" nil)))
(append! params (st-tok-value p))))))
;; Keyword header: (kw ident)+
((at-type? "keyword")
(let ((sel-parts (list)))
(begin
(define
kh-loop
(fn
()
(when
(at-type? "keyword")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(append! sel-parts (st-tok-value t))
(let ((p (consume! "ident" nil)))
(append! params (st-tok-value p)))
(kh-loop))))))
(kh-loop)
(set! sel (join "" sel-parts)))))
(else
(error
(str
"st-parse-method: expected selector header, got "
(st-tok-type (peek-tok))))))
;; Pragmas and temps may appear in either order. Allow many
;; pragmas; one temps section.
(define
parse-temps!
(fn
()
(begin
(advance-tok!)
(define
th-loop
(fn
()
(when
(at-type? "ident")
(let ((t (peek-tok)))
(begin
(advance-tok!)
(append! temps (st-tok-value t))
(th-loop))))))
(th-loop)
(consume! "bar" nil))))
(define
pt-loop
(fn
()
(cond
((and
(at? "binary" "<")
(= (st-tok-type (peek-tok-at 1)) "keyword"))
(begin (append! pragmas (parse-pragma)) (pt-loop)))
((and (at? "bar" nil) (= (len temps) 0))
(begin (parse-temps!) (pt-loop)))
(else nil))))
(pt-loop)
;; Body statements
(set! body (parse-statements "eof"))
{:type "method"
:selector sel
:params params
:temps temps
:pragmas pragmas
:body body}))))
;; Top-level program: statements separated by '.'
(cond
((= mode "expr") (parse-expression))
((= mode "method") (parse-method))
(else
{:type "seq" :exprs (parse-statements "eof")}))))))

View File

@@ -1,291 +0,0 @@
;; Smalltalk runtime — class table, bootstrap hierarchy, type→class mapping,
;; instance construction. Method dispatch / eval-ast live in a later layer.
;;
;; Class record shape:
;; {:name "Foo"
;; :superclass "Object" ; or nil for Object itself
;; :ivars (list "x" "y") ; instance variable names declared on this class
;; :methods (dict selector→method-record)
;; :class-methods (dict selector→method-record)}
;;
;; A method record is the AST returned by st-parse-method, plus a :defining-class
;; field so super-sends can resolve from the right place. (Methods are registered
;; via runtime helpers that fill the field.)
;;
;; The class table is a single dict keyed by class name. Bootstrap installs the
;; canonical hierarchy. Test code resets it via (st-bootstrap-classes!).
(define st-class-table {})
(define st-class-table-clear! (fn () (set! st-class-table {})))
(define
st-class-define!
(fn
(name superclass ivars)
(begin
(set!
st-class-table
(assoc
st-class-table
name
{:name name
:superclass superclass
:ivars ivars
:methods {}
:class-methods {}}))
name)))
(define
st-class-get
(fn (name) (if (has-key? st-class-table name) (get st-class-table name) nil)))
(define
st-class-exists?
(fn (name) (has-key? st-class-table name)))
(define
st-class-superclass
(fn
(name)
(let
((c (st-class-get name)))
(cond ((= c nil) nil) (else (get c :superclass))))))
;; Walk class chain root-to-leaf? No, follow superclass chain leaf-to-root.
;; Returns list of class names starting at `name` and ending with the root.
(define
st-class-chain
(fn
(name)
(let ((acc (list)) (cur name))
(begin
(define
ch-loop
(fn
()
(when
(and (not (= cur nil)) (st-class-exists? cur))
(begin
(append! acc cur)
(set! cur (st-class-superclass cur))
(ch-loop)))))
(ch-loop)
acc))))
;; Inherited + own ivars in declaration order from root to leaf.
(define
st-class-all-ivars
(fn
(name)
(let ((chain (reverse (st-class-chain name))) (out (list)))
(begin
(for-each
(fn
(cn)
(let
((c (st-class-get cn)))
(when
(not (= c nil))
(for-each (fn (iv) (append! out iv)) (get c :ivars)))))
chain)
out))))
;; Method install. The defining-class field is stamped on the method record
;; so super-sends look up from the right point in the chain.
(define
st-class-add-method!
(fn
(cls-name selector method-ast)
(let
((cls (st-class-get cls-name)))
(cond
((= cls nil) (error (str "st-class-add-method!: unknown class " cls-name)))
(else
(let
((m (assoc method-ast :defining-class cls-name)))
(begin
(set!
st-class-table
(assoc
st-class-table
cls-name
(assoc
cls
:methods
(assoc (get cls :methods) selector m))))
selector)))))))
(define
st-class-add-class-method!
(fn
(cls-name selector method-ast)
(let
((cls (st-class-get cls-name)))
(cond
((= cls nil) (error (str "st-class-add-class-method!: unknown class " cls-name)))
(else
(let
((m (assoc method-ast :defining-class cls-name)))
(begin
(set!
st-class-table
(assoc
st-class-table
cls-name
(assoc
cls
:class-methods
(assoc (get cls :class-methods) selector m))))
selector)))))))
;; Method lookup: walk superclass chain starting at `cls-name`.
;; class-side? = true searches :class-methods, false searches :methods.
;; Returns the method record (with :defining-class) or nil.
(define
st-method-lookup
(fn
(cls-name selector class-side?)
(let
((found nil))
(begin
(define
ml-loop
(fn
(cur)
(when
(and (= found nil) (not (= cur nil)) (st-class-exists? cur))
(let
((c (st-class-get cur)))
(let
((dict (if class-side? (get c :class-methods) (get c :methods))))
(cond
((has-key? dict selector) (set! found (get dict selector)))
(else (ml-loop (get c :superclass)))))))))
(ml-loop cls-name)
found))))
;; SX value → Smalltalk class name. Native types are not boxed.
(define
st-class-of
(fn
(v)
(cond
((= v nil) "UndefinedObject")
((= v true) "True")
((= v false) "False")
((integer? v) "SmallInteger")
((number? v) "Float")
((string? v) "String")
((symbol? v) "Symbol")
((list? v) "Array")
((and (dict? v) (has-key? v :type) (= (get v :type) "st-instance"))
(get v :class))
((and (dict? v) (has-key? v :type) (= (get v :type) "block"))
"BlockClosure")
((and (dict? v) (has-key? v :st-block?) (get v :st-block?))
"BlockClosure")
((dict? v) "Dictionary")
((lambda? v) "BlockClosure")
(else "Object"))))
;; Construct a fresh instance of cls-name. Ivars (own + inherited) start as nil.
(define
st-make-instance
(fn
(cls-name)
(cond
((not (st-class-exists? cls-name))
(error (str "st-make-instance: unknown class " cls-name)))
(else
(let
((iv-names (st-class-all-ivars cls-name)) (ivars {}))
(begin
(for-each (fn (n) (set! ivars (assoc ivars n nil))) iv-names)
{:type "st-instance" :class cls-name :ivars ivars}))))))
(define
st-instance?
(fn
(v)
(and (dict? v) (has-key? v :type) (= (get v :type) "st-instance"))))
(define
st-iv-get
(fn
(inst name)
(let ((ivs (get inst :ivars)))
(if (has-key? ivs name) (get ivs name) nil))))
(define
st-iv-set!
(fn
(inst name value)
(let
((new-ivars (assoc (get inst :ivars) name value)))
(assoc inst :ivars new-ivars))))
;; Inherits-from check: is `descendant` either equal to `ancestor` or a subclass?
(define
st-class-inherits-from?
(fn
(descendant ancestor)
(let ((found false) (cur descendant))
(begin
(define
ih-loop
(fn
()
(when
(and (not found) (not (= cur nil)) (st-class-exists? cur))
(cond
((= cur ancestor) (set! found true))
(else
(begin
(set! cur (st-class-superclass cur))
(ih-loop)))))))
(ih-loop)
found))))
;; Bootstrap the canonical class hierarchy. Reset and rebuild.
(define
st-bootstrap-classes!
(fn
()
(begin
(st-class-table-clear!)
;; Root
(st-class-define! "Object" nil (list))
;; Class side machinery
(st-class-define! "Behavior" "Object" (list "superclass" "methodDict" "format"))
(st-class-define! "ClassDescription" "Behavior" (list "instanceVariables" "organization"))
(st-class-define! "Class" "ClassDescription" (list "name" "subclasses"))
(st-class-define! "Metaclass" "ClassDescription" (list "thisClass"))
;; Pseudo-variable types
(st-class-define! "UndefinedObject" "Object" (list))
(st-class-define! "Boolean" "Object" (list))
(st-class-define! "True" "Boolean" (list))
(st-class-define! "False" "Boolean" (list))
;; Magnitudes
(st-class-define! "Magnitude" "Object" (list))
(st-class-define! "Number" "Magnitude" (list))
(st-class-define! "Integer" "Number" (list))
(st-class-define! "SmallInteger" "Integer" (list))
(st-class-define! "LargePositiveInteger" "Integer" (list))
(st-class-define! "Float" "Number" (list))
(st-class-define! "Character" "Magnitude" (list "value"))
;; Collections
(st-class-define! "Collection" "Object" (list))
(st-class-define! "SequenceableCollection" "Collection" (list))
(st-class-define! "ArrayedCollection" "SequenceableCollection" (list))
(st-class-define! "Array" "ArrayedCollection" (list))
(st-class-define! "String" "ArrayedCollection" (list))
(st-class-define! "Symbol" "String" (list))
(st-class-define! "OrderedCollection" "SequenceableCollection" (list "array" "firstIndex" "lastIndex"))
(st-class-define! "Dictionary" "Collection" (list))
;; Blocks / contexts
(st-class-define! "BlockClosure" "Object" (list))
"ok")))
;; Initialise on load. Tests can re-bootstrap to reset state.
(st-bootstrap-classes!)

View File

@@ -1,137 +0,0 @@
#!/usr/bin/env bash
# Fast Smalltalk-on-SX test runner — pipes directly to sx_server.exe.
# Mirrors lib/haskell/test.sh.
#
# Usage:
# bash lib/smalltalk/test.sh # run all tests
# bash lib/smalltalk/test.sh -v # verbose
# bash lib/smalltalk/test.sh tests/tokenize.sx # run one file
set -uo pipefail
cd "$(git rev-parse --show-toplevel)"
SX_SERVER="hosts/ocaml/_build/default/bin/sx_server.exe"
if [ ! -x "$SX_SERVER" ]; then
MAIN_ROOT=$(git worktree list | head -1 | awk '{print $1}')
if [ -x "$MAIN_ROOT/$SX_SERVER" ]; then
SX_SERVER="$MAIN_ROOT/$SX_SERVER"
else
echo "ERROR: sx_server.exe not found. Run: cd hosts/ocaml && dune build"
exit 1
fi
fi
VERBOSE=""
FILES=()
for arg in "$@"; do
case "$arg" in
-v|--verbose) VERBOSE=1 ;;
*) FILES+=("$arg") ;;
esac
done
if [ ${#FILES[@]} -eq 0 ]; then
# tokenize.sx must load first — it defines the st-test helpers reused by
# subsequent test files. Sort enforces this lexicographically.
mapfile -t FILES < <(find lib/smalltalk/tests -maxdepth 2 -name '*.sx' | sort)
fi
TOTAL_PASS=0
TOTAL_FAIL=0
FAILED_FILES=()
for FILE in "${FILES[@]}"; do
[ -f "$FILE" ] || { echo "skip $FILE (not found)"; continue; }
TMPFILE=$(mktemp)
if [ "$(basename "$FILE")" = "tokenize.sx" ]; then
cat > "$TMPFILE" <<EPOCHS
(epoch 1)
(load "lib/smalltalk/tokenizer.sx")
(epoch 2)
(load "$FILE")
(epoch 3)
(eval "(list st-test-pass st-test-fail)")
EPOCHS
else
cat > "$TMPFILE" <<EPOCHS
(epoch 1)
(load "lib/smalltalk/tokenizer.sx")
(epoch 2)
(load "lib/smalltalk/parser.sx")
(epoch 3)
(load "lib/smalltalk/runtime.sx")
(epoch 4)
(load "lib/smalltalk/tests/tokenize.sx")
(epoch 5)
(load "$FILE")
(epoch 6)
(eval "(list st-test-pass st-test-fail)")
EPOCHS
fi
OUTPUT=$(timeout 60 "$SX_SERVER" < "$TMPFILE" 2>&1 || true)
rm -f "$TMPFILE"
# Final epoch's value: either (ok N (P F)) on one line or
# (ok-len N M)\n(P F) where the value is on the following line.
LINE=$(echo "$OUTPUT" | awk '/^\(ok-len [0-9]+ / {getline; print}' | tail -1)
if [ -z "$LINE" ]; then
LINE=$(echo "$OUTPUT" | grep -E '^\(ok [0-9]+ \([0-9]+ [0-9]+\)\)' | tail -1 \
| sed -E 's/^\(ok [0-9]+ //; s/\)$//')
fi
if [ -z "$LINE" ]; then
echo "X $FILE: could not extract summary"
echo "$OUTPUT" | tail -30
TOTAL_FAIL=$((TOTAL_FAIL + 1))
FAILED_FILES+=("$FILE")
continue
fi
P=$(echo "$LINE" | sed -E 's/^\(([0-9]+) ([0-9]+)\).*/\1/')
F=$(echo "$LINE" | sed -E 's/^\(([0-9]+) ([0-9]+)\).*/\2/')
TOTAL_PASS=$((TOTAL_PASS + P))
TOTAL_FAIL=$((TOTAL_FAIL + F))
if [ "$F" -gt 0 ]; then
FAILED_FILES+=("$FILE")
printf 'X %-40s %d/%d\n' "$FILE" "$P" "$((P+F))"
TMPFILE2=$(mktemp)
if [ "$(basename "$FILE")" = "tokenize.sx" ]; then
cat > "$TMPFILE2" <<EPOCHS
(epoch 1)
(load "lib/smalltalk/tokenizer.sx")
(epoch 2)
(load "$FILE")
(epoch 3)
(eval "(map (fn (f) (get f :name)) st-test-fails)")
EPOCHS
else
cat > "$TMPFILE2" <<EPOCHS
(epoch 1)
(load "lib/smalltalk/tokenizer.sx")
(epoch 2)
(load "lib/smalltalk/parser.sx")
(epoch 3)
(load "lib/smalltalk/runtime.sx")
(epoch 4)
(load "lib/smalltalk/tests/tokenize.sx")
(epoch 5)
(load "$FILE")
(epoch 6)
(eval "(map (fn (f) (get f :name)) st-test-fails)")
EPOCHS
fi
FAILS=$(timeout 60 "$SX_SERVER" < "$TMPFILE2" 2>&1 | grep -E '^\(ok [0-9]+ \(' | tail -1 || true)
rm -f "$TMPFILE2"
echo " $FAILS"
elif [ "$VERBOSE" = "1" ]; then
printf 'OK %-40s %d passed\n' "$FILE" "$P"
fi
done
TOTAL=$((TOTAL_PASS + TOTAL_FAIL))
if [ $TOTAL_FAIL -eq 0 ]; then
echo "OK $TOTAL_PASS/$TOTAL smalltalk-on-sx tests passed"
else
echo "FAIL $TOTAL_PASS/$TOTAL passed, $TOTAL_FAIL failed in: ${FAILED_FILES[*]}"
fi
[ $TOTAL_FAIL -eq 0 ]

