Chez Phase 3 inc 5a: port reader atom layer to jolt.reader (Clojure)
Starts taking the reader off Janet (src/jolt/reader.janet, 831 lines) into
portable jolt-core Clojure. jolt.reader holds the lexing/parsing LOGIC; form
construction + string->number parsing delegate to the jolt.host contract — a
Clojure source file can't write a {:jolt/type :symbol} literal (parses as a
tagged form) and the concrete representation is the host's to own. Same split the
analyzer/emitter already use. Once cross-compiled this runs on Chez so compile-
from-source needs no Janet reader.
inc 5a = the atom layer: whitespace/comments, symbols (+ nil/true/false),
keywords, strings (escapes), numbers (sign/hex/radix/ratio/fractional/exponent,
trailing N/M), characters. Collections, quote/deref/meta and dispatch (#) follow
in 5b/5c (throw not-yet-ported). Positions are char indices (Janet uses bytes);
identical for ASCII and the gate compares form VALUES, not positions.
host_iface.janet gains four reader primitives on the contract: form-make-symbol,
form-make-char, form-char-from-name, form-scan-number (the irreducible host bits
the portable reader rests on). Additive — new jolt.host interns, nothing else
changed.
Surfaced jolt-if19 (Janet seed reader: +N literals error instead of reading as N;
read-number strips only the - sign). The port reproduces it; both-throw counts as
faithful parity in the gate.
Gate: reader-parity 64/64 (symbols/keywords/strings/ints/hex/radix/ratio/floats/
exponent/N-M/chars). Full jpm gate green after clean rebuild, conformance 355x3.
jolt-50xx.
This commit is contained in:
Yogthos
parent
2a33da87d8
commit
4de586089c
3 changed files with 277 additions and 0 deletions
195
jolt-core/jolt/reader.clj
Normal file
195
jolt-core/jolt/reader.clj
Normal file
|
|
@ -0,0 +1,195 @@
|
|||
(ns jolt.reader
|
||||
"Portable Clojure reader: source text -> reader forms (Chez Phase 3, jolt-cf1q.4).
|
||||
|
||||
The no-Janet replacement for src/jolt/reader.janet. All the lexing/parsing LOGIC
|
||||
is portable Clojure; form CONSTRUCTION and string->number parsing delegate to the
|
||||
jolt.host contract (form-make-symbol/char, form-char-from-name, form-scan-number)
|
||||
— a Clojure source file cannot write a {:jolt/type :symbol} literal (it parses as
|
||||
a tagged reader form), and the concrete form representation is the host's to own.
|
||||
Same split the analyzer uses for the form-* readers. Once cross-compiled this runs
|
||||
ON Chez so compile-from-source needs no Janet reader.
|
||||
|
||||
Positions are CHARACTER indices (the Janet reader uses byte indices); for ASCII
|
||||
source they coincide, and form VALUES are identical either way — the parity gate
|
||||
compares values, not positions.
|
||||
|
||||
INCREMENT 5a (jolt-50xx): the ATOM layer — whitespace/comments, symbols (+ nil/
|
||||
true/false), keywords, strings, numbers (sign/hex/radix/ratio/fractional/
|
||||
exponent, trailing N/M), characters. Collections, quote/deref/meta, and dispatch
|
||||
(#) land in 5b/5c (they throw not-yet-ported so a hit is loud)."
|
||||
(:require [clojure.string :as str]
|
||||
[jolt.host :refer [form-make-symbol form-make-char form-char-from-name
|
||||
form-scan-number]]))
|
||||
|
||||
;; Source access by CHARACTER codepoint, mirroring the Janet reader's byte access
|
||||
;; (identical for ASCII). cp = codepoint at i; len = character count.
