🚀 Start Phase 15.3: Nyash compiler MVP implementation

Major milestone:
- Set up apps/selfhost-compiler/ directory structure
- Implement basic Nyash compiler in Nyash (CompilerBox)
- Stage-1: Basic arithmetic parser (int/string/+/-/*/括弧/return)
- JSON v0 output compatible with --ny-parser-pipe
- Runner integration with NYASH_USE_NY_COMPILER=1 flag
- Comprehensive smoke tests for PHI/Bridge/Stage-2

Technical updates:
- Updated CLAUDE.md with Phase 15.3 status and MIR14 details
- Statement separation policy: newline-based with minimal ASI
- Fixed runaway ny-parser-pipe processes (CPU 94.9%)
- Clarified MIR14 as canonical instruction set (not 13/18)
- LoopForm strategy: PHI auto-generation during reverse lowering

Collaborative development:
- ChatGPT5 implementing compiler skeleton
- Codex provided LoopForm PHI generation guidance
- Claude maintaining documentation and coordination

🎉 セルフホスティングの歴史的一歩!自分自身をコンパイルする日が近いにゃ!

Co-Authored-By: ChatGPT <noreply@openai.com>
This commit is contained in:
Selfhosting Dev
2025-09-15 01:21:37 +09:00
parent d01f9b9c93
commit af11c6855b
28 changed files with 1007 additions and 40 deletions

View File

@ -0,0 +1,18 @@
# Nyash Selfhost Compiler (MVP scaffold)
This is the Phase 15.3 work-in-progress Nyash compiler implemented in Ny.
Layout
- `compiler.nyash`: entry (CompilerBox). Reads `tmp/ny_parser_input.ny`, prints JSON v0.
- `parser/`: lexer/parser/ast (scaffolds; to be filled as we extend Stage2)
- `mir/`: builder/optimizer stubs (future; current target is JSON v0 emit)
- `tests/`: Stage1/2 samples (TBD)
Run (behind flag)
- `NYASH_USE_NY_COMPILER=1 target/release/nyash --backend vm <program.nyash>`
- The runner writes the input to `tmp/ny_parser_input.ny` and invokes this program.
- It captures a JSON v0 line from stdout and executes it via the JSON bridge.
Notes
- Early MVP emits a minimal JSON v0 (currently a placeholder: return 0). We will gradually wire lexer/parser/emitter.
- Keep JSON v0 spec in `docs/reference/ir/json_v0.md`.

