Chez Phase 0a+0b: value model + host-neutral contract gate

0a (host/chez/values.ss): Jolt value model on Chez — nil sentinel distinct from
#f/'(), interned keywords, ns+meta symbols, exactness-aware = and consistent
hash. Chez numeric tower gives ratios/bignums free. 33/33 tests.

0b (test/chez/): extract test/spec/*.janet defspec tables into corpus.edn (2655
cases, valid as both EDN and Janet data), and a runner that drives ANY jolt
binary via the CLI boundary with per-case subprocess isolation. Pluggable target
(JOLT_BIN) so the same corpus gates every host. Baseline vs Janet build/jolt:
2641/2655, 14 known CLI divergences allowlisted; gate fails only on NEW ones.

test/chez/README.md

@@ -0,0 +1,38 @@
+# Chez port — Phase 0 test contract harness
+
+The host-neutral correctness gate for the Chez re-host (epic jolt-cf1q). The
+spec corpus is data, so the SAME contract validates every host.
+
+## Files
+- `extract-corpus.janet` — parses `test/spec/*.janet` `(defspec …)` tables as
+  data and writes `corpus.edn` (2655 `[label expected actual]` cases). The file
+  is valid as BOTH EDN (a future Chez-jolt runner) and Janet data (the runner
+  below). Regenerate: `janet test/chez/extract-corpus.janet`.
+- `corpus.edn` — the extracted contract (generated; checked in for convenience).
+- `run-corpus.janet` — drives a TARGET jolt binary, one fresh subprocess per case
+  (fresh ctx = per-case isolation), checking `(= expected actual)` prints `true`
+  at the CLI, or that a `:throws` case exits non-zero. Pluggable target:
+  - `janet test/chez/run-corpus.janet`                    # default build/jolt
+  - `JOLT_BIN=build/jolt-chez janet test/chez/run-corpus.janet`   # Phase 1+
+  - `JOLT_CORPUS_LIMIT=400 …`                              # every-Nth stride, fast
+- `known-divergences.edn` — allowlist of cases that diverge at the CLI boundary.
+  The gate fails only on a NEW divergence; known ones are reported but tolerated.
+- `values-test.ss` / `../../host/chez/values.ss` — Phase 0a value model + tests.
+
+## The reference baseline (2026-06-17, Janet `build/jolt`, compile mode)
+2641/2655 pass; 14 known divergences. They split into:
+- **interpret-vs-compile leniency** — `:throws` cases where interpret mode raises
+  but compile mode returns (`< nil`, `> with nil`, `neg? keyword`, `max`/`min-key`
+  on non-numbers). Several are also non-canonical vs JVM Clojure.
+- **invoke-collection-as-fn** — the `transient / invokable lookup` suite invokes
+  transients/collections as fns (`((transient {:x 7}) :x)`); compile mode (and
+  JVM Clojure) reject it.
+- **`xml-seq walks`** — one structural case.
+
+The compile-only Chez host (JVM-canonical oracle) should MATCH OR FIX these. The
+gate's job is to catch *regressions* the port introduces, not to bless these.
+
+## Why the CLI boundary
+The runner tests through `jolt -e`, exactly how the Chez host will be exercised —
+not the in-process `eval-string` the Janet `defspec` harness uses. The two differ
+on a handful of cases (the allowlist), and the CLI boundary is the portable one.