|
||||
(defn- cp [s i] (int (nth s i)))
|
||||
(defn- len [s] (count s))
|
||||
|
||||
(defn- whitespace? [c] (or (= c 32) (= c 9) (= c 10) (= c 13) (= c 44))) ; space tab nl cr ,
|
||||
(defn- digit? [c] (and (>= c 48) (<= c 57)))
|
||||
(defn- hex-digit? [c]
|
||||
(or (digit? c) (and (>= c 65) (<= c 70)) (and (>= c 97) (<= c 102))))
|
||||
(defn- symbol-start? [c]
|
||||
(or (and (>= c 65) (<= c 90)) (and (>= c 97) (<= c 122))
|
||||
(= c 42) (= c 43) (= c 33) (= c 95) (= c 45) (= c 63) (= c 46)
|
||||
(= c 60) (= c 62) (= c 61) (= c 38) (= c 124) (= c 36) (= c 37) (= c 47)))
|
||||
(defn- symbol-char? [c]
|
||||
(or (symbol-start? c) (digit? c) (= c 35) (= c 39) (= c 58))) ; + # ' :
|
||||
|
||||
(defn- skip-whitespace [s pos]
|
||||
(if (and (< pos (len s)) (whitespace? (cp s pos)))
|
||||
(recur s (inc pos))
|
||||
pos))
|
||||
|
||||
(defn- read-until-newline [s pos]
|
||||
(if (or (>= pos (len s)) (= (cp s pos) 10)) pos (recur s (inc pos))))
|
||||
|
||||
;; --- symbols -----------------------------------------------------------------
|
||||
(defn- read-symbol-name [s pos end]
|
||||
(if (and (< end (len s)) (symbol-char? (cp s end))) (recur s pos (inc end)) end))
|
||||
|
||||
(defn- read-symbol* [s pos]
|
||||
(let [end (read-symbol-name s pos pos)]
|
||||
(when (= end pos)
|
||||
(throw (ex-info (str "Unrecognized character: " (char (cp s pos))) {})))
|
||||
(let [nm (subs s pos end)]
|
||||
(cond
|
||||
(= nm "nil") [nil end]
|
||||
(= nm "true") [true end]
|
||||
(= nm "false") [false end]
|
||||
:else [(form-make-symbol nm) end]))))
|
||||
|
||||
;; --- keywords ----------------------------------------------------------------
|
||||
(defn- read-keyword-name [s pos end]
|
||||
(if (and (< end (len s)) (symbol-char? (cp s end))) (recur s pos (inc end)) end))
|
||||
|
||||
(defn- read-keyword* [s pos]
|
||||
;; pos is at the first colon; ::foo is treated as :foo (no auto-resolution),
|
||||
;; matching the Janet reader.
|
||||
(let [start (if (and (< (inc pos) (len s)) (= (cp s (inc pos)) 58)) (+ pos 2) (inc pos))
|
||||
end (read-keyword-name s start start)]
|
||||
[(keyword (subs s start end)) end]))
|
||||
|
||||
;; --- strings -----------------------------------------------------------------
|
||||
(defn- escape-char [c]
|
||||
(cond (= c 110) "\n" (= c 116) "\t" (= c 114) "\r" (= c 92) "\\" (= c 34) "\""
|
||||
:else (str (char c))))
|
||||
|
||||
(defn- read-string* [s pos]
|
||||
;; pos at opening double-quote
|
||||
(loop [p (inc pos) acc []]
|
||||
(when (>= p (len s)) (throw (ex-info "Unterminated string" {})))
|
||||
(let [c (cp s p)]
|
||||
(cond
|
||||
(= c 92) (let [np (inc p)]
|
||||
(when (>= np (len s)) (throw (ex-info "Unterminated escape" {})))
|
||||
(recur (+ p 2) (conj acc (escape-char (cp s np)))))
|
||||
(= c 34) [(apply str acc) (inc p)]
|
||||
:else (recur (inc p) (conj acc (str (char c))))))))
|
||||
|
||||
;; --- numbers -----------------------------------------------------------------
|
||||
(defn- read-digits [s pos end]
|
||||
(if (and (< end (len s)) (digit? (cp s end))) (recur s pos (inc end)) end))
|
||||
(defn- read-hex-digits [s pos end]
|
||||
(if (and (< end (len s)) (hex-digit? (cp s end))) (recur s pos (inc end)) end))
|
||||
|
||||
;; Value of an alphanumeric digit for radix parsing (0-9, a-z/A-Z = 10-35).