View File

@@ -1,369 +0,0 @@
;; Smalltalk parser tests.
;;
;; Reuses helpers (st-test, st-deep=?) from tokenize.sx. Counters reset
;; here so this file's summary covers parse tests only.
(set! st-test-pass 0)
(set! st-test-fail 0)
(set! st-test-fails (list))
;; ── 1. Atoms ──
(st-test "int" (st-parse-expr "42") {:type "lit-int" :value 42})
(st-test "float" (st-parse-expr "3.14") {:type "lit-float" :value 3.14})
(st-test "string" (st-parse-expr "'hi'") {:type "lit-string" :value "hi"})
(st-test "char" (st-parse-expr "$x") {:type "lit-char" :value "x"})
(st-test "symbol" (st-parse-expr "#foo") {:type "lit-symbol" :value "foo"})
(st-test "binary symbol" (st-parse-expr "#+") {:type "lit-symbol" :value "+"})
(st-test "keyword symbol" (st-parse-expr "#at:put:") {:type "lit-symbol" :value "at:put:"})
(st-test "nil" (st-parse-expr "nil") {:type "lit-nil"})
(st-test "true" (st-parse-expr "true") {:type "lit-true"})
(st-test "false" (st-parse-expr "false") {:type "lit-false"})
(st-test "self" (st-parse-expr "self") {:type "self"})
(st-test "super" (st-parse-expr "super") {:type "super"})
(st-test "ident" (st-parse-expr "x") {:type "ident" :name "x"})
(st-test "negative int" (st-parse-expr "-3") {:type "lit-int" :value -3})
;; ── 2. Literal arrays ──
(st-test
"literal array of ints"
(st-parse-expr "#(1 2 3)")
{:type "lit-array"
:elements (list
{:type "lit-int" :value 1}
{:type "lit-int" :value 2}
{:type "lit-int" :value 3})})
(st-test
"literal array mixed"
(st-parse-expr "#(1 #foo 'x' true)")
{:type "lit-array"
:elements (list
{:type "lit-int" :value 1}
{:type "lit-symbol" :value "foo"}
{:type "lit-string" :value "x"}
{:type "lit-true"})})
(st-test
"literal array bare ident is symbol"
(st-parse-expr "#(foo bar)")
{:type "lit-array"
:elements (list
{:type "lit-symbol" :value "foo"}
{:type "lit-symbol" :value "bar"})})
(st-test
"nested literal array"
(st-parse-expr "#(1 (2 3) 4)")
{:type "lit-array"
:elements (list
{:type "lit-int" :value 1}
{:type "lit-array"
:elements (list
{:type "lit-int" :value 2}
{:type "lit-int" :value 3})}
{:type "lit-int" :value 4})})
(st-test
"byte array"
(st-parse-expr "#[1 2 3]")
{:type "lit-byte-array" :elements (list 1 2 3)})
;; ── 3. Unary messages ──
(st-test
"unary single"
(st-parse-expr "x foo")
{:type "send"
:receiver {:type "ident" :name "x"}
:selector "foo"
:args (list)})
(st-test
"unary chain"
(st-parse-expr "x foo bar baz")
{:type "send"
:receiver {:type "send"
:receiver {:type "send"
:receiver {:type "ident" :name "x"}
:selector "foo"
:args (list)}
:selector "bar"
:args (list)}
:selector "baz"
:args (list)})
(st-test
"unary on literal"
(st-parse-expr "42 printNl")
{:type "send"
:receiver {:type "lit-int" :value 42}
:selector "printNl"
:args (list)})
;; ── 4. Binary messages ──
(st-test
"binary single"
(st-parse-expr "1 + 2")
{:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})})
(st-test
"binary left-assoc"
(st-parse-expr "1 + 2 + 3")
{:type "send"
:receiver {:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})}
:selector "+"
:args (list {:type "lit-int" :value 3})})
(st-test
"binary same precedence l-to-r"
(st-parse-expr "1 + 2 * 3")
{:type "send"
:receiver {:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})}
:selector "*"
:args (list {:type "lit-int" :value 3})})
;; ── 5. Precedence: unary binds tighter than binary ──
(st-test
"unary tighter than binary"
(st-parse-expr "3 + 4 factorial")
{:type "send"
:receiver {:type "lit-int" :value 3}
:selector "+"
:args (list
{:type "send"
:receiver {:type "lit-int" :value 4}
:selector "factorial"
:args (list)})})
;; ── 6. Keyword messages ──
(st-test
"keyword single"
(st-parse-expr "x at: 1")
{:type "send"
:receiver {:type "ident" :name "x"}
:selector "at:"
:args (list {:type "lit-int" :value 1})})
(st-test
"keyword chain"
(st-parse-expr "x at: 1 put: 'a'")
{:type "send"
:receiver {:type "ident" :name "x"}
:selector "at:put:"
:args (list {:type "lit-int" :value 1} {:type "lit-string" :value "a"})})
;; ── 7. Precedence: binary tighter than keyword ──
(st-test
"binary tighter than keyword"
(st-parse-expr "x at: 1 + 2")
{:type "send"
:receiver {:type "ident" :name "x"}
:selector "at:"
:args (list
{:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})})})
(st-test
"keyword absorbs trailing unary"
(st-parse-expr "a foo: b bar")
{:type "send"
:receiver {:type "ident" :name "a"}
:selector "foo:"
:args (list
{:type "send"
:receiver {:type "ident" :name "b"}
:selector "bar"
:args (list)})})
;; ── 8. Parens override precedence ──
(st-test
"paren forces grouping"
(st-parse-expr "(1 + 2) * 3")
{:type "send"
:receiver {:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})}
:selector "*"
:args (list {:type "lit-int" :value 3})})
;; ── 9. Cascade ──
(st-test
"simple cascade"
(st-parse-expr "x m1; m2")
{:type "cascade"
:receiver {:type "ident" :name "x"}
:messages (list
{:selector "m1" :args (list)}
{:selector "m2" :args (list)})})
(st-test
"cascade with binary and keyword"
(st-parse-expr "Stream new nl; tab; print: 1")
{:type "cascade"
:receiver {:type "send"
:receiver {:type "ident" :name "Stream"}
:selector "new"
:args (list)}
:messages (list
{:selector "nl" :args (list)}
{:selector "tab" :args (list)}
{:selector "print:" :args (list {:type "lit-int" :value 1})})})
;; ── 10. Blocks ──
(st-test
"empty block"
(st-parse-expr "[]")
{:type "block" :params (list) :temps (list) :body (list)})
(st-test
"block one expr"
(st-parse-expr "[1 + 2]")
{:type "block"
:params (list)
:temps (list)
:body (list
{:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})})})
(st-test
"block with params"
(st-parse-expr "[:a :b | a + b]")
{:type "block"
:params (list "a" "b")
:temps (list)
:body (list
{:type "send"
:receiver {:type "ident" :name "a"}
:selector "+"
:args (list {:type "ident" :name "b"})})})
(st-test
"block with temps"
(st-parse-expr "[| t | t := 1. t]")
{:type "block"
:params (list)
:temps (list "t")
:body (list
{:type "assign" :name "t" :expr {:type "lit-int" :value 1}}
{:type "ident" :name "t"})})
(st-test
"block with params and temps"
(st-parse-expr "[:x | | t | t := x + 1. t]")
{:type "block"
:params (list "x")
:temps (list "t")
:body (list
{:type "assign"
:name "t"
:expr {:type "send"
:receiver {:type "ident" :name "x"}
:selector "+"
:args (list {:type "lit-int" :value 1})}}
{:type "ident" :name "t"})})
;; ── 11. Assignment / return / statements ──
(st-test
"assignment"
(st-parse-expr "x := 1")
{:type "assign" :name "x" :expr {:type "lit-int" :value 1}})
(st-test
"return"
(st-parse-expr "1")
{:type "lit-int" :value 1})
(st-test
"return statement at top level"
(st-parse "^ 1")
{:type "seq"
:exprs (list {:type "return" :expr {:type "lit-int" :value 1}})})
(st-test
"two statements"
(st-parse "x := 1. y := 2")
{:type "seq"
:exprs (list
{:type "assign" :name "x" :expr {:type "lit-int" :value 1}}
{:type "assign" :name "y" :expr {:type "lit-int" :value 2}})})
(st-test
"trailing dot allowed"
(st-parse "1. 2.")
{:type "seq"
:exprs (list {:type "lit-int" :value 1} {:type "lit-int" :value 2})})
;; ── 12. Method headers ──
(st-test
"unary method"
(st-parse-method "factorial ^ self * (self - 1) factorial")
{:type "method"
:selector "factorial"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return"
:expr {:type "send"
:receiver {:type "self"}
:selector "*"
:args (list
{:type "send"
:receiver {:type "send"
:receiver {:type "self"}
:selector "-"
:args (list {:type "lit-int" :value 1})}
:selector "factorial"
:args (list)})}})})
(st-test
"binary method"
(st-parse-method "+ other ^ 'plus'")
{:type "method"
:selector "+"
:params (list "other")
:temps (list)
:pragmas (list)
:body (list {:type "return" :expr {:type "lit-string" :value "plus"}})})
(st-test
"keyword method"
(st-parse-method "at: i put: v ^ v")
{:type "method"
:selector "at:put:"
:params (list "i" "v")
:temps (list)
:pragmas (list)
:body (list {:type "return" :expr {:type "ident" :name "v"}})})
(st-test
"method with temps"
(st-parse-method "twice: x | t | t := x + x. ^ t")
{:type "method"
:selector "twice:"
:params (list "x")
:temps (list "t")
:pragmas (list)
:body (list
{:type "assign"
:name "t"
:expr {:type "send"
:receiver {:type "ident" :name "x"}
:selector "+"
:args (list {:type "ident" :name "x"})}}
{:type "return" :expr {:type "ident" :name "t"}})})
(list st-test-pass st-test-fail)