test/chez/extract-corpus.janet

@@ -0,0 +1,69 @@
+# Phase 0b — extract the spec corpus into a host-neutral contract file.
+#
+# Parses every test/spec/*.janet as DATA (no eval), pulls each
+# (defspec "suite" [label expected actual] ...) triple, and writes a corpus that
+# is valid BOTH as EDN (a future Chez-jolt runner reads it) and as Janet data
+# (the current runner reads it via `parse`). Run from repo root:
+#   janet test/chez/extract-corpus.janet
+(use ../../src/jolt/reader)   # not needed for parse, but keeps paths obvious
+
+(defn parse-all [src]
+  (def p (parser/new))
+  (parser/consume p src)
+  (parser/eof p)
+  (def out @[])
+  (while (parser/has-more p) (array/push out (parser/produce p)))
+  out)
+
+(defn edn-str [s]
+  # escape a Clojure-source string into an EDN/Janet string literal
+  (def b @`"`)
+  (each c s
+    (cond
+      (= c (chr `"`)) (buffer/push b `\"`)
+      (= c (chr "\\")) (buffer/push b "\\\\")
+      (= c (chr "\n")) (buffer/push b "\\n")
+      (= c (chr "\t")) (buffer/push b "\\t")
+      (buffer/push-byte b c)))
+  (buffer/push b `"`)
+  (string b))
+
+(defn triples-from [form]
+  # form is (defspec "suite" [l e a] ...) as a parsed tuple
+  (when (and (indexed? form) (> (length form) 0) (= (first form) 'defspec))
+    (def suite (in form 1))
+    (def rows @[])
+    (each case (slice form 2)
+      (when (and (indexed? case) (>= (length case) 3))
+        (def label (in case 0))
+        (def expected (in case 1))
+        (def actual (in case 2))
+        # expected is either a Clojure-source string or the :throws keyword;
+        # actual is always a Clojure-source string. Skip non-literal rows.
+        (when (and (string? label) (string? actual)
+                   (or (string? expected) (keyword? expected)))
+          (array/push rows {:suite suite :label label
+                            :expected expected :actual actual})))
+      )
+    rows))
+
+(def spec-dir "test/spec")
+(def all @[])
+(each f (sort (os/dir spec-dir))
+  (when (string/has-suffix? "-spec.janet" f)
+    (def path (string spec-dir "/" f))
+    (each form (parse-all (slurp path))
+      (when-let [rows (triples-from form)]
+        (array/concat all rows)))))
+
+# emit EDN-and-Janet-valid corpus
+(def out @"[\n")
+(each row all
+  (buffer/push out
+    (string "  {:suite " (edn-str (row :suite))
+            " :label " (edn-str (row :label))
+            " :expected " (if (keyword? (row :expected)) ":throws" (edn-str (row :expected)))
+            " :actual " (edn-str (row :actual)) "}\n")))
+(buffer/push out "]\n")
+(spit "test/chez/corpus.edn" out)
+(printf "extracted %d cases from %s into test/chez/corpus.edn" (length all) spec-dir)

test/chez/known-divergences.edn

@@ -0,0 +1,14 @@
+["max non-number throws"
+ "min-key keys nonnum"
+ "neg? keyword"
+ "< nil"
+ "> with nil"
+ "max non-number"
+ "neg? throws"
+ "vector index"
+ "map key as fn"
+ "map key default"
+ "set membership"
+ "set miss default"
+ "collection key"
+ "xml-seq walks"]

test/chez/run-corpus.janet

@@ -0,0 +1,53 @@
+# Phase 0b — host-neutral parity runner.
+#
+# Feeds each corpus case to a TARGET jolt binary as a fresh subprocess (fresh
+# ctx = the per-case isolation the in-process harness gives), checking
+# (= expected actual) prints `true`, or that a :throws case exits non-zero. The
+# target is pluggable so the SAME corpus gates every host:
+#   JOLT_BIN=build/jolt           janet test/chez/run-corpus.janet   # Janet ref
+#   JOLT_BIN=build/jolt-chez ...  (Phase 1+)                         # Chez host
+# Env: JOLT_CORPUS_LIMIT=N caps the run (every-Nth stride) for fast iteration.
+(def corpus (parse (slurp "test/chez/corpus.edn")))
+(def jolt-bin (or (os/getenv "JOLT_BIN") "build/jolt"))
+(def known (let [k @{}] (each l (parse (slurp "test/chez/known-divergences.edn")) (put k l true)) k))
+
+(def cases
+  (if-let [n (os/getenv "JOLT_CORPUS_LIMIT")]
+    (let [stride (max 1 (math/floor (/ (length corpus) (scan-number n))))]
+      (seq [i :range [0 (length corpus) stride]] (in corpus i)))
+    corpus))
+
+(defn run-capture [args]
+  (def proc (os/spawn [jolt-bin ;args] :p {:out :pipe :err :pipe}))
+  (def out (ev/read (proc :out) 0x100000))
+  (ev/read (proc :err) 0x100000)
+  (def code (os/proc-wait proc))
+  # take the LAST non-empty line: a case may print side effects before its value
+  (def lines (filter (fn [l] (not (empty? l))) (string/split "\n" (string/trim (if out (string out) "")))))
+  [code (if (empty? lines) "" (last lines))])
+
+(var pass 0)
+(def fails @[])          # NEW divergences — these fail the gate
+(def known-hit @[])      # expected (allowlisted) divergences
+(defn record-fail [l m]
+  (if (known l) (array/push known-hit l) (array/push fails [l m])))
+(each row cases
+  (def {:expected e :actual a :label l} row)
+  (if (= e :throws)
+    (let [[code _] (run-capture ["-e" a])]
+      (if (not= code 0) (++ pass)
+        (record-fail l "expected an error, exited 0")))
+    (let [[code out] (run-capture ["-e" (string "(= " e " " a ")")])]
+      (cond
+        (not= code 0) (record-fail l (string "errored (exit " code ")"))
+        (= out "true") (++ pass)
+        (record-fail l (string "want true, got " out))))))
+
+(printf "\ncorpus parity [%s]: %d/%d passed  (%d known divergences)"
+        jolt-bin pass (length cases) (length known-hit))
+(when (> (length fails) 0)
+  (printf "%d NEW divergence(s) — gate FAILED:" (length fails))
+  (each [l m] (slice fails 0 (min 20 (length fails)))
+    (printf "  FAIL [%s] %s" l m)))
+(flush)
+(os/exit (if (> (length fails) 0) 1 0))