|
||||
(defn- radix-digit-val [c]
|
||||
(cond
|
||||
(and (>= c 48) (<= c 57)) (- c 48)
|
||||
(and (>= c 97) (<= c 122)) (+ 10 (- c 97))
|
||||
(and (>= c 65) (<= c 90)) (+ 10 (- c 65))
|
||||
:else nil))
|
||||
(defn- read-alnum [s pos end]
|
||||
(if (and (< end (len s)) (radix-digit-val (cp s end))) (recur s pos (inc end)) end))
|
||||
|
||||
(defn- read-exponent [s end]
|
||||
;; if s[end] is e/E (optionally signed) followed by digits, return index past it
|
||||
(if (and (< end (len s)) (let [c (cp s end)] (or (= c 101) (= c 69))))
|
||||
(let [p (if (and (< (inc end) (len s)) (let [c (cp s (inc end))] (or (= c 43) (= c 45))))
|
||||
(+ end 2) (inc end))
|
||||
de (read-digits s p p)]
|
||||
(if (> de p) de end))
|
||||
end))
|
||||
|
||||
;; Jolt has no bignum/ratio: N (bigint) / M (bigdec) suffixes read as the plain
|
||||
;; number, a ratio a/b reads as the double quotient, radixed ints by base.
|
||||
(defn- read-number* [s pos]
|
||||
(let [length (len s)
|
||||
neg (and (< pos length) (= (cp s pos) 45))
|
||||
start (if neg (inc pos) pos)
|
||||
hex? (and (< (inc start) length) (= (cp s start) 48)
|
||||
(let [c1 (cp s (inc start))] (or (= c1 120) (= c1 88))))] ; 0x / 0X
|
||||
(if hex?
|
||||
(let [hs (+ start 2) he (read-hex-digits s hs hs)]
|
||||
(when (= he hs) (throw (ex-info "Expected hex digits" {})))
|
||||
(let [he2 (if (and (< he length) (= (cp s he) 78)) (inc he) he) ; trailing N
|
||||
val (form-scan-number (str "0x" (subs s hs he)))]
|
||||
[(if neg (- val) val) he2]))
|
||||
(let [iend (read-digits s start start)]
|
||||
(when (= iend start) (throw (ex-info "Expected number" {})))
|
||||
(cond
|
||||
;; radix integer <base>r<digits>
|
||||
(and (< iend length) (let [c (cp s iend)] (or (= c 114) (= c 82))))
|
||||
(let [base (form-scan-number (subs s start iend))
|
||||
ds (inc iend) de (read-alnum s ds ds)]
|
||||
(when (= de ds) (throw (ex-info "Expected radix digits" {})))
|
||||
(let [acc (reduce (fn [a i] (+ (* a base) (radix-digit-val (cp s i)))) 0 (range ds de))]
|
||||
[(if neg (- acc) acc) de]))
|
||||
;; ratio <int>/<int> (only when a digit follows the slash)
|
||||
(and (< (inc iend) length) (= (cp s iend) 47) (digit? (cp s (inc iend))))
|
||||
(let [ds (inc iend) de (read-digits s ds ds)
|
||||
numr (form-scan-number (subs s start iend))
|
||||
den (form-scan-number (subs s ds de))]
|
||||
[(if neg (- (/ numr den)) (/ numr den)) de])
|
||||
;; fractional and/or exponent, optional trailing N/M
|
||||
:else
|
||||
(let [frac-end (if (and (< iend length) (= (cp s iend) 46))
|
||||
(let [fs (inc iend) fe (read-digits s fs fs)]
|
||||
(when (= fe fs) (throw (ex-info "Expected digit after ." {})))
|
||||
fe)
|
||||
iend)
|
||||
exp-end (read-exponent s frac-end)
|
||||
val (form-scan-number (subs s start exp-end))
|
||||
fin (if (and (< exp-end length) (let [c (cp s exp-end)] (or (= c 78) (= c 77))))
|
||||
(inc exp-end) exp-end)]
|
||||
[(if neg (- val) val) fin]))))))
|
||||
|
||||