View File

@@ -1,294 +0,0 @@
;; Smalltalk chunk-stream parser + pragma tests.
;;
;; Reuses helpers (st-test, st-deep=?) from tokenize.sx. Counters reset
;; here so this file's summary covers chunk + pragma tests only.
(set! st-test-pass 0)
(set! st-test-fail 0)
(set! st-test-fails (list))
;; ── 1. Raw chunk reader ──
(st-test "empty source" (st-read-chunks "") (list))
(st-test "single chunk" (st-read-chunks "foo!") (list "foo"))
(st-test "two chunks" (st-read-chunks "a! b!") (list "a" "b"))
(st-test "trailing no bang" (st-read-chunks "a! b") (list "a" "b"))
(st-test "empty chunk" (st-read-chunks "a! ! b!") (list "a" "" "b"))
(st-test
"doubled bang escapes"
(st-read-chunks "yes!! no!yes!")
(list "yes! no" "yes"))
(st-test
"whitespace trimmed"
(st-read-chunks " \n hello \n !")
(list "hello"))
;; ── 2. Chunk parser — do-it mode ──
(st-test
"single do-it chunk"
(st-parse-chunks "1 + 2!")
(list
{:kind "expr"
:ast {:type "send"
:receiver {:type "lit-int" :value 1}
:selector "+"
:args (list {:type "lit-int" :value 2})}}))
(st-test
"two do-it chunks"
(st-parse-chunks "x := 1! y := 2!")
(list
{:kind "expr"
:ast {:type "assign" :name "x" :expr {:type "lit-int" :value 1}}}
{:kind "expr"
:ast {:type "assign" :name "y" :expr {:type "lit-int" :value 2}}}))
(st-test
"blank chunk outside methods"
(st-parse-chunks "1! ! 2!")
(list
{:kind "expr" :ast {:type "lit-int" :value 1}}
{:kind "blank"}
{:kind "expr" :ast {:type "lit-int" :value 2}}))
;; ── 3. Methods batch ──
(st-test
"methodsFor opens method batch"
(st-parse-chunks
"Foo methodsFor: 'access'! foo ^ 1! bar ^ 2! !")
(list
{:kind "expr"
:ast {:type "send"
:receiver {:type "ident" :name "Foo"}
:selector "methodsFor:"
:args (list {:type "lit-string" :value "access"})}}
{:kind "method"
:class "Foo"
:class-side? false
:category "access"
:ast {:type "method"
:selector "foo"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return" :expr {:type "lit-int" :value 1}})}}
{:kind "method"
:class "Foo"
:class-side? false
:category "access"
:ast {:type "method"
:selector "bar"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return" :expr {:type "lit-int" :value 2}})}}
{:kind "end-methods"}))
(st-test
"class-side methodsFor"
(st-parse-chunks
"Foo class methodsFor: 'creation'! make ^ self new! !")
(list
{:kind "expr"
:ast {:type "send"
:receiver {:type "send"
:receiver {:type "ident" :name "Foo"}
:selector "class"
:args (list)}
:selector "methodsFor:"
:args (list {:type "lit-string" :value "creation"})}}
{:kind "method"
:class "Foo"
:class-side? true
:category "creation"
:ast {:type "method"
:selector "make"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return"
:expr {:type "send"
:receiver {:type "self"}
:selector "new"
:args (list)}})}}
{:kind "end-methods"}))
(st-test
"method batch returns to do-it after empty chunk"
(st-parse-chunks
"Foo methodsFor: 'a'! m1 ^ 1! ! 99!")
(list
{:kind "expr"
:ast {:type "send"
:receiver {:type "ident" :name "Foo"}
:selector "methodsFor:"
:args (list {:type "lit-string" :value "a"})}}
{:kind "method"
:class "Foo"
:class-side? false
:category "a"
:ast {:type "method"
:selector "m1"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return" :expr {:type "lit-int" :value 1}})}}
{:kind "end-methods"}
{:kind "expr" :ast {:type "lit-int" :value 99}}))
;; ── 4. Pragmas in method bodies ──
(st-test
"single pragma"
(st-parse-method "primAt: i <primitive: 60> ^ self")
{:type "method"
:selector "primAt:"
:params (list "i")
:temps (list)
:pragmas (list
{:selector "primitive:"
:args (list {:type "lit-int" :value 60})})
:body (list {:type "return" :expr {:type "self"}})})
(st-test
"pragma with two keyword pairs"
(st-parse-method "fft <primitive: 1 module: 'fft'> ^ nil")
{:type "method"
:selector "fft"
:params (list)
:temps (list)
:pragmas (list
{:selector "primitive:module:"
:args (list
{:type "lit-int" :value 1}
{:type "lit-string" :value "fft"})})
:body (list {:type "return" :expr {:type "lit-nil"}})})
(st-test
"pragma with negative number"
(st-parse-method "neg <primitive: -1> ^ nil")
{:type "method"
:selector "neg"
:params (list)
:temps (list)
:pragmas (list
{:selector "primitive:"
:args (list {:type "lit-int" :value -1})})
:body (list {:type "return" :expr {:type "lit-nil"}})})
(st-test
"pragma with symbol arg"
(st-parse-method "tagged <category: #algebra> ^ nil")
{:type "method"
:selector "tagged"
:params (list)
:temps (list)
:pragmas (list
{:selector "category:"
:args (list {:type "lit-symbol" :value "algebra"})})
:body (list {:type "return" :expr {:type "lit-nil"}})})
(st-test
"pragma then temps"
(st-parse-method "calc <primitive: 1> | t | t := 5. ^ t")
{:type "method"
:selector "calc"
:params (list)
:temps (list "t")
:pragmas (list
{:selector "primitive:"
:args (list {:type "lit-int" :value 1})})
:body (list
{:type "assign" :name "t" :expr {:type "lit-int" :value 5}}
{:type "return" :expr {:type "ident" :name "t"}})})
(st-test
"temps then pragma"
(st-parse-method "calc | t | <primitive: 1> t := 5. ^ t")
{:type "method"
:selector "calc"
:params (list)
:temps (list "t")
:pragmas (list
{:selector "primitive:"
:args (list {:type "lit-int" :value 1})})
:body (list
{:type "assign" :name "t" :expr {:type "lit-int" :value 5}}
{:type "return" :expr {:type "ident" :name "t"}})})
(st-test
"two pragmas"
(st-parse-method "m <primitive: 1> <category: 'a'> ^ self")
{:type "method"
:selector "m"
:params (list)
:temps (list)
:pragmas (list
{:selector "primitive:"
:args (list {:type "lit-int" :value 1})}
{:selector "category:"
:args (list {:type "lit-string" :value "a"})})
:body (list {:type "return" :expr {:type "self"}})})
;; ── 5. End-to-end: a small "filed-in" snippet ──
(st-test
"small filed-in class snippet"
(st-parse-chunks
"Object subclass: #Account
instanceVariableNames: 'balance'!
!Account methodsFor: 'access'!
balance
^ balance!
deposit: amount
balance := balance + amount.
^ self! !")
(list
{:kind "expr"
:ast {:type "send"
:receiver {:type "ident" :name "Object"}
:selector "subclass:instanceVariableNames:"
:args (list
{:type "lit-symbol" :value "Account"}
{:type "lit-string" :value "balance"})}}
{:kind "blank"}
{:kind "expr"
:ast {:type "send"
:receiver {:type "ident" :name "Account"}
:selector "methodsFor:"
:args (list {:type "lit-string" :value "access"})}}
{:kind "method"
:class "Account"
:class-side? false
:category "access"
:ast {:type "method"
:selector "balance"
:params (list)
:temps (list)
:pragmas (list)
:body (list
{:type "return"
:expr {:type "ident" :name "balance"}})}}
{:kind "method"
:class "Account"
:class-side? false
:category "access"
:ast {:type "method"
:selector "deposit:"
:params (list "amount")
:temps (list)
:pragmas (list)
:body (list
{:type "assign"
:name "balance"
:expr {:type "send"
:receiver {:type "ident" :name "balance"}
:selector "+"
:args (list {:type "ident" :name "amount"})}}
{:type "return" :expr {:type "self"}})}}
{:kind "end-methods"}))
(list st-test-pass st-test-fail)

View File

@@ -1,182 +0,0 @@
;; Smalltalk runtime tests — class table, type→class mapping, instances.
;;
;; Reuses helpers (st-test, st-deep=?) from tokenize.sx. Counters reset
;; here so this file's summary covers runtime tests only.
(set! st-test-pass 0)
(set! st-test-fail 0)
(set! st-test-fails (list))
;; Fresh hierarchy for every test file.
(st-bootstrap-classes!)
;; ── 1. Bootstrap installed expected classes ──
(st-test "Object exists" (st-class-exists? "Object") true)
(st-test "Behavior exists" (st-class-exists? "Behavior") true)
(st-test "Metaclass exists" (st-class-exists? "Metaclass") true)
(st-test "True/False/UndefinedObject"
(and
(st-class-exists? "True")
(st-class-exists? "False")
(st-class-exists? "UndefinedObject"))
true)
(st-test "SmallInteger / Float / Symbol exist"
(and
(st-class-exists? "SmallInteger")
(st-class-exists? "Float")
(st-class-exists? "Symbol"))
true)
(st-test "BlockClosure exists" (st-class-exists? "BlockClosure") true)
;; ── 2. Superclass chain ──
(st-test "Object has no superclass" (st-class-superclass "Object") nil)
(st-test "Behavior super = Object" (st-class-superclass "Behavior") "Object")
(st-test "True super = Boolean" (st-class-superclass "True") "Boolean")
(st-test "Symbol super = String" (st-class-superclass "Symbol") "String")
(st-test
"String chain"
(st-class-chain "String")
(list "String" "ArrayedCollection" "SequenceableCollection" "Collection" "Object"))
(st-test
"SmallInteger chain"
(st-class-chain "SmallInteger")
(list "SmallInteger" "Integer" "Number" "Magnitude" "Object"))
;; ── 3. inherits-from? ──
(st-test "True inherits from Boolean" (st-class-inherits-from? "True" "Boolean") true)
(st-test "True inherits from Object" (st-class-inherits-from? "True" "Object") true)
(st-test "True inherits from True" (st-class-inherits-from? "True" "True") true)
(st-test
"True does not inherit from Number"
(st-class-inherits-from? "True" "Number")
false)
(st-test
"Object does not inherit from Number"
(st-class-inherits-from? "Object" "Number")
false)
;; ── 4. type→class mapping ──
(st-test "class-of nil" (st-class-of nil) "UndefinedObject")
(st-test "class-of true" (st-class-of true) "True")
(st-test "class-of false" (st-class-of false) "False")
(st-test "class-of int" (st-class-of 42) "SmallInteger")
(st-test "class-of zero" (st-class-of 0) "SmallInteger")
(st-test "class-of negative int" (st-class-of -3) "SmallInteger")
(st-test "class-of float" (st-class-of 3.14) "Float")
(st-test "class-of string" (st-class-of "hi") "String")
(st-test "class-of symbol" (st-class-of (quote foo)) "Symbol")
(st-test "class-of list" (st-class-of (list 1 2)) "Array")
(st-test "class-of empty list" (st-class-of (list)) "Array")
(st-test "class-of lambda" (st-class-of (fn (x) x)) "BlockClosure")
(st-test "class-of dict" (st-class-of {:a 1}) "Dictionary")
;; ── 5. User class definition ──
(st-class-define! "Account" "Object" (list "balance" "owner"))
(st-class-define! "SavingsAccount" "Account" (list "rate"))
(st-test "Account exists" (st-class-exists? "Account") true)
(st-test "Account super = Object" (st-class-superclass "Account") "Object")
(st-test
"SavingsAccount chain"
(st-class-chain "SavingsAccount")
(list "SavingsAccount" "Account" "Object"))
(st-test
"SavingsAccount own ivars"
(get (st-class-get "SavingsAccount") :ivars)
(list "rate"))
(st-test
"SavingsAccount inherited+own ivars"
(st-class-all-ivars "SavingsAccount")
(list "balance" "owner" "rate"))
;; ── 6. Instance construction ──
(define a1 (st-make-instance "Account"))
(st-test "instance is st-instance" (st-instance? a1) true)
(st-test "instance class" (get a1 :class) "Account")
(st-test "instance ivars start nil" (st-iv-get a1 "balance") nil)
(st-test
"instance has all expected ivars"
(sort (keys (get a1 :ivars)))
(sort (list "balance" "owner")))
(define a2 (st-iv-set! a1 "balance" 100))
(st-test "iv-set! returns updated copy" (st-iv-get a2 "balance") 100)
(st-test "iv-set! does not mutate original" (st-iv-get a1 "balance") nil)
(st-test "class-of instance" (st-class-of a1) "Account")
(define s1 (st-make-instance "SavingsAccount"))
(st-test
"subclass instance has all inherited ivars"
(sort (keys (get s1 :ivars)))
(sort (list "balance" "owner" "rate")))
;; ── 7. Method install + lookup ──
(st-class-add-method!
"Account"
"balance"
(st-parse-method "balance ^ balance"))
(st-class-add-method!
"Account"
"deposit:"
(st-parse-method "deposit: amount balance := balance + amount. ^ self"))
(st-test
"method registered"
(has-key? (get (st-class-get "Account") :methods) "balance")
true)
(st-test
"method lookup direct"
(= (st-method-lookup "Account" "balance" false) nil)
false)
(st-test
"method lookup walks superclass"
(= (st-method-lookup "SavingsAccount" "deposit:" false) nil)
false)
(st-test
"method lookup unknown selector"
(st-method-lookup "Account" "frobnicate" false)
nil)
(st-test
"method lookup records defining class"
(get (st-method-lookup "SavingsAccount" "balance" false) :defining-class)
"Account")
;; SavingsAccount overrides deposit:
(st-class-add-method!
"SavingsAccount"
"deposit:"
(st-parse-method "deposit: amount ^ super deposit: amount + 1"))
(st-test
"subclass override picked first"
(get (st-method-lookup "SavingsAccount" "deposit:" false) :defining-class)
"SavingsAccount")
(st-test
"Account still finds its own deposit:"
(get (st-method-lookup "Account" "deposit:" false) :defining-class)
"Account")
;; ── 8. Class-side methods ──
(st-class-add-class-method!
"Account"
"new"
(st-parse-method "new ^ super new"))
(st-test
"class-side lookup"
(= (st-method-lookup "Account" "new" true) nil)
false)
(st-test
"instance-side does not find class method"
(st-method-lookup "Account" "new" false)
nil)
;; ── 9. Re-bootstrap resets table ──
(st-bootstrap-classes!)
(st-test "after re-bootstrap Account gone" (st-class-exists? "Account") false)
(st-test "after re-bootstrap Object stays" (st-class-exists? "Object") true)
(list st-test-pass st-test-fail)