View File

@ -0,0 +1,309 @@
// Selfhost Compiler MVP (Phase 15.3)
// Reads tmp/ny_parser_input.ny and prints a minimal JSON v0 program.
static box Main {
// ---- IO helper ----
read_all(path) {
local fb = new FileBox()
fb.open(path, "r")
local s = fb.read()
fb.close()
return s
}
// ---- JSON helpers ----
esc_json(s) {
local out = ""
local i = 0
local n = s.length()
loop(i < n) {
local ch = s.substring(i, i+1)
if ch == "\\" { out = out + "\\\\" } else {
if ch == "\"" { out = out + "\\\"" } else { out = out + ch }
}
i = i + 1
}
return out
}
// ---- Lexer helpers ----
is_digit(ch) {
return ch == "0" || ch == "1" || ch == "2" || ch == "3" || ch == "4" || ch == "5" || ch == "6" || ch == "7" || ch == "8" || ch == "9"
}
is_space(ch) { return ch == " " || ch == "\t" || ch == "\n" || ch == "\r" || ch == ";" }
// ---- Parser (Stage-1/mini Stage-2) ----
// Global cursor for second-pass parser (no pack strings)
gpos_set(i) {
me.gpos = i
return 0
}
gpos_get() { return me.gpos }
parse_number2(src, i) {
local n = src.length()
local j = i
loop(j < n && me.is_digit(src.substring(j, j+1))) { j = j + 1 }
local s = src.substring(i, j)
me.gpos_set(j)
return "{\"type\":\"Int\",\"value\":" + s + "}"
}
parse_string2(src, i) {
local n = src.length()
local j = i + 1
local out = ""
local done = 0
loop(j < n && done == 0) {
local ch = src.substring(j, j+1)
if ch == "\"" {
j = j + 1
done = 1
} else {
if ch == "\\" && j + 1 < n {
local nx = src.substring(j+1, j+2)
if nx == "\"" { out = out + "\"" } else { if nx == "\\" { out = out + "\\" } else { out = out + nx } }
j = j + 2
} else {
out = out + ch
j = j + 1
}
}
}
me.gpos_set(j)
return "{\"type\":\"Str\",\"value\":\"" + me.esc_json(out) + "\"}"
}
parse_factor2(src, i) {
local j = me.skip_ws(src, i)
local ch = src.substring(j, j+1)
if ch == "(" {
local inner = me.parse_expr2(src, j + 1)
local k = me.gpos_get()
k = me.skip_ws(src, k)
if src.substring(k, k+1) == ")" { k = k + 1 }
me.gpos_set(k)
return inner
}
if ch == "\"" { return me.parse_string2(src, j) }
return me.parse_number2(src, j)
}
parse_term2(src, i) {
local lhs = me.parse_factor2(src, i)
local j = me.gpos_get()
local cont = 1
loop(cont == 1) {
j = me.skip_ws(src, j)
if j >= src.length() { cont = 0 } else {
local op = src.substring(j, j+1)
if op != "*" && op != "/" { cont = 0 } else {
local rhs = me.parse_factor2(src, j+1)
j = me.gpos_get()
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
}
}
}
me.gpos_set(j)
return lhs
}
parse_expr2(src, i) {
local lhs = me.parse_term2(src, i)
local j = me.gpos_get()
local cont = 1
loop(cont == 1) {
j = me.skip_ws(src, j)
if j >= src.length() { cont = 0 } else {
local op = src.substring(j, j+1)
if op != "+" && op != "-" { cont = 0 } else {
local rhs = me.parse_term2(src, j+1)
j = me.gpos_get()
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
}
}
}
me.gpos_set(j)
return lhs
}
parse_program2(src) {
local i = me.skip_ws(src, 0)
local j = me.skip_return_kw(src, i)
if j == i { j = i } // optional 'return'
local expr = me.parse_expr2(src, j)
return "{\"version\":0,\"kind\":\"Program\",\"body\":[{\"type\":\"Return\",\"expr\":" + expr + "}]}"
}
i2s(n) {
// integer to decimal string (non-negative only for MVP)
if n == 0 { return "0" }
local x = n
if x < 0 { x = 0 } // MVP: clamp negatives to 0 to avoid surprises
local out = ""
loop(x > 0) {
local q = x / 10
local d = x - q * 10
local ch = "0"
if d == 1 { ch = "1" } else {
if d == 2 { ch = "2" } else {
if d == 3 { ch = "3" } else {
if d == 4 { ch = "4" } else {
if d == 5 { ch = "5" } else {
if d == 6 { ch = "6" } else {
if d == 7 { ch = "7" } else {
if d == 8 { ch = "8" } else {
if d == 9 { ch = "9" }
}
}
}
}
}
}
}
}
out = ch + out
x = q
}
return out
}
to_int(s) {
local n = s.length()
if n == 0 { return 0 }
local i = 0
local acc = 0
loop(i < n) {
local d = s.substring(i, i+1)
local dv = 0
if d == "1" { dv = 1 } else { if d == "2" { dv = 2 } else { if d == "3" { dv = 3 } else { if d == "4" { dv = 4 } else { if d == "5" { dv = 5 } else { if d == "6" { dv = 6 } else { if d == "7" { dv = 7 } else { if d == "8" { dv = 8 } else { if d == "9" { dv = 9 } } } } } } } } }
acc = acc * 10 + dv
i = i + 1
}
return acc
}
parse_number(src, i) {
local n = src.length()
local j = i