;; --- characters --------------------------------------------------------------
|
||||
(defn- read-char-name-end [s pos]
|
||||
(if (and (< pos (len s)) (symbol-char? (cp s pos))) (recur s (inc pos)) pos))
|
||||
|
||||
(defn- read-char* [s pos]
|
||||
(when (>= (inc pos) (len s)) (throw (ex-info "unexpected end of input after \\" {})))
|
||||
(let [end (read-char-name-end s (inc pos))]
|
||||
(if (= end (inc pos))
|
||||
;; a non-symbol char right after \ is a one-character literal of itself
|
||||
[(form-make-char (cp s (inc pos))) (+ pos 2)]
|
||||
[(form-char-from-name (subs s (inc pos) end)) end])))
|
||||
|
||||
;; --- dispatcher --------------------------------------------------------------
|
||||
(defn- number-start? [s pos c]
|
||||
(or (digit? c)
|
||||
(and (= c 45) (< (inc pos) (len s)) (digit? (cp s (inc pos))))
|
||||
(and (= c 43) (< (inc pos) (len s)) (digit? (cp s (inc pos))))))
|
||||
|
||||
(defn read-form [s pos]
|
||||
(let [pos (skip-whitespace s pos)]
|
||||
(if (>= pos (len s))
|
||||
[nil pos]
|
||||
(let [c (cp s pos)]
|
||||
(cond
|
||||
(= c 59) (read-form s (read-until-newline s pos)) ; ; comment: skip + reread
|
||||
(= c 34) (read-string* s pos)
|
||||
(= c 58) (read-keyword* s pos)
|
||||
(= c 92) (read-char* s pos)
|
||||
(number-start? s pos c) (read-number* s pos)
|
||||
(symbol-start? c) (read-symbol* s pos)
|
||||
:else (throw (ex-info (str "read-form: not yet ported (inc 5a atoms): '"
|
||||
(char c) "' (" c ")") {})))))))
|
||||
|
||||
(defn read-one
|
||||
"Read the first form of `s` (skipping leading trivia). Returns the form."
|
||||
[s]
|
||||
(first (read-form s 0)))
|
||||
|
|
@ -282,8 +282,24 @@
|
|||
(defn h-record-shapes [ctx]
|
||||
(or (get (ctx :env) :record-shapes) @{}))
|
||||
|
||||
# Reader form CONSTRUCTORS (Chez Phase 3, jolt-cf1q.4). The portable Clojure
|
||||
# reader (jolt.reader) holds all the LEXING/PARSING logic but must DELEGATE form
|
||||
# construction to the host — a Clojure source file cannot write a {:jolt/type
|
||||
# :symbol} literal (it parses as a tagged reader form, CLAUDE.md), and the
|
||||
# concrete representation (Janet struct/array/tuple here; something else on Chez)
|
||||
# is the host's to own. Same split the analyzer already uses for form-* readers.
|
||||
(defn h-make-symbol [name] (rdr/make-symbol name))
|
||||
(defn h-make-char [code] (make-char code))
|
||||
(defn h-char-from-name [name] (char-from-name name))
|
||||
# Parse a numeric literal substring to a number. The reader keeps the radix/ratio/
|
||||
# sign/exponent LOGIC portable; only the final string->number bottoms out here
|
||||
# (Janet scan-number; Chez string->number). Returns nil on a non-number.
|
||||
(defn h-scan-number [str] (scan-number str))
|
||||
|
||||
(def- exports
|
||||
{"form-sym?" h-sym? "form-sym-name" h-sym-name "form-sym-ns" h-sym-ns
|
||||
"form-make-symbol" h-make-symbol "form-make-char" h-make-char
|
||||
"form-char-from-name" h-char-from-name "form-scan-number" h-scan-number
|
||||
"ref-put!" h-ref-put!