View File

@@ -1,362 +0,0 @@
;; Smalltalk tokenizer tests.
;;
;; Lightweight runner: each test checks actual vs expected with structural
;; equality and accumulates pass/fail counters. Final summary read by
;; lib/smalltalk/test.sh.
(define
st-deep=?
(fn
(a b)
(cond
((= a b) true)
((and (dict? a) (dict? b))
(let
((ak (keys a)) (bk (keys b)))
(if
(not (= (len ak) (len bk)))
false
(every?
(fn
(k)
(and (has-key? b k) (st-deep=? (get a k) (get b k))))
ak))))
((and (list? a) (list? b))
(if
(not (= (len a) (len b)))
false
(let
((i 0) (ok true))
(begin
(define
de-loop
(fn
()
(when
(and ok (< i (len a)))
(begin
(when
(not (st-deep=? (nth a i) (nth b i)))
(set! ok false))
(set! i (+ i 1))
(de-loop)))))
(de-loop)
ok))))
(:else false))))
(define st-test-pass 0)
(define st-test-fail 0)
(define st-test-fails (list))
(define
st-test
(fn
(name actual expected)
(if
(st-deep=? actual expected)
(set! st-test-pass (+ st-test-pass 1))
(begin
(set! st-test-fail (+ st-test-fail 1))
(append! st-test-fails {:actual actual :expected expected :name name})))))
;; Strip eof and project to just :type/:value.
(define
st-toks
(fn
(src)
(map
(fn (tok) {:type (get tok :type) :value (get tok :value)})
(filter
(fn (tok) (not (= (get tok :type) "eof")))
(st-tokenize src)))))
;; ── 1. Whitespace / empty ──
(st-test "empty input" (st-toks "") (list))
(st-test "all whitespace" (st-toks " \t\n ") (list))
;; ── 2. Identifiers ──
(st-test
"lowercase ident"
(st-toks "foo")
(list {:type "ident" :value "foo"}))
(st-test
"capitalised ident"
(st-toks "Foo")
(list {:type "ident" :value "Foo"}))
(st-test
"underscore ident"
(st-toks "_x")
(list {:type "ident" :value "_x"}))
(st-test
"digits in ident"
(st-toks "foo123")
(list {:type "ident" :value "foo123"}))
(st-test
"two idents separated"
(st-toks "foo bar")
(list {:type "ident" :value "foo"} {:type "ident" :value "bar"}))
;; ── 3. Keyword selectors ──
(st-test
"keyword selector"
(st-toks "foo:")
(list {:type "keyword" :value "foo:"}))
(st-test
"keyword call"
(st-toks "x at: 1")
(list
{:type "ident" :value "x"}
{:type "keyword" :value "at:"}
{:type "number" :value 1}))
(st-test
"two-keyword chain stays separate"
(st-toks "at: 1 put: 2")
(list
{:type "keyword" :value "at:"}
{:type "number" :value 1}
{:type "keyword" :value "put:"}
{:type "number" :value 2}))
(st-test
"ident then assign — not a keyword"
(st-toks "x := 1")
(list
{:type "ident" :value "x"}
{:type "assign" :value ":="}
{:type "number" :value 1}))
;; ── 4. Numbers ──
(st-test
"integer"
(st-toks "42")
(list {:type "number" :value 42}))
(st-test
"float"
(st-toks "3.14")
(list {:type "number" :value 3.14}))
(st-test
"hex radix"
(st-toks "16rFF")
(list
{:type "number"
:value
{:radix 16 :digits "FF" :value 255 :kind "radix"}}))
(st-test
"binary radix"
(st-toks "2r1011")
(list
{:type "number"
:value
{:radix 2 :digits "1011" :value 11 :kind "radix"}}))
(st-test
"exponent"
(st-toks "1e3")
(list {:type "number" :value 1000}))
(st-test
"negative exponent (parser handles minus)"
(st-toks "1.5e-2")
(list {:type "number" :value 0.015}))
;; ── 5. Strings ──
(st-test
"simple string"
(st-toks "'hi'")
(list {:type "string" :value "hi"}))
(st-test
"empty string"
(st-toks "''")
(list {:type "string" :value ""}))
(st-test
"doubled-quote escape"
(st-toks "'a''b'")
(list {:type "string" :value "a'b"}))
;; ── 6. Characters ──
(st-test
"char literal letter"
(st-toks "$a")
(list {:type "char" :value "a"}))
(st-test
"char literal punct"
(st-toks "$$")
(list {:type "char" :value "$"}))
(st-test
"char literal space"
(st-toks "$ ")
(list {:type "char" :value " "}))
;; ── 7. Symbols ──
(st-test
"symbol ident"
(st-toks "#foo")
(list {:type "symbol" :value "foo"}))
(st-test
"symbol binary"
(st-toks "#+")
(list {:type "symbol" :value "+"}))
(st-test
"symbol arrow"
(st-toks "#->")
(list {:type "symbol" :value "->"}))
(st-test
"symbol keyword chain"
(st-toks "#at:put:")
(list {:type "symbol" :value "at:put:"}))
(st-test
"quoted symbol with spaces"
(st-toks "#'foo bar'")
(list {:type "symbol" :value "foo bar"}))
;; ── 8. Literal arrays / byte arrays ──
(st-test
"literal array open"
(st-toks "#(1 2)")
(list
{:type "array-open" :value "#("}
{:type "number" :value 1}
{:type "number" :value 2}
{:type "rparen" :value ")"}))
(st-test
"byte array open"
(st-toks "#[1 2 3]")
(list
{:type "byte-array-open" :value "#["}
{:type "number" :value 1}
{:type "number" :value 2}
{:type "number" :value 3}
{:type "rbracket" :value "]"}))
;; ── 9. Binary selectors ──
(st-test "plus" (st-toks "+") (list {:type "binary" :value "+"}))
(st-test "minus" (st-toks "-") (list {:type "binary" :value "-"}))
(st-test "star" (st-toks "*") (list {:type "binary" :value "*"}))
(st-test "double-equal" (st-toks "==") (list {:type "binary" :value "=="}))
(st-test "leq" (st-toks "<=") (list {:type "binary" :value "<="}))
(st-test "geq" (st-toks ">=") (list {:type "binary" :value ">="}))
(st-test "neq" (st-toks "~=") (list {:type "binary" :value "~="}))
(st-test "arrow" (st-toks "->") (list {:type "binary" :value "->"}))
(st-test "comma" (st-toks ",") (list {:type "binary" :value ","}))
(st-test
"binary in expression"
(st-toks "a + b")
(list
{:type "ident" :value "a"}
{:type "binary" :value "+"}
{:type "ident" :value "b"}))
;; ── 10. Punctuation ──
(st-test "lparen" (st-toks "(") (list {:type "lparen" :value "("}))
(st-test "rparen" (st-toks ")") (list {:type "rparen" :value ")"}))
(st-test "lbracket" (st-toks "[") (list {:type "lbracket" :value "["}))
(st-test "rbracket" (st-toks "]") (list {:type "rbracket" :value "]"}))
(st-test "lbrace" (st-toks "{") (list {:type "lbrace" :value "{"}))
(st-test "rbrace" (st-toks "}") (list {:type "rbrace" :value "}"}))
(st-test "period" (st-toks ".") (list {:type "period" :value "."}))
(st-test "semi" (st-toks ";") (list {:type "semi" :value ";"}))
(st-test "bar" (st-toks "|") (list {:type "bar" :value "|"}))
(st-test "caret" (st-toks "^") (list {:type "caret" :value "^"}))
(st-test "bang" (st-toks "!") (list {:type "bang" :value "!"}))
(st-test "colon" (st-toks ":") (list {:type "colon" :value ":"}))
(st-test "assign" (st-toks ":=") (list {:type "assign" :value ":="}))
;; ── 11. Comments ──
(st-test "comment skipped" (st-toks "\"hello\"") (list))
(st-test
"comment between tokens"
(st-toks "a \"comment\" b")
(list {:type "ident" :value "a"} {:type "ident" :value "b"}))
(st-test
"multi-line comment"
(st-toks "\"line1\nline2\"42")
(list {:type "number" :value 42}))
;; ── 12. Compound expressions ──
(st-test
"block with params"
(st-toks "[:a :b | a + b]")
(list
{:type "lbracket" :value "["}
{:type "colon" :value ":"}
{:type "ident" :value "a"}
{:type "colon" :value ":"}
{:type "ident" :value "b"}
{:type "bar" :value "|"}
{:type "ident" :value "a"}
{:type "binary" :value "+"}
{:type "ident" :value "b"}
{:type "rbracket" :value "]"}))
(st-test
"cascade"
(st-toks "x m1; m2")
(list
{:type "ident" :value "x"}
{:type "ident" :value "m1"}
{:type "semi" :value ";"}
{:type "ident" :value "m2"}))
(st-test
"method body return"
(st-toks "^ self foo")
(list
{:type "caret" :value "^"}
{:type "ident" :value "self"}
{:type "ident" :value "foo"}))
(st-test
"class declaration head"
(st-toks "Object subclass: #Foo")
(list
{:type "ident" :value "Object"}
{:type "keyword" :value "subclass:"}
{:type "symbol" :value "Foo"}))
(st-test
"temp declaration"
(st-toks "| t1 t2 |")
(list
{:type "bar" :value "|"}
{:type "ident" :value "t1"}
{:type "ident" :value "t2"}
{:type "bar" :value "|"}))
(st-test
"chunk separator"
(st-toks "Foo bar !")
(list
{:type "ident" :value "Foo"}
{:type "ident" :value "bar"}
{:type "bang" :value "!"}))
(st-test
"keyword call with binary precedence"
(st-toks "x foo: 1 + 2")
(list
{:type "ident" :value "x"}
{:type "keyword" :value "foo:"}
{:type "number" :value 1}
{:type "binary" :value "+"}
{:type "number" :value 2}))
(list st-test-pass st-test-fail)