test/chez/values-test.ss

@@ -0,0 +1,60 @@
+;; Phase 0a tests — Jolt value model on Chez. Run from repo root:
+;;   chez --script test/chez/values-test.ss
+(import (chezscheme))
+(load "host/chez/values.ss")
+
+(define total 0)
+(define fails 0)
+(define (ok name pred)
+  (set! total (+ total 1))
+  (unless pred (set! fails (+ fails 1)) (printf "FAIL: ~a\n" name)))
+
+;; nil distinct from #f and '()
+(ok "nil not #f"        (not (eq? jolt-nil #f)))
+(ok "nil not '()"       (not (eq? jolt-nil '())))
+(ok "nil? jolt-nil"     (jolt-nil? jolt-nil))
+(ok "nil? not on #f"    (not (jolt-nil? #f)))
+
+;; truthiness: only nil and false falsey
+(ok "nil falsey"        (not (jolt-truthy? jolt-nil)))
+(ok "false falsey"      (not (jolt-truthy? #f)))
+(ok "true truthy"       (jolt-truthy? #t))
+(ok "0 truthy"          (jolt-truthy? 0))
+(ok "empty-str truthy"  (jolt-truthy? ""))
+(ok "empty-list truthy" (jolt-truthy? '()))
+
+;; keywords interned -> identity
+(ok "kw eq"             (eq? (keyword #f "foo") (keyword #f "foo")))
+(ok "kw ns eq"          (eq? (keyword "a" "foo") (keyword "a" "foo")))
+(ok "kw diff ns"        (not (eq? (keyword "a" "foo") (keyword #f "foo"))))
+(ok "kw?"               (keyword? (keyword #f "x")))
+(ok "kw not sym"        (not (jolt-symbol? (keyword #f "x"))))
+
+;; symbols NOT interned but jolt= by ns/name
+(ok "sym not eq"        (not (eq? (jolt-symbol #f "x") (jolt-symbol #f "x"))))
+(ok "sym jolt="         (jolt= (jolt-symbol #f "x") (jolt-symbol #f "x")))
+(ok "sym diff name"     (not (jolt= (jolt-symbol #f "x") (jolt-symbol #f "y"))))
+(ok "sym?"              (jolt-symbol? (jolt-symbol "ns" "n")))
+
+;; numbers: exactness-aware = (Clojure semantics)
+(ok "1 = 1"             (jolt= 1 1))
+(ok "1 not= 1.0"        (not (jolt= 1 1.0)))
+(ok "1.0 = 1.0"         (jolt= 1.0 1.0))
+(ok "ratio ="           (jolt= 1/2 1/2))
+(ok "bigint=int exact"  (jolt= 2 (expt 2 1)))
+(ok "= variadic"        (jolt= 3 3 3))
+(ok "= variadic false"  (not (jolt= 3 3 4)))
+
+;; strings / chars
+(ok "str ="             (jolt= "ab" "ab"))
+(ok "str !="            (not (jolt= "ab" "ac")))
+(ok "char ="            (jolt= #\a #\a))
+
+;; hashing consistent with =
+(ok "hash kw stable"    (= (jolt-hash (keyword #f "k")) (jolt-hash (keyword #f "k"))))
+(ok "hash sym stable"   (= (jolt-hash (jolt-symbol #f "k")) (jolt-hash (jolt-symbol #f "k"))))
+(ok "hash 1 != 1.0"     (not (= (jolt-hash 1) (jolt-hash 1.0))))
+(ok "hash str stable"   (= (jolt-hash "abc") (jolt-hash "abc")))
+
+(printf "values-test: ~a/~a passed\n" (- total fails) total)
+(exit (if (> fails 0) 1 0))