loop(j < n && me.is_digit(src.substring(j, j+1))) { j = j + 1 }
local s = src.substring(i, j)
local json = "{\"type\":\"Int\",\"value\":" + s + "}"
return json + "@" + me.i2s(j)
}
parse_string(src, i) {
local n = src.length()
local j = i + 1
local out = ""
loop(j < n) {
local ch = src.substring(j, j+1)
if ch == "\"" {
j = j + 1
return "{\"type\":\"Str\",\"value\":\"" + me.esc_json(out) + "\"}@" + j
}
if ch == "\\" && j + 1 < n {
local nx = src.substring(j+1, j+2)
if nx == "\"" { out = out + "\"" } else { if nx == "\\" { out = out + "\\" } else { out = out + nx } }
j = j + 2
} else {
out = out + ch
j = j + 1
}
}
return "{\"type\":\"Str\",\"value\":\"" + me.esc_json(out) + "\"}@" + me.i2s(j)
}
skip_ws(src, i) {
local n = src.length()
loop(i < n && me.is_space(src.substring(i, i+1))) { i = i + 1 }
return i
}
skip_return_kw(src, i) {
// If source at i starts with "return", advance; otherwise return i unchanged
local n = src.length()
local j = i
if j < n && src.substring(j, j+1) == "r" { j = j + 1 } else { return i }
if j < n && src.substring(j, j+1) == "e" { j = j + 1 } else { return i }
if j < n && src.substring(j, j+1) == "t" { j = j + 1 } else { return i }
if j < n && src.substring(j, j+1) == "u" { j = j + 1 } else { return i }
if j < n && src.substring(j, j+1) == "r" { j = j + 1 } else { return i }
if j < n && src.substring(j, j+1) == "n" { j = j + 1 } else { return i }
return j
}
parse_factor(src, i) {
i = me.skip_ws(src, i)
local ch = src.substring(i, i+1)
if ch == "(" {
local p = me.parse_expr(src, i + 1)
local at = p.lastIndexOf("@")
local ej = p.substring(0, at)
local j = me.to_int(p.substring(at+1, p.length()))
j = me.skip_ws(src, j)
if src.substring(j, j+1) == ")" { j = j + 1 }
return ej + "@" + me.i2s(j)
}
if ch == "\"" { return me.parse_string(src, i) }
return me.parse_number(src, i)
}
parse_term(src, i) {
local p = me.parse_factor(src, i)
local at = p.lastIndexOf("@")
local lhs = p.substring(0, at)
local j = me.to_int(p.substring(at+1, p.length()))
local cont = 1
loop(cont == 1) {
j = me.skip_ws(src, j)
if j >= src.length() { cont = 0 } else {
local op = src.substring(j, j+1)
if op != "*" && op != "/" { cont = 0 } else {
local q = me.parse_factor(src, j+1)
local at2 = q.lastIndexOf("@")
local rhs = q.substring(0, at2)
j = me.to_int(q.substring(at2+1, q.length()))
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
}
}
}
return lhs + "@" + me.i2s(j)
}
parse_expr(src, i) {
local p = me.parse_term(src, i)
local at = p.lastIndexOf("@")
local lhs = p.substring(0, at)
local j = me.to_int(p.substring(at+1, p.length()))
local cont = 1
loop(cont == 1) {
j = me.skip_ws(src, j)
if j >= src.length() { cont = 0 } else {
local op = src.substring(j, j+1)
if op != "+" && op != "-" { cont = 0 } else {
local q = me.parse_term(src, j+1)
local at2 = q.lastIndexOf("@")
local rhs = q.substring(0, at2)
j = me.to_int(q.substring(at2+1, q.length()))
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
}
}
}
return lhs + "@" + me.i2s(j)
}
parse_program(src) {
// Legacy packed path (debug) removed; use parser2
return me.parse_program2(src)
}
main(args) {
// Parse the input and emit JSON v0
local src = me.read_all("tmp/ny_parser_input.ny")
if src == null { src = "return 1+2*3" }
local json = me.parse_program(src)
local console = new ConsoleBox()
// console.println(json) -- final output only
console.println(json)
return 0
}
}

View File

@ -0,0 +1,8 @@
// JSON v0 emitter (MVP placeholder)
static box JsonV0Emitter {
// Emit a minimal Program{return 0}
program_return0() {
return "{\"version\":0,\"kind\":\"Program\",\"body\":[{\"type\":\"Return\",\"expr\":{\"type\":\"Int\",\"value\":0}}]}"
}
}

View File

@ -0,0 +1,5 @@
static box MirBuilder {
// placeholder
main(args) { return 0 }
}

View File

@ -0,0 +1,5 @@
static box Optimizer {
// placeholder
main(args) { return 0 }
}

View File

@ -0,0 +1,5 @@
static box AST {
// scaffold for future AST node constructors
main(args) { return 0 }
}

View File

@ -0,0 +1,5 @@
static box Lexer {
// scaffold for future implementation
main(args) { return 0 }
}

View File

@ -0,0 +1,5 @@
static box Parser {
// scaffold for future implementation
main(args) { return 0 }
}

View File

@ -0,0 +1,4 @@
Stage1 tests (scaffold)
Add minimal Ny source samples here. Harness TBD.