View File

@@ -1,366 +0,0 @@
;; Smalltalk tokenizer.
;;
;; Token types:
;; ident identifier (foo, Foo, _x)
;; keyword selector keyword (foo:) — value is "foo:" with the colon
;; binary binary selector chars run together (+, ==, ->, <=, ~=, ...)
;; number integer or float; radix integers like 16rFF supported
;; string 'hello''world' style
;; char $c
;; symbol #foo, #foo:bar:, #+, #'with spaces'
;; array-open #(
;; byte-array-open #[
;; lparen rparen lbracket rbracket lbrace rbrace
;; period semi bar caret colon assign bang
;; eof
;;
;; Comments "…" are skipped.
(define st-make-token (fn (type value pos) {:type type :value value :pos pos}))
(define st-digit? (fn (c) (and (not (= c nil)) (>= c "0") (<= c "9"))))
(define
st-letter?
(fn
(c)
(and
(not (= c nil))
(or (and (>= c "a") (<= c "z")) (and (>= c "A") (<= c "Z"))))))
(define st-ident-start? (fn (c) (or (st-letter? c) (= c "_"))))
(define st-ident-char? (fn (c) (or (st-ident-start? c) (st-digit? c))))
(define st-ws? (fn (c) (or (= c " ") (= c "\t") (= c "\n") (= c "\r"))))
(define
st-binary-chars
(list "+" "-" "*" "/" "\\" "~" "<" ">" "=" "@" "%" "&" "?" ","))
(define
st-binary-char?
(fn (c) (and (not (= c nil)) (contains? st-binary-chars c))))
(define
st-radix-digit?
(fn
(c)
(and
(not (= c nil))
(or (st-digit? c) (and (>= c "A") (<= c "Z"))))))
(define
st-tokenize
(fn
(src)
(let
((tokens (list)) (pos 0) (src-len (len src)))
(define
pk
(fn
(offset)
(if (< (+ pos offset) src-len) (nth src (+ pos offset)) nil)))
(define cur (fn () (pk 0)))
(define advance! (fn (n) (set! pos (+ pos n))))
(define
push!
(fn
(type value start)
(append! tokens (st-make-token type value start))))
(define
skip-comment!
(fn
()
(cond
((>= pos src-len) nil)
((= (cur) "\"") (advance! 1))
(else (begin (advance! 1) (skip-comment!))))))
(define
skip-ws!
(fn
()
(cond
((>= pos src-len) nil)
((st-ws? (cur)) (begin (advance! 1) (skip-ws!)))
((= (cur) "\"") (begin (advance! 1) (skip-comment!) (skip-ws!)))
(else nil))))
(define
read-ident-chars!
(fn
()
(when
(and (< pos src-len) (st-ident-char? (cur)))
(begin (advance! 1) (read-ident-chars!)))))
(define
read-decimal-digits!
(fn
()
(when
(and (< pos src-len) (st-digit? (cur)))
(begin (advance! 1) (read-decimal-digits!)))))
(define
read-radix-digits!
(fn
()
(when
(and (< pos src-len) (st-radix-digit? (cur)))
(begin (advance! 1) (read-radix-digits!)))))
(define
read-exp-part!
(fn
()
(when
(and
(< pos src-len)
(or (= (cur) "e") (= (cur) "E"))
(let
((p1 (pk 1)) (p2 (pk 2)))
(or
(st-digit? p1)
(and (or (= p1 "+") (= p1 "-")) (st-digit? p2)))))
(begin
(advance! 1)
(when
(and (< pos src-len) (or (= (cur) "+") (= (cur) "-")))
(advance! 1))
(read-decimal-digits!)))))
(define
read-number
(fn
(start)
(begin
(read-decimal-digits!)
(cond
((and (< pos src-len) (= (cur) "r"))
(let
((base-str (slice src start pos)))
(begin
(advance! 1)
(let
((rstart pos))
(begin
(read-radix-digits!)
(let
((digits (slice src rstart pos)))
{:radix (parse-number base-str)
:digits digits
:value (parse-radix base-str digits)
:kind "radix"}))))))
((and
(< pos src-len)
(= (cur) ".")
(st-digit? (pk 1)))
(begin
(advance! 1)
(read-decimal-digits!)
(read-exp-part!)
(parse-number (slice src start pos))))
(else
(begin
(read-exp-part!)
(parse-number (slice src start pos))))))))
(define
parse-radix
(fn
(base-str digits)
(let
((base (parse-number base-str))
(chars digits)
(n-len (len digits))
(idx 0)
(acc 0))
(begin
(define
rd-loop
(fn
()
(when
(< idx n-len)
(let
((c (nth chars idx)))
(let
((d (cond
((and (>= c "0") (<= c "9")) (- (char-code c) 48))
((and (>= c "A") (<= c "Z")) (- (char-code c) 55))
(else 0))))
(begin
(set! acc (+ (* acc base) d))
(set! idx (+ idx 1))
(rd-loop)))))))
(rd-loop)
acc))))
(define
read-string
(fn
()
(let
((chars (list)))
(begin
(advance! 1)
(define
loop
(fn
()
(cond
((>= pos src-len) nil)
((= (cur) "'")
(cond
((= (pk 1) "'")
(begin
(append! chars "'")
(advance! 2)
(loop)))
(else (advance! 1))))
(else
(begin (append! chars (cur)) (advance! 1) (loop))))))
(loop)
(join "" chars)))))
(define
read-binary-run!
(fn
()
(let
((start pos))
(begin
(define
bin-loop
(fn
()
(when
(and (< pos src-len) (st-binary-char? (cur)))
(begin (advance! 1) (bin-loop)))))
(bin-loop)
(slice src start pos)))))
(define
read-symbol
(fn
(start)
(cond
;; Quoted symbol: #'whatever'
((= (cur) "'")
(let ((s (read-string))) (push! "symbol" s start)))
;; Binary-char symbol: #+, #==, #->, #|
((or (st-binary-char? (cur)) (= (cur) "|"))
(let ((b (read-binary-run!)))
(cond
((= b "")
;; lone | wasn't binary; consume it
(begin (advance! 1) (push! "symbol" "|" start)))
(else (push! "symbol" b start)))))
;; Identifier or keyword chain: #foo, #foo:bar:
((st-ident-start? (cur))
(let ((id-start pos))
(begin
(read-ident-chars!)
(define
kw-loop
(fn
()
(when
(and (< pos src-len) (= (cur) ":"))
(begin
(advance! 1)
(when
(and (< pos src-len) (st-ident-start? (cur)))
(begin (read-ident-chars!) (kw-loop)))))))
(kw-loop)
(push! "symbol" (slice src id-start pos) start))))
(else
(error
(str "st-tokenize: bad symbol at " pos))))))
(define
step
(fn
()
(begin
(skip-ws!)
(when
(< pos src-len)
(let
((start pos) (c (cur)))
(cond
;; Identifier or keyword
((st-ident-start? c)
(begin
(read-ident-chars!)
(let
((word (slice src start pos)))
(cond
;; ident immediately followed by ':' (and not ':=') => keyword
((and
(< pos src-len)
(= (cur) ":")
(not (= (pk 1) "=")))
(begin
(advance! 1)
(push!
"keyword"
(str word ":")
start)))
(else (push! "ident" word start))))
(step)))
;; Number
((st-digit? c)
(let
((v (read-number start)))
(begin (push! "number" v start) (step))))
;; String
((= c "'")
(let
((s (read-string)))
(begin (push! "string" s start) (step))))
;; Character literal
((= c "$")
(cond
((>= (+ pos 1) src-len)
(error (str "st-tokenize: $ at end of input")))
(else
(begin
(advance! 1)
(push! "char" (cur) start)
(advance! 1)
(step)))))
;; Symbol or array literal
((= c "#")
(cond
((= (pk 1) "(")
(begin (advance! 2) (push! "array-open" "#(" start) (step)))
((= (pk 1) "[")
(begin (advance! 2) (push! "byte-array-open" "#[" start) (step)))
(else
(begin (advance! 1) (read-symbol start) (step)))))
;; Assignment := or bare colon
((= c ":")
(cond
((= (pk 1) "=")
(begin (advance! 2) (push! "assign" ":=" start) (step)))
(else
(begin (advance! 1) (push! "colon" ":" start) (step)))))
;; Single-char structural punctuation
((= c "(") (begin (advance! 1) (push! "lparen" "(" start) (step)))
((= c ")") (begin (advance! 1) (push! "rparen" ")" start) (step)))
((= c "[") (begin (advance! 1) (push! "lbracket" "[" start) (step)))
((= c "]") (begin (advance! 1) (push! "rbracket" "]" start) (step)))
((= c "{") (begin (advance! 1) (push! "lbrace" "{" start) (step)))
((= c "}") (begin (advance! 1) (push! "rbrace" "}" start) (step)))
((= c ".") (begin (advance! 1) (push! "period" "." start) (step)))
((= c ";") (begin (advance! 1) (push! "semi" ";" start) (step)))
((= c "|") (begin (advance! 1) (push! "bar" "|" start) (step)))
((= c "^") (begin (advance! 1) (push! "caret" "^" start) (step)))
((= c "!") (begin (advance! 1) (push! "bang" "!" start) (step)))
;; Binary selector run
((st-binary-char? c)
(let
((b (read-binary-run!)))
(begin (push! "binary" b start) (step))))
(else
(error
(str
"st-tokenize: unexpected char "
c
" at "
pos)))))))))
(step)
(push! "eof" nil pos)
tokens)))

View File

@@ -1,77 +0,0 @@
# smalltalk-on-sx loop agent (single agent, queue-driven)
Role: iterates `plans/smalltalk-on-sx.md` forever. Message-passing OO + **blocks with non-local return** on delimited continuations. Non-local return is the headline showcase — every other Smalltalk reinvents it on the host stack; on SX it falls out of the captured method-return continuation.
```
description: smalltalk-on-sx queue loop
subagent_type: general-purpose
run_in_background: true
isolation: worktree
```
## Prompt
You are the sole background agent working `/root/rose-ash/plans/smalltalk-on-sx.md`. Isolated worktree, forever, one commit per feature. Never push.
## Restart baseline — check before iterating
1. Read `plans/smalltalk-on-sx.md` — roadmap + Progress log.
2. `ls lib/smalltalk/` — pick up from the most advanced file.
3. If `lib/smalltalk/tests/*.sx` exist, run them. Green before new work.
4. If `lib/smalltalk/scoreboard.md` exists, that's your baseline.
## The queue
Phase order per `plans/smalltalk-on-sx.md`:
- **Phase 1** — tokenizer + parser (chunk format, identifiers, keywords `foo:`, binary selectors, `#sym`, `#(…)`, `$c`, blocks `[:a | …]`, cascades, message precedence)
- **Phase 2** — object model + sequential eval (class table bootstrap, message dispatch, `super`, `doesNotUnderstand:`, instance variables)
- **Phase 3** — **THE SHOWCASE**: blocks with non-local return via captured method-return continuation. `whileTrue:` / `ifTrue:ifFalse:` as block sends. 5 classic programs (eight-queens, quicksort, mandelbrot, life, fibonacci) green.
- **Phase 4** — reflection + MOP: `perform:`, `respondsTo:`, runtime method addition, `becomeForward:`, `Exception` / `on:do:` / `ensure:` on top of `handler-bind`/`raise`
- **Phase 5** — collections + numeric tower + streams
- **Phase 6** — port SUnit, vendor Pharo Kernel-Tests slice, drive corpus to 200+
- **Phase 7** — speed (optional): inline caching, block intrinsification
Within a phase, pick the checkbox that unlocks the most tests per effort.
Every iteration: implement → test → commit → tick `[ ]` → Progress log → next.
## Ground rules (hard)
- **Scope:** only `lib/smalltalk/**` and `plans/smalltalk-on-sx.md`. Do **not** edit `spec/`, `hosts/`, `shared/`, other `lib/<lang>/` dirs, `lib/stdlib.sx`, or `lib/` root. Smalltalk primitives go in `lib/smalltalk/runtime.sx`.
- **NEVER call `sx_build`.** 600s watchdog. If sx_server binary broken → Blockers entry, stop.
- **Shared-file issues** → plan's Blockers with minimal repro.
- **Delimited continuations** are in `lib/callcc.sx` + `spec/evaluator.sx` Step 5. `sx_summarise` spec/evaluator.sx first — 2300+ lines.
- **SX files:** `sx-tree` MCP tools ONLY. `sx_validate` after edits.
- **Worktree:** commit locally. Never push. Never touch `main`.
- **Commit granularity:** one feature per commit.
- **Plan file:** update Progress log + tick boxes every commit.
## Smalltalk-specific gotchas
- **Method invocation captures `^k`** — the return continuation. Bind it as the block's escape token. `^expr` from inside any nested block invokes that captured `^k`. Escape past method return raises `BlockContext>>cannotReturn:`.
- **Blocks are lambdas + escape token**, not bare lambdas. `value`/`value:`/… invoke the lambda; `^` invokes the escape.
- **`ifTrue:` / `ifFalse:` / `whileTrue:` are ordinary block sends** — no special form. The runtime intrinsifies them in the JIT path (Tier 1 of bytecode expansion already covers this pattern).
- **Cascade** `r m1; m2; m3` desugars to `(let ((tmp r)) (st-send tmp 'm1 ()) (st-send tmp 'm2 ()) (st-send tmp 'm3 ()))`. Result is the cascade's last send (or first, depending on parser variant — pick one and document).
- **`super` send** looks up starting from the *defining* class's superclass, not the receiver class. Stash the defining class on the method record.
- **Selectors are interned symbols.** Use SX symbols.
- **Receiver dispatch:** tagged ints / floats / strings / symbols / `nil` / `true` / `false` aren't boxed. Their classes (`SmallInteger`, `Float`, `String`, `Symbol`, `UndefinedObject`, `True`, `False`) are looked up by SX type-of, not by an `:class` field.
- **Method precedence:** unary > binary > keyword. `3 + 4 factorial` is `3 + (4 factorial)`. `a foo: b bar` is `a foo: (b bar)` (keyword absorbs trailing unary).
- **Image / fileIn / become: between sessions** = out of scope. One-way `becomeForward:` only.
- **Test corpus:** ~200 hand-written + a slice of Pharo Kernel-Tests. Place programs in `lib/smalltalk/tests/programs/`.
## General gotchas (all loops)
- SX `do` = R7RS iteration. Use `begin` for multi-expr sequences.
- `cond`/`when`/`let` clauses evaluate only the last expr.
- `type-of` on user fn returns `"lambda"`.
- Shell heredoc `||` gets eaten — escape or use `case`.
## Style
- No comments in `.sx` unless non-obvious.
- No new planning docs — update `plans/smalltalk-on-sx.md` inline.
- Short, factual commit messages (`smalltalk: tokenizer + 56 tests`).
- One feature per iteration. Commit. Log. Next.
Go. Read the plan; find first `[ ]`; implement.

