giles
680cdf62aa
lib/scheme/test.sh — single-process test runner. Loads parser/eval/
runtime + lib/guest/reflective/env.sx once, then for each test
suite loads its file and calls its (*-tests-run!) function. Parses
the {:passed N :failed N ...} dict output and aggregates.
Usage:
bash lib/scheme/test.sh # summary
bash lib/scheme/test.sh -v # per-suite breakdown
Output: "ok 296/296 scheme-on-sx tests passed (9 suites)"
lib/scheme/scoreboard.md — per-suite passing counts, phase status,
deferred items, reflective-kit consumption ledger.
The scoreboard documents the chisel value of the Scheme port:
three reflective kits unlocked (env.sx — already extracted with
Scheme as third consumer; evaluator.sx + quoting.sx — second-
consumer-ready for extraction whenever a follow-up commit is run).
Loop status: 11 phases done (1, 2, 3, 3.5, 4, 5abc, 6ab, 7, 8, 9,
10, 11). Two deferred (6c hygiene, full call/cc-wind interaction).
296 tests, 1830 LoC of Scheme implementation. Zero substrate fixes
required across the loop.
4.1 KiB
Scheme-on-SX Scoreboard
All tests pass: 296 / 296 across 9 suites.
Per-suite breakdown
| Suite | Passing | Covers |
|---|---|---|
| parse | 62 | R7RS lexer: numbers, strings, chars, vectors, lists, quote/quasi/unquote, line/block/datum comments |
| eval | 23 | Self-evaluating literals, symbol lookup, quote, primitive application |
| syntax | 49 | if/define/set!/begin/lambda/closures + let/let*/cond/when/unless/and/or |
| runtime | 78 | Standard env: variadic arithmetic, type predicates, list/string/char/vector ops, higher-order combinators |
| control | 25 | call/cc (escape), raise/guard/with-exception-handler, dynamic-wind |
| macros | 20 | define-syntax / syntax-rules incl. tail-rest ellipsis |
| reflection | 23 | eval / interaction-environment / null-environment / scheme-report-environment + quasiquote runtime |
| records | 9 | define-record-type with constructor / predicate / accessor / mutator |
| modules | 7 | define-library + import (minimal — no cond-expand / include / rename) |
Phases implemented
- Phase 1 — Parser
- Phase 2 — Evaluator + env.sx third consumer
- Phase 3 — Syntactic operators (if/lambda/define/set!/begin)
- Phase 3.5 — let/let*/cond/when/unless/and/or
- Phase 4 — Standard environment + set! cond-bugfix
- Phase 5a — call/cc
- Phase 5b — exceptions (raise/guard/with-exception-handler/error)
- Phase 5c — dynamic-wind (basic, no call/cc-escape tracking)
- Phase 6a — define-syntax + syntax-rules (no ellipsis)
- Phase 6b — syntax-rules ellipsis (tail-rest, single variable)
- Phase 7 — eval / interaction-environment second consumer for evaluator.sx
- Phase 8 — define-library + import (minimal)
- Phase 9 — define-record-type
- Phase 10 — quasiquote runtime second consumer for quoting.sx
- Phase 11 — test.sh + scoreboard
Deferred
-
Phase 6c — hygiene (scope-set / lifted-symbol Dybvig-style algorithm). Would be the second consumer for the deferred
lib/guest/reflective/hygiene.sxresearch-grade kit. Current macros work for common patterns but can capture caller bindings if a macro introduces same-named identifiers. -
Nested quasiquote depth tracking —
`\`x\`is not properly depth-aware; matches Kernel's deferred state. -
R7RS module rich features: cond-expand, include, include-library-declarations,
(only ...)/(except ...)/(prefix ...)/(rename ...)import sets. -
Dotted-pair
(a b . rest)syntax at the parser level. Lambda rest-args currently use the(lambda args ...)form (bare symbol) instead. -
Full call/cc + dynamic-wind interaction: re-entry/re-exit of dynamic extents via continuations is not tracked. Pure-eval programs work; call/cc- heavy code with dynamic-wind interleaving doesn't.
Reflective-kit consumption (chisel ledger)
This Scheme port unlocks three reflective-kit extractions from the kernel-on-sx loop's original six-candidate list:
| Kit | Status |
|---|---|
| env.sx | Extracted (third consumer; no adapter) |
| class-chain.sx | n/a (no OO in Scheme) |
| evaluator.sx | Unblocked (second consumer ready) |
| quoting.sx | Unblocked (second consumer ready) |
| hygiene.sx | Awaiting Phase 6c (research-grade) |
| combiner.sx | n/a (no fexprs in Scheme) |
| short-circuit.sx | n/a (Scheme and/or are syntactic, not operative) |
The kit-extraction commits themselves are follow-on work — kit code is staged
in the proposed sections of plans/kernel-on-sx.md; Scheme's consumer code
satisfies the two-consumer rule for evaluator.sx and quoting.sx.
Substrate stats
- parser.sx — 281 LoC
- eval.sx — ~970 LoC
- runtime.sx — ~580 LoC
- Tests — ~1500 LoC across 9 files
Total Scheme implementation ≈ 1830 LoC.