|
||||
"ref-get" h-ref-get
|
||||
"tagged-table" h-tagged-table
|
||||
|
|
|
|||
66
test/chez/reader-parity.janet
Normal file
66
test/chez/reader-parity.janet
Normal file
|
|
@ -0,0 +1,66 @@
|
|||
# Chez Phase 3 inc 5a (jolt-50xx) — value-parity gate for the PORTABLE Clojure
|
||||
# reader (jolt.reader) vs the Janet seed reader (src/jolt/reader.janet, oracle).
|
||||
#
|
||||
# jolt.reader holds the lexing/parsing LOGIC in portable Clojure and delegates
|
||||
# form construction + number parsing to the jolt.host contract. Here it runs
|
||||
# interpreted ON THE JANET HOST (loaded via bootstrap-load-source); each input is
|
||||
# read by BOTH readers and the resulting FORMS compared with jolt's own = (so the
|
||||
# representation, host-built either way, matches structurally). Positions differ
|
||||
# (char vs byte indices) and are not compared.
|
||||
#
|
||||
# janet test/chez/reader-parity.janet (from repo root)
|
||||
(import ../../src/jolt/api :as api)
|
||||
(import ../../src/jolt/backend :as backend)
|
||||
(import ../../src/jolt/reader :as r)
|
||||
(import ../../src/jolt/types_ctx :as tctx)
|
||||
(import ../../src/jolt/types_ns :as tns)
|
||||
(import ../../src/jolt/types_var :as tvar)
|
||||
(import ../../src/jolt/core :as core)
|
||||
|
||||
(var total 0) (var fails 0)
|
||||
(defn ok [name pred &opt extra]
|
||||
(++ total)
|
||||
(if pred (printf "ok: %s" name)
|
||||
(do (++ fails) (printf "FAIL: %s %s" name (or extra "")))))
|
||||
|
||||
(def ctx (api/init {:compile? true}))
|
||||
(def src (get (get (ctx :env) :embedded-sources) "jolt.reader"))
|
||||
(assert src "jolt.reader not embedded (check stdlib_embed collect)")
|
||||
(backend/bootstrap-load-source ctx "jolt.reader" src)
|
||||
(def read-one (tvar/var-get (tns/ns-find (tctx/ctx-find-ns ctx "jolt.reader") "read-one")))
|
||||
(assert read-one "jolt.reader/read-one not found")
|
||||
|
||||
# jolt's own value equality (the host =), the same comparator the corpus gate uses.
|
||||
(defn jeq [a b] (core/jolt-equal? a b))
|
||||
|
||||
(defn check [input]
|
||||
(def w (protect (in (r/parse-next input) 0))) # Janet seed reader (oracle)
|
||||
(def g (protect (read-one input))) # portable Clojure reader
|
||||
(cond
|
||||
# both readers throw on the same input = faithful parity (e.g. the +N latent
|
||||
# bug, jolt-if19 — reader.janet dispatches +digit to read-number but never
|
||||
# strips the +, so "+5" errors; the port reproduces it).
|
||||
(and (not (w 0)) (not (g 0))) (ok input true)
|
||||
(not (w 0)) (ok input false (string "janet threw, clj didn't: clj=" (string/format "%p" (g 1))))
|
||||
(not (g 0)) (ok input false (string "clj threw, janet ok: " (string (g 1))))
|
||||
(ok input (jeq (w 1) (g 1))
|
||||
(string "clj=" (string/format "%p" (g 1)) " janet=" (string/format "%p" (w 1))))))
|
||||
|
||||
# --- inc 5a: atoms -------------------------------------------------------------
|
||||
# nil / bool
|
||||
(each i ["nil" "true" "false"] (check i))
|
||||
# symbols (plain, ns'd, punctuation, special chars)
|
||||
(each i ["foo" "foo-bar" "my.ns/bar" "+" "-" "*" "->" "<=" "some?" "a1" "x'" "ns/+"] (check i))
|
||||
# keywords
|
||||
(each i [":a" ":foo-bar" ":my.ns/key" "::auto" ":a1" ":+"] (check i))
|
||||
# strings (escapes)
|
||||
(each i [`"hello"` `"with space"` `"tab\there"` `"nl\nhere"` `"q\"q"` `"back\\slash"` `""`] (check i))
|
||||
# integers / signs / hex / radix
|
||||
(each i ["0" "42" "-7" "+5" "123456" "0xFF" "0x10" "-0xff" "2r1010" "16rFF" "36rZ" "8r17"] (check i))
|
||||
# floats / exponent / ratio / N|M suffix
|
||||
(each i ["3.14" "-2.5" "0.0" "1e10" "1.5e-3" "2E5" "10N" "3.14M" "1/2" "-3/4" "22/7"] (check i))
|
||||
# characters
|
||||
(each i [`\a` `\Z` `\0` `\newline` `\tab` `\space` `\return` `\\` `\(` `\{` `\%` `A` `\o101`] (check i))
|
||||
|
||||
(printf "\n%d/%d ok" (- total fails) total)
|
||||
(when (> fails 0) (os/exit 1))
|
||||
Loading…
Add table
Add a link
Reference in a new issue