View File

@@ -55,25 +55,32 @@ Key mappings:
### Phase 1 — tokenizer + parser + layout rule
- [x] Tokenizer: reserved words, qualified names, operators, numbers (int, float, Rational later), chars/strings, comments (`--` and `{-` nested)
- [ ] Layout algorithm: turn indentation into virtual `{`, `;`, `}` tokens per Haskell 98 §10.3
- [ ] Parser: modules, imports (stub), top-level decls, type sigs, function clauses with patterns + guards + where-clauses, expressions with operator precedence, lambdas, `let`, `if`, `case`, `do`, list comp, sections
- [ ] AST design modelled on GHC's HsSyn at a surface level
- [x] Layout algorithm: turn indentation into virtual `{`, `;`, `}` tokens per Haskell 98 §10.3
- Parser (split into sub-items — implement one per iteration):
- [x] Expressions: atoms, parens, tuples, lists, ranges, application, infix with full Haskell-98 precedence table, unary `-`, backtick operators, lambdas, `if`, `let`
- [x] `case … of` and `do`-notation expressions (plus minimal patterns needed for arms/binds: var, wildcard, literal, 0-arity and applied constructor, tuple, list)
- [x] Patterns — full: `as` patterns, nested, negative literal, `~` lazy, infix constructor (`:` / consym), extend lambdas/let with non-var patterns
- [x] Top-level decls: function clauses (simple — no guards/where yet), pattern bindings, multi-name type signatures, `data` with type vars and recursive constructors, `type` synonyms, `newtype`, fixity (`infix`/`infixl`/`infixr` with optional precedence, comma-separated ops, backtick names). Types: vars / constructors / application / `->` (right-assoc) / tuples / lists. `hk-parse-top` entry.
- [x] `where` clauses + guards (on fun-clauses, case alts, and let/do-let bindings — with the let funclause shorthand `let f x = …` now supported)
- [x] Module header + imports — `module NAME [exports] where …`, qualified/as/hiding/explicit imports, operator exports, `module Foo` exports, dotted names, headerless-with-imports
- [x] List comprehensions + operator sections — `(op)` / `(op e)` / `(e op)` (excluding `-` from right sections), `[e | q1, q2, …]` with `q-gen` / `q-guard` / `q-let` qualifiers
- [x] AST design modelled on GHC's HsSyn at a surface level — keyword-tagged lists cover modules/imports/decls/types/patterns/expressions; see parser.sx docstrings for the full node catalogue
- [x] Unit tests in `lib/haskell/tests/parse.sx` (43 tokenizer tests, all green)
### Phase 2 — desugar + eager-ish eval + ADTs (untyped)
- [ ] Desugar: guards → nested `if`s; `where``let`; list comp → `concatMap`-based; do-notation stays for now (desugared in phase 3)
- [ ] `data` declarations register constructors in runtime
- [ ] Pattern match (tag-based, value-level): atoms, vars, wildcards, constructor patterns, `as` patterns, nested
- [ ] Evaluator (still strict internally — laziness in phase 3): `let`, `lambda`, application, `case`, literals, constructors
- [ ] 30+ eval tests in `lib/haskell/tests/eval.sx`
- [x] Desugar: guards → nested `if`s; `where``let`; list comp → `concatMap`-based; do-notation stays for now (desugared in phase 3)
- [x] `data` declarations register constructors in runtime
- [x] Pattern match (tag-based, value-level): atoms, vars, wildcards, constructor patterns, `as` patterns, nested
- [x] Evaluator (still strict internally — laziness in phase 3): `let`, `lambda`, application, `case`, literals, constructors
- [x] 30+ eval tests in `lib/haskell/tests/eval.sx`
### Phase 3 — laziness + classic programs
- [ ] Transpile to thunk-wrapped SX: every application arg becomes `(make-thunk (lambda () <arg>))`
- [ ] `force` = SX eval-thunk-to-WHNF primitive
- [ ] Pattern match forces scrutinee before matching
- [ ] Infinite structures: `repeat x`, `iterate f x`, `[1..]`, Fibonacci stream, sieve of Eratosthenes
- [ ] `seq`, `deepseq` from Prelude
- [ ] Do-notation for a stub `IO` monad (just threading, no real side effects yet)
- [x] Transpile to thunk-wrapped SX: every application arg becomes `(make-thunk (lambda () <arg>))`
- [x] `force` = SX eval-thunk-to-WHNF primitive
- [x] Pattern match forces scrutinee before matching
- [x] Infinite structures: `repeat x`, `iterate f x`, `[1..]`, Fibonacci stream (sieve deferred — needs lazy `++` and is exercised under `Classic programs`)
- [x] `seq`, `deepseq` from Prelude
- [x] Do-notation for a stub `IO` monad (just threading, no real side effects yet)
- [ ] Classic programs in `lib/haskell/tests/programs/`:
- [ ] `fib.hs` — infinite Fibonacci stream
- [ ] `sieve.hs` — lazy sieve of Eratosthenes
@@ -107,6 +114,381 @@ Key mappings:
_Newest first._
- **2026-04-25** — Phase 3 do-notation + stub IO monad. Added a
`hk-desugar-do` pass that follows Haskell 98 §3.14 verbatim:
`do { e } = e`, `do { e ; ss } = e >> do { ss }`,
`do { p <- e ; ss } = e >>= \p -> do { ss }`, and
`do { let ds ; ss } = let ds in do { ss }`. The desugarer's
`:do` branch now invokes this pass directly so the surface
AST forms (`:do-expr`, `:do-bind`, `:do-let`) never reach the
evaluator. IO is represented as a tagged value
`("IO" payload)``return` (lazy builtin) wraps; `>>=` (lazy
builtin) forces the action, unwraps, and calls the bound
function on the payload; `>>` (lazy builtin) forces the
action and returns the second one. All three are non-strict
in their action arguments so deeply nested do-blocks don't
walk the whole chain at construction time. 14 new tests in
`lib/haskell/tests/do-io.sx` cover single-stmt do, single
and multi-bind, `>>` sequencing (last action wins), do-let
(single, multi, interleaved with bind), bind-to-`Just`,
bind-to-tuple, do inside a top-level fun, nested do, and
using `(>>=)`/`(>>)` directly as functions. 382/382 green.
- **2026-04-25** — Phase 3 `seq` + `deepseq`. Built-ins were strict
in all args by default (every collected thunk forced before
invoking the underlying SX fn) — that defeats `seq`'s purpose,
which is strict in its first argument and lazy in its second.
Added a tiny `lazy` flag on the builtin record (set by a new
`hk-mk-lazy-builtin` constructor) and routed `hk-apply-builtin`
to skip the auto-force when the flag is true. `seq a b` calls
`hk-force a` then returns `b` unchanged so its laziness is
preserved; `deepseq` does the same with `hk-deep-force`. 9 new
tests in `lib/haskell/tests/seq.sx` cover primitive, computed,
and let-bound first args, deepseq on a list / `Just` /
tuple, seq inside arithmetic, seq via a fun-clause, and
`[seq 1 10, seq 2 20]` to confirm seq composes inside list
literals. The lazy-when-unused negative case is also tested:
`let x = error "never" in 42 == 42`. 368/368 green.
- **2026-04-24** — Phase 3 infinite structures + Prelude. Two
evaluator changes turn the lazy primitives into a working
language:
1. Op-form `:` is now non-strict in both args — `hk-eval-op`
special-cases it before the eager force-and-binop path, so a
cons-cell holds two thunks. This is what makes `repeat x =
x : repeat x`, `iterate f x = x : iterate f (f x)`, and the
classic `fibs = 0 : 1 : zipWith plus fibs (tail fibs)`
terminate when only a finite prefix is consumed.
2. Operators are now first-class values via a small
`hk-make-binop-builtin` helper, so `(+)`, `(*)`, `(==)` etc.
can be passed to `zipWith` and `map`.
Added range support across parser + evaluator: `[from..to]` and
`[from,next..to]` evaluate eagerly via `hk-build-range` (handles
step direction); `[from..]` parses to a new `:range-from` node
that the evaluator desugars to `iterate (+ 1) from`. New
`hk-load-into!` runs the regular pipeline (parse → desugar →
register data → bind decls) on a source string, and `hk-init-env`
preloads `hk-prelude-src` with the Phase-3 Prelude:
`head`, `tail`, `fst`, `snd`, `take`, `drop`, `repeat`, `iterate`,
`length`, `map`, `filter`, `zipWith`, plus `fibs` and `plus`.
25 new tests in `lib/haskell/tests/infinite.sx`, including
`take 10 fibs == [0,1,1,2,3,5,8,13,21,34]`,
`head (drop 99 [1..])`, `iterate (\x -> x * 2) 1` powers of two,
user-defined `ones = 1 : ones`, `naturalsFrom`, range edge cases,
composed `map`/`filter`, and a custom `mySum`. 359/359 green.
Sieve of Eratosthenes is deferred — it needs lazy `++` plus a
`mod` primitive — and lives under `Classic programs` anyway.
- **2026-04-24** — Phase 3 laziness foundation. Added a thunk type to
`lib/haskell/eval.sx` (`hk-mk-thunk` / `hk-is-thunk?`) backed by a
one-shot memoizing `hk-force` that evaluates the deferred AST, then
flips a `forced` flag and caches the value on the thunk dict; the
shared `hk-deep-force` walks the result tree at the test/output
boundary. Three single-line wiring changes in the evaluator make
every application argument lazy: `:app` now wraps its argument in
`hk-mk-thunk` rather than evaluating it. To preserve correctness
where values must be inspected, `hk-apply`, `hk-eval-op`,
`hk-eval-if`, `hk-eval-case`, and `hk-eval` for `:neg` now force
their operand. `hk-apply-builtin` forces every collected arg
before invoking the underlying SX fn so built-ins (`error`, `not`,
`id`) stay strict. The pattern matcher in `match.sx` now forces
the scrutinee just-in-time only for patterns that need to inspect
shape — `p-wild`, `p-var`, `p-as`, and `p-lazy` are no-force
paths, so the value flows through as a thunk and binding
preserves laziness. `hk-match-list-pat` forces at every cons-spine
step. 6 new lazy-specific tests in `lib/haskell/tests/eval.sx`
verify that `(\x y -> x) 1 (error …)` and `(\x y -> y) (error …) 99`
return without diverging, that `case Just (error …) of Just _ -> 7`
short-circuits, that `const` drops its second arg, that
`myHead (1 : error … : [])` returns 1 without touching the tail,
and that `Just (error …)` survives a wildcard-arm `case`. 333/333
green, all prior eval tests preserved by deep-forcing the result
in `hk-eval-expr-source` and `hk-prog-val`.
- **2026-04-24** — Phase 2 evaluator (`lib/haskell/eval.sx`) — ties
the whole pipeline together. Strict semantics throughout (laziness
is Phase 3). Function values are tagged dicts: `closure`,
`multi`(fun), `con-partial`, `builtin`. `hk-apply` unifies dispatch
across all four; closures and multifuns curry one argument at a
time, multifuns trying each clause's pat-list in order once arity
is reached. Top-level `hk-bind-decls!` is three-pass —
collect groups + pre-seed names → install multifuns (so closures
observe later names) → eval 0-arity bodies and pat-binds — making
forward and mutually recursive references work. `hk-eval-let` does
the same trick with a mutable child env. Built-ins:
`error`/`not`/`id`, plus `otherwise = True`. Operators wired:
arithmetic, comparison (returning Bool conses), `&&`, `||`, `:`,
`++`. Sections evaluate the captured operand once and return a
closure synthesized via the existing AST. `hk-eval-program`
registers data decls then binds, returning the env; `hk-run`
fetches `main` if present. Also extended `runtime.sx` to
pre-register the standard Prelude conses (`Maybe`, `Either`,
`Ordering`) so expression-level eval doesn't need a leading
`data` decl. 48 new tests in `lib/haskell/tests/eval.sx` cover
literals, arithmetic precedence, comparison/Bool, `if`, `let`
(incl. recursive factorial), lambdas (incl. constructor pattern
args), constructors, `case` (Just/Nothing/literal/tuple/wildcard),
list literals + cons + `++`, tuples, sections, multi-clause
top-level (factorial, list length via cons pattern, Maybe handler
with default), user-defined `data` with case-style matching, a
binary-tree height program, currying, higher-order (`twice`),
short-circuit `error` via `if`, and the three built-ins. 329/329
green. Phase 2 is now complete; Phase 3 (laziness) is next.
- **2026-04-24** — Phase 2: value-level pattern matcher
(`lib/haskell/match.sx`). Core entry `hk-match pat val env` returns
an extended env dict on success or `nil` on failure (uses `assoc`
rather than `dict-set!` so failed branches never pollute the
caller's env). Constructor values are tagged lists with the
constructor name as the first element; tuples use the tag `"Tuple"`,
lists are chained `(":" h t)` cons cells terminated by `("[]")`.
Value builders `hk-mk-con` / `hk-mk-tuple` / `hk-mk-nil` /
`hk-mk-cons` / `hk-mk-list` keep tests readable. The matcher
handles every pattern node the parser emits:
- `:p-wild` (always matches), `:p-var` (binds), `:p-int` /
`:p-float` / `:p-string` / `:p-char` (literal equality)
- `:p-as` (sub-match then bind whole), `:p-lazy` (eager for now;
laziness wired in phase 3)
- `:p-con` with arity check + recursive arg matching, including
deeply nested patterns and infix `:` cons (uses the same
code path as named constructors)
- `:p-tuple` against `"Tuple"` values, `:p-list` against an
exact-length cons spine.
Helper `hk-parse-pat-source` lifts a real Haskell pattern out of
`case _ of <pat> -> 0`, letting tests drive against parser output.
31 new tests in `lib/haskell/tests/match.sx` cover atomic
patterns, success/failure for each con/tuple/list shape, nested
`Just (Just x)`, cons-vs-empty, `as` over con / wildcard /
failing-sub, `~` lazy, plus four parser-driven cases (`Just x`,
`x : xs`, `(a, b)`, `n@(Just x)`). 281/281 green.
- **2026-04-24** — Phase 2: runtime constructor registry
(`lib/haskell/runtime.sx`). A mutable dict `hk-constructors` keyed
by constructor name, each entry carrying arity and owning type.
`hk-register-data!` walks a `:data` AST and registers every
`:con-def` with its arity (= number of field types) and the type
name; `hk-register-newtype!` does the one-constructor variant;
`hk-register-decls!` / `hk-register-program!` filter a decls list
(or a `:program` / `:module` AST) and call the appropriate
registrar. `hk-load-source!` composes it with `hk-core`
(tokenize → layout → parse → desugar → register). Pre-registers
five built-ins tied to Haskell syntactic forms: `True` / `False`
(Bool), `[]` and `:` (List), `()` (Unit) — everything else comes
from user declarations or the eventual Prelude. Query helpers:
`hk-is-con?`, `hk-con-arity`, `hk-con-type`, `hk-con-names`. 24
new tests in `lib/haskell/tests/runtime.sx` cover each built-in
(arity + type), unknown-name probes, registration of `MyBool` /
`Maybe` / `Either` / recursive `Tree` / `newtype Age`, multi-data
programs, a module-header body, ignoring non-data decls, and
last-wins re-registration. 250/250 green.
- **2026-04-24** — Phase 2 kicks off with `lib/haskell/desugar.sx` — a
tree-walking rewriter that eliminates the three surface-only forms
produced by the parser, leaving a smaller core AST for the evaluator:
- `:where BODY DECLS``:let DECLS BODY`
- `:guarded ((:guard C1 E1) (:guard C2 E2) …)` → right-folded
`(:if C1 E1 (:if C2 E2 … (:app (:var "error") (:string "…"))))`
- `:list-comp E QUALS` → Haskell 98 §3.11 translation:
empty quals → `(:list (E))`, `:q-guard``(:if … (:list (E)) (:list ()))`,
`:q-gen PAT SRC``(concatMap (\PAT -> …) SRC)`, `:q-let BINDS`
`(:let BINDS …)`. Nested generators compile to nested concatMap.
Every other expression, decl, pattern, and type node is recursed
into and passed through unchanged. Public entries `hk-desugar`,
`hk-core` (tokenize → layout → parse → desugar on a module), and
`hk-core-expr` (the same for an expression). 15 new tests in
`lib/haskell/tests/desugar.sx` cover two- and three-way guards,
case-alt guards, single/multi-binding `where`, guards + `where`
combined, the four list-comprehension cases (single-gen, gen +
filter, gen + let, nested gens), and pass-through for literals,
lambdas, simple fun-clauses, `data` decls, and a module header
wrapping a guarded function. 226/226 green.
- **2026-04-24** — Phase 1 parser is now complete. This iteration adds
operator sections and list comprehensions, the two remaining
aexp-level forms, plus ticks the “AST design” item (the keyword-
tagged list shape has accumulated a full HsSyn-level surface).
Changes:
- `hk-parse-infix` now bails on `op )` without consuming the op, so
the paren parser can claim it as a left section.
- `hk-parse-parens` rewritten to recognise five new forms:
`()` (unit), `(op)``(:var OP)`, `(op e)``(:sect-right OP E)`
(excluded for `-` so that `(- 5)` stays `(:neg 5)`), `(e op)`
`(:sect-left OP E)`, plus regular parens and tuples. Works for
varsym, consym, reservedop `:`, and backtick-quoted varids.
- `hk-section-op-info` inspects the current token and returns a
`{:name :len}` dict, so the same logic handles 1-token ops and
3-token backtick ops uniformly.
- `hk-parse-list-lit` now recognises a `|` after the first element
and dispatches to `hk-parse-qual` per qualifier (comma-separated),
producing `(:list-comp EXPR QUALS)`. Qualifiers are:
`(:q-gen PAT EXPR)` when a paren-balanced lookahead
(`hk-comp-qual-is-gen?`) finds `<-` before the next `,`/`]`,
`(:q-let BINDS)` for `let …`, and `(:q-guard EXPR)` otherwise.
- `hk-parse-comp-let` accepts `]` or `,` as an implicit block close
(single-line comprehensions never see layout's vrbrace before the
qualifier terminator arrives); explicit `{ }` still closes
strictly.
22 new tests in `lib/haskell/tests/parser-sect-comp.sx` cover
op-references (inc. `(-)`, `(:)`, backtick), right sections (inc.
backtick), left sections, the `(- 5)``:neg` corner, plain parens
and tuples, six comprehension shapes (simple, filter, let,
nested-generators, constructor pattern bind, tuple pattern bind,
and a three-qualifier mix). 211/211 green.
- **2026-04-24** — Phase 1: module header + imports. Added
`hk-parse-module-header`, `hk-parse-import`, plus shared helpers for
import/export entity lists (`hk-parse-ent`, `hk-parse-ent-member`,
`hk-parse-ent-list`). New AST:
- `(:module NAME EXPORTS IMPORTS DECLS)` — NAME `nil` means no header,
EXPORTS `nil` means no export list (distinct from empty `()`)
- `(:import QUALIFIED NAME AS SPEC)` — QUALIFIED bool, AS alias or nil,
SPEC nil / `(:spec-items ENTS)` / `(:spec-hiding ENTS)`
- Entity refs: `:ent-var`, `:ent-all` (`Tycon(..)`), `:ent-with`
(`Tycon(m1, m2, …)`), `:ent-module` (exports only).
`hk-parse-program` now dispatches on the leading token: `module`
keyword → full header-plus-body parse (consuming the `where` layout
brace around the module body); otherwise collect any leading
`import` decls and then remaining decls with the existing logic.
The outer shell is `(:module …)` as soon as any header or import is
present, and stays as `(:program DECLS)` otherwise — preserving every
previous test expectation untouched. Handles operator exports `((+:))`,
dotted module names (`Data.Map`), and the Haskell-98 context-sensitive
keywords `qualified`/`as`/`hiding` (all lexed as ordinary varids and
matched only in import position). 16 new tests in
`lib/haskell/tests/parser-module.sx` covering simple/exports/empty
headers, dotted names, operator exports, `module Foo` exports,
qualified/aliased/items/hiding imports, and a headerless-with-imports
file. 189/189 green.
- **2026-04-24** — Phase 1: guards + where clauses. Factored a single
`hk-parse-rhs sep` that all body-producing sites now share: it reads
a plain `sep expr` body or a chain of `| cond sep expr` guards, then
— regardless of which form — looks for an optional `where` block and
wraps accordingly. AST additions:
- `:guarded GUARDS` where each GUARD is `:guard COND EXPR`
- `:where BODY DECLS` where BODY is a plain expr or a `:guarded`
Both can nest (guards inside where). `hk-parse-alt` now routes through
`hk-parse-rhs "->"`, `hk-parse-fun-clause` and `hk-parse-bind` through
`hk-parse-rhs "="`. `hk-parse-where-decls` reuses `hk-parse-decl` so
where-blocks accept any decl form (signatures, fixity, nested funs).
As a side effect, `hk-parse-bind` now also picks up the Haskell-native
`let f x = …` funclause shorthand: a varid followed by one or more
apats produces `(:fun-clause NAME APATS BODY)` instead of a
`(:bind (:p-var …) …)` — keeping the simple `let x = e` shape
unchanged for existing tests. 11 new tests in
`lib/haskell/tests/parser-guards-where.sx` cover two- and three-way
guards, mixed guarded + equality clauses, single- and multi-binding
where blocks, guards plus where, case-alt guards, case-alt where,
let with funclause shorthand, let with guards, and a where containing
a type signature alongside a fun-clause. 173/173 green.
- **2026-04-24** — Phase 1: top-level decls. Refactored `hk-parse-expr` into a
`hk-parser tokens mode` with `:expr` / `:module` dispatch so the big lexical
state is shared (peek/advance/pat/expr helpers all reachable); added public
wrappers `hk-parse-expr`, `hk-parse-module`, and source-level entry
`hk-parse-top`. New type parser (`hk-parse-type` / `hk-parse-btype` /
`hk-parse-atype`): type variables (`:t-var`), type constructors (`:t-con`),
type application (`:t-app`, left-assoc), right-associative function arrow
(`:t-fun`), unit/tuples (`:t-tuple`), and lists (`:t-list`). New decl parser
(`hk-parse-decl` / `hk-parse-program`) producing a `(:program DECLS)` shell:
- `:type-sig NAMES TYPE` — comma-separated multi-name support
- `:fun-clause NAME APATS BODY` — patterns for args, body via existing expr
- `:pat-bind PAT BODY` — top-level pattern bindings like `(a, b) = pair`
- `:data NAME TVARS CONS` with `:con-def CNAME FIELDS` for nullary and
multi-arg constructors, including recursive references
- `:type-syn NAME TVARS TYPE`, `:newtype NAME TVARS CNAME FIELD`
- `:fixity ASSOC PREC OPS` — assoc one of `"l"`/`"r"`/`"n"`, default prec 9,
comma-separated operator names, including backtick-quoted varids.
Sig vs fun-clause disambiguated by a paren-balanced top-level scan for
`::` before the next `;`/`}` (`hk-has-top-dcolon?`). 24 new tests in
`lib/haskell/tests/parser-decls.sx` cover all decl forms, signatures with
application / tuples / lists / right-assoc arrows, nullary and recursive
data types, multi-clause functions, and a mixed program with data + type-
synonym + signature + two function clauses. Not yet: guards, where
clauses, module header, imports, deriving, contexts, GADTs. 162/162 green.
- **2026-04-24** — Phase 1: full patterns. Added `as` patterns
(`name@apat``(:p-as NAME PAT)`), lazy patterns (`~apat`
`(:p-lazy PAT)`), negative literal patterns (`-N` / `-F` resolving
eagerly in the parser so downstream passes see a plain `(:p-int -1)`),
and infix constructor patterns via a right-associative single-band
layer on top of `hk-parse-pat-lhs` for any `consym` or reservedop `:`
(so `x : xs` parses as `(:p-con ":" [x, xs])`, `a :+: b` likewise).
Extended `hk-apat-start?` with `-` and `~` so the pattern-argument
loops in lambdas and constructor applications pick these up.
Lambdas now parse apat parameters instead of bare varids — so the
`:lambda` AST is `(:lambda APATS BODY)` with apats as pattern nodes.
`hk-parse-bind` became a plain `pat = expr` form, so `:bind` now has
a pattern LHS throughout (simple `x = 1``(:bind (:p-var "x") …)`);
this picks up `let (x, y) = pair in …` and `let Just x = m in x`
automatically, and flows through `do`-notation lets. Eight existing
tests updated to the pattern-flavoured AST. Also fixed a pragmatic
layout issue that surfaced in multi-line `let`s: when a layout-indent
would emit a spurious `;` just before an `in` token (because the
let block had already been closed by dedent), `hk-peek-next-reserved`
now lets the layout pass skip that indent and leave closing to the
existing `in` handler. 18 new tests in
`lib/haskell/tests/parser-patterns.sx` cover every pattern variant,
lambda with mixed apats, let pattern-bindings (tuple / constructor /
cons), and do-bind with a tuple pattern. 138/138 green.
- **2026-04-24** — Phase 1: `case … of` and `do`-notation parsers. Added `hk-parse-case`
/ `hk-parse-alt`, `hk-parse-do` / `hk-parse-do-stmt` / `hk-parse-do-let`, plus the
minimal pattern language needed to make arms and binds meaningful:
`hk-parse-apat` (var, wildcard `_`, int/float/string/char literal, 0-arity
conid/qconid, paren+tuple, list) and `hk-parse-pat` (conid applied to
apats greedily). AST nodes: `:case SCRUT ALTS`, `:alt PAT BODY`, `:do STMTS`
with stmts `:do-expr E` / `:do-bind PAT E` / `:do-let BINDS`, and pattern
tags `:p-wild` / `:p-int` / `:p-float` / `:p-string` / `:p-char` / `:p-var`
/ `:p-con NAME ARGS` / `:p-tuple` / `:p-list`. `do`-stmts disambiguate
`pat <- e` vs bare expression with a forward paren/bracket/brace-balanced
scan for `<-` before the next `;`/`}` — no backtracking, no AST rewrite.
`case` and `do` accept both implicit (`vlbrace`/`vsemi`/`vrbrace`) and
explicit braces. Added to `hk-parse-lexp` so they participate fully in
operator-precedence expressions. 19 new tests in
`lib/haskell/tests/parser-case-do.sx` cover every pattern variant,
explicit-brace `case`, expression scrutinees, do with bind/let/expr,
multi-binding `let` in `do`, constructor patterns in binds, and
`case`/`do` nested inside `let` and lambda. The full pattern item (as
patterns, negative literals, `~` lazy, lambda/let pattern extension)
remains a separate sub-item. 119/119 green.
- **2026-04-24** — Phase 1: expression parser (`lib/haskell/parser.sx`, ~380 lines).
Pratt-style precedence climbing against a Haskell-98-default op table (24
operators across precedence 09, left/right/non assoc, default infixl 9 for
anything unlisted). Supports literals (int/float/string/char), varid/conid
(qualified variants folded into `:var` / `:con`), parens / unit / tuples,
list literals, ranges `[a..b]` and `[a,b..c]`, left-associative application,
unary `-`, backtick operators (`x \`mod\` 3`), lambdas, `if-then-else`, and
`let … in` consuming both virtual and explicit braces. AST uses keyword
tags (`:var`, `:op`, `:lambda`, `:let`, `:bind`, `:tuple`, `:range`,
`:range-step`, `:app`, `:neg`, `:if`, `:list`, `:int`, `:float`, `:string`,
`:char`, `:con`). The parser skips a leading `vlbrace` / `lbrace` so it can
be called on full post-layout output, and uses a `raise`-based error channel
with location-lite messages. 42 new tests in `lib/haskell/tests/parser-expr.sx`
cover literals, identifiers, parens/tuple/unit, list + range, app associativity,
operator precedence (mul over add, cons right-assoc, function-composition
right-assoc, `$` lowest), backtick ops, unary `-`, lambda multi-param,
`if` with infix condition, single- and multi-binding `let` (both implicit
and explicit braces), plus a few mixed nestings. 100/100 green.
- **2026-04-24** — Phase 1: layout algorithm (`lib/haskell/layout.sx`, ~260 lines)
implementing Haskell 98 §10.3. Two-pass design: a pre-pass augments the raw
token stream with explicit `layout-open` / `layout-indent` markers (suppressing
`<n>` when `{n}` already applies, per note 3), then an L pass consumes the
augmented stream against a stack of implicit/explicit layout contexts and
emits `vlbrace` / `vsemi` / `vrbrace` tokens; newlines are dropped. Supports
the initial module-level implicit open (skipped when the first token is
`module` or `{`), the four layout keywords (`let`/`where`/`do`/`of`), explicit
braces disabling layout, dedent closing nested implicit blocks while also
emitting `vsemi` at the enclosing level, and the pragmatic single-line
`let … in` rule (emit `}` when `in` meets an implicit let). 15 new tests
in `lib/haskell/tests/layout.sx` cover module-start, do/let/where/case/of,
explicit braces, multi-level dedent, line continuation, and EOF close-down.
Shared test helpers moved to `lib/haskell/testlib.sx` so both test files
can share one `hk-test`. `test.sh` preloads tokenizer + layout + testlib.
58/58 green.
- **2026-04-24** — Phase 1: Haskell 98 tokenizer (`lib/haskell/tokenizer.sx`, 490 lines)
covering idents (lower/upper/qvarid/qconid), 23 reserved words, 11 reserved ops,
varsym/consym operator chains, integer/hex/octal/float literals incl. exponent

View File

@@ -1,120 +0,0 @@
# Smalltalk-on-SX: blocks with non-local return on delimited continuations
The headline showcase is **blocks** — Smalltalk's closures with non-local return (`^expr` aborts the enclosing *method*, not the block). Every other Smalltalk on top of a host VM (RSqueak on PyPy, GemStone on C, Maxine on Java) reinvents non-local return on whatever stack discipline the host gives them. On SX it's a one-liner: a block holds a captured continuation; `^` just invokes it. Message-passing OO falls out cheaply on top of the existing component / dispatch machinery.
End-state goal: ANSI-ish Smalltalk-80 subset, SUnit working, ~200 hand-written tests + a vendored slice of the Pharo kernel tests, classic corpus (eight queens, quicksort, mandelbrot, Conway's Life).
## Scope decisions (defaults — override by editing before we spawn)
- **Syntax:** Pharo / Squeak chunk format (`!` separators, `Object subclass: #Foo …`). No fileIn/fileOut images — text source only.
- **Conformance:** ANSI X3J20 *as a target*, not bug-for-bug Squeak. "Reads like Smalltalk, runs like Smalltalk."
- **Test corpus:** SUnit ported to SX-Smalltalk + custom programs + a curated slice of Pharo `Kernel-Tests` / `Collections-Tests`.
- **Image:** out of scope. Source-only. No `become:` between sessions, no snapshotting.
- **Reflection:** `class`, `respondsTo:`, `perform:`, `doesNotUnderstand:` in. `become:` (object-identity swap) **in** — it's a good CEK exercise. Method modification at runtime in.
- **GUI / Morphic / threads:** out entirely.
## Ground rules
- **Scope:** only touch `lib/smalltalk/**` and `plans/smalltalk-on-sx.md`. Don't edit `spec/`, `hosts/`, `shared/`, or any other `lib/<lang>/**`. Smalltalk primitives go in `lib/smalltalk/runtime.sx`.
- **SX files:** use `sx-tree` MCP tools only.
- **Commits:** one feature per commit. Keep `## Progress log` updated and tick roadmap boxes.
## Architecture sketch
```
Smalltalk source
lib/smalltalk/tokenizer.sx — selectors, keywords, literals, $c, #sym, #(…), $'…'
lib/smalltalk/parser.sx — AST: classes, methods, blocks, cascades, sends
lib/smalltalk/transpile.sx — AST → SX AST (entry: smalltalk-eval-ast)
lib/smalltalk/runtime.sx — class table, MOP, dispatch, primitives
```
Core mapping:
- **Class** = SX dict `{:name :superclass :ivars :methods :class-methods :metaclass}`. Class table is a flat dict keyed by class name.
- **Object** = SX dict `{:class :ivars}``ivars` keyed by symbol. Tagged ints / floats / strings / symbols are not boxed; their class is looked up by SX type.
- **Method** = SX lambda closing over a `self` binding + temps. Body wrapped in a delimited continuation so `^` can escape.
- **Message send** = `(st-send receiver selector args)` — does class-table lookup, walks superclass chain, falls back to `doesNotUnderstand:` with a `Message` object.
- **Block** `[:x | … ^v … ]` = lambda + captured `^k` (the method-return continuation). Invoking `^` calls `k`; outer block invocation past method return raises `BlockContext>>cannotReturn:`.
- **Cascade** `r m1; m2; m3` = `(let ((tmp r)) (st-send tmp 'm1 ()) (st-send tmp 'm2 ()) (st-send tmp 'm3 ()))`.
- **`ifTrue:ifFalse:` / `whileTrue:`** = ordinary block sends; the runtime intrinsifies them in the JIT path so they compile to native branches (Tier 1 of bytecode expansion already covers this pattern).
- **`become:`** = swap two object identities everywhere — in SX this is a heap walk, but we restrict to `oneWayBecome:` (cheap: rewrite class field) by default.
## Roadmap
### Phase 1 — tokenizer + parser
- [x] Tokenizer: identifiers, keywords (`foo:`), binary selectors (`+`, `==`, `,`, `->`, `~=` etc.), numbers (radix `16r1F`; **scaled `1.5s2` deferred**), strings `'…''…'`, characters `$c`, symbols `#foo` `#'foo bar'` `#+`, byte arrays `#[1 2 3]` (open token), literal arrays `#(1 #foo 'x')` (open token), comments `"…"`
- [x] Parser (expression level): blocks `[:a :b | | t1 t2 | …]`, cascades, message precedence (unary > binary > keyword), assignment, return, statement sequences, literal arrays, byte arrays, paren grouping, method headers (`+ other`, `at:put:`, unary, with temps and body). Class-definition keyword messages parse as ordinary keyword sends — no special-case needed.
- [x] Parser (chunk-stream level): `st-read-chunks` splits source on `!` (with `!!` doubling) and `st-parse-chunks` runs the Pharo file-in state machine — `methodsFor:` / `class methodsFor:` opens a method batch, an empty chunk closes it. Pragmas `<primitive: …>` (incl. multiple keyword pairs, before or after temps, multiple per method) parsed into the method AST.
- [x] Unit tests in `lib/smalltalk/tests/parse.sx`
### Phase 2 — object model + sequential eval
- [x] Class table + bootstrap (`lib/smalltalk/runtime.sx`): canonical hierarchy installed (`Object`, `Behavior`, `ClassDescription`, `Class`, `Metaclass`, `UndefinedObject`, `Boolean`/`True`/`False`, `Magnitude`/`Number`/`Integer`/`SmallInteger`/`Float`/`Character`, `Collection`/`SequenceableCollection`/`ArrayedCollection`/`Array`/`String`/`Symbol`/`OrderedCollection`/`Dictionary`, `BlockClosure`). User class definition via `st-class-define!`, methods via `st-class-add-method!` (stamps `:defining-class` for super), method lookup walks chain, ivars accumulated through superclass chain, native SX value types map to Smalltalk classes via `st-class-of`.
- [ ] `smalltalk-eval-ast`: literals, variable reference, assignment, message send, cascade, sequence, return
- [ ] Method lookup: walk class → superclass; cache hit-class on `(class, selector)`
- [ ] `doesNotUnderstand:` fallback constructing `Message` object
- [ ] `super` send (lookup starts at superclass of *defining* class, not receiver class)
- [ ] 30+ tests in `lib/smalltalk/tests/eval.sx`
### Phase 3 — blocks + non-local return (THE SHOWCASE)
- [ ] Method invocation captures a `^k` (the return continuation) and binds it as the block's escape
- [ ] `^expr` from inside a block invokes that captured `^k`
- [ ] `BlockContext>>value`, `value:`, `value:value:`, …, `valueWithArguments:`
- [ ] `whileTrue:` / `whileTrue` / `whileFalse:` / `whileFalse` as ordinary block sends — runtime intrinsifies the loop in the bytecode JIT
- [ ] `ifTrue:` / `ifFalse:` / `ifTrue:ifFalse:` as block sends, similarly intrinsified
- [ ] Escape past returned-from method raises `BlockContext>>cannotReturn:`
- [ ] Classic programs in `lib/smalltalk/tests/programs/`:
- [ ] `eight-queens.st`
- [ ] `quicksort.st`
- [ ] `mandelbrot.st`
- [ ] `life.st` (Conway's Life, glider gun)
- [ ] `fibonacci.st` (recursive + memoised)
- [ ] `lib/smalltalk/conformance.sh` + runner, `scoreboard.json` + `scoreboard.md`
### Phase 4 — reflection + MOP
- [ ] `Object>>class`, `class>>name`, `class>>superclass`, `class>>methodDict`, `class>>selectors`
- [ ] `Object>>perform:` / `perform:with:` / `perform:withArguments:`
- [ ] `Object>>respondsTo:`, `Object>>isKindOf:`, `Object>>isMemberOf:`
- [ ] `Behavior>>compile:` — runtime method addition
- [ ] `Object>>becomeForward:` (one-way become; rewrites the class field of `aReceiver`)
- [ ] Exceptions: `Exception`, `Error`, `signal`, `signal:`, `on:do:`, `ensure:`, `ifCurtailed:` — built on top of SX `handler-bind`/`raise`
### Phase 5 — collections + numeric tower
- [ ] `SequenceableCollection`/`OrderedCollection`/`Array`/`String`/`Symbol`
- [ ] `HashedCollection`/`Set`/`Dictionary`/`IdentityDictionary`
- [ ] `Stream` hierarchy: `ReadStream`/`WriteStream`/`ReadWriteStream`
- [ ] `Number` tower: `SmallInteger`/`LargePositiveInteger`/`Float`/`Fraction`
- [ ] `String>>format:`, `printOn:` for everything
### Phase 6 — SUnit + corpus to 200+
- [ ] Port SUnit (TestCase, TestSuite, TestResult) — written in SX-Smalltalk, runs in itself
- [ ] Vendor a slice of Pharo `Kernel-Tests` and `Collections-Tests`
- [ ] Drive the scoreboard up: aim for 200+ green tests
- [ ] Stretch: ANSI Smalltalk validator subset
### Phase 7 — speed (optional)
- [ ] Method-dictionary inline caching (already in CEK as a primitive; just wire selector cache)
- [ ] Block intrinsification beyond `whileTrue:` / `ifTrue:`
- [ ] Compare against GNU Smalltalk on the corpus
## Progress log
_Newest first. Agent appends on every commit._
- 2026-04-25: class table + bootstrap (`lib/smalltalk/runtime.sx`, `lib/smalltalk/tests/runtime.sx`). Canonical hierarchy, type→class mapping for native SX values, instance construction, ivar inheritance, method install with `:defining-class` stamp, instance- and class-side method lookup walking the superclass chain. 54 new tests, 185/185 total.
- 2026-04-25: chunk-stream parser + pragmas + 21 chunk/pragma tests (`lib/smalltalk/tests/parse_chunks.sx`). `st-read-chunks` (with `!!` doubling), `st-parse-chunks` state machine for `methodsFor:` batches incl. class-side. Pragmas with multiple keyword pairs, signed numeric / string / symbol args, in either pragma-then-temps or temps-then-pragma order. 131/131 tests pass.
- 2026-04-25: expression-level parser + 47 parse tests (`lib/smalltalk/parser.sx`, `lib/smalltalk/tests/parse.sx`). Full message precedence (unary > binary > keyword), cascades, blocks with params/temps, literal/byte arrays, assignment chain, method headers (unary/binary/keyword). Chunk-format `! !` driver deferred to a follow-up box. 110/110 tests pass.
- 2026-04-25: tokenizer + 63 tests (`lib/smalltalk/tokenizer.sx`, `lib/smalltalk/tests/tokenize.sx`, `lib/smalltalk/test.sh`). All token types covered except scaled decimals `1.5s2` (deferred). `#(` and `#[` emit open tokens; literal-array contents lexed as ordinary tokens for the parser to interpret.
## Blockers
_Shared-file issues that need someone else to fix. Minimal repro only._
- _(none yet)_

View File

@@ -30,7 +30,7 @@ fi
if [ "$CLEAN" = "1" ]; then
cd "$(dirname "$0")/.."
for lang in lua prolog forth erlang haskell js hs smalltalk; do
for lang in lua prolog forth erlang haskell js hs; do
wt="$WORKTREE_BASE/$lang"
if [ -d "$wt" ]; then
git worktree remove --force "$wt" 2>/dev/null || rm -rf "$wt"
@@ -39,5 +39,5 @@ if [ "$CLEAN" = "1" ]; then
done
git worktree prune
echo "Worktree branches (loops/<lang>) are preserved. Delete manually if desired:"
echo " git branch -D loops/lua loops/prolog loops/forth loops/erlang loops/haskell loops/js loops/hs loops/smalltalk"
echo " git branch -D loops/lua loops/prolog loops/forth loops/erlang loops/haskell loops/js loops/hs"
fi

View File

@@ -1,5 +1,5 @@
#!/usr/bin/env bash
# Spawn 8 claude sessions in tmux, one per language loop.
# Spawn 7 claude sessions in tmux, one per language loop.
# Each runs in its own git worktree rooted at /root/rose-ash-loops/<lang>,
# on branch loops/<lang>. No two loops share a working tree, so there's
# zero risk of file collisions between languages.
@@ -9,7 +9,7 @@
#
# After the script prints done:
# tmux a -t sx-loops
# Ctrl-B + <window-number> to switch (0=lua ... 7=smalltalk)
# Ctrl-B + <window-number> to switch (0=lua ... 6=hs)
# Ctrl-B + d to detach (loops keep running, SSH-safe)
#
# Stop: ./scripts/sx-loops-down.sh
@@ -38,9 +38,8 @@ declare -A BRIEFING=(
[haskell]=haskell-loop.md
[js]=loop.md
[hs]=hs-loop.md
[smalltalk]=smalltalk-loop.md
)
ORDER=(lua prolog forth erlang haskell js hs smalltalk)
ORDER=(lua prolog forth erlang haskell js hs)
mkdir -p "$WORKTREE_BASE"
@@ -67,7 +66,7 @@ for lang in "${ORDER[@]:1}"; do
tmux new-window -t "$SESSION" -n "$lang" -c "$WORKTREE_BASE/$lang"
done
echo "Starting 8 claude sessions..."
echo "Starting 7 claude sessions..."
for lang in "${ORDER[@]}"; do
tmux send-keys -t "$SESSION:$lang" "claude" C-m
done
@@ -90,10 +89,10 @@ for lang in "${ORDER[@]}"; do
done
echo ""
echo "Done. 8 loops started in tmux session '$SESSION', each in its own worktree."
echo "Done. 7 loops started in tmux session '$SESSION', each in its own worktree."
echo ""
echo " Attach: tmux a -t $SESSION"
echo " Switch: Ctrl-B <0..7> (0=lua 1=prolog 2=forth 3=erlang 4=haskell 5=js 6=hs 7=smalltalk)"
echo " Switch: Ctrl-B <0..6> (0=lua 1=prolog 2=forth 3=erlang 4=haskell 5=js 6=hs)"
echo " List: Ctrl-B w"
echo " Detach: Ctrl-B d"
echo " Stop: ./scripts/sx-loops-down.sh"