restore(lang/compiler): bring back lang/src/compiler from e917d400; add Hako index canaries and docs; implement Rust-side index operator (Array/Map get/set) with Fail‑Fast diagnostics
- restore: lang/src/compiler/** (parser/emit/builder/pipeline_v2) from e917d400 - docs: docs/development/selfhosting/index-operator-hako.md - smokes(hako): tools/smokes/v2/profiles/quick/core/index_operator_hako.sh (opt-in) - smokes(vm): adjust index_operator_vm.sh for semicolon gate + stable error text - rust/parser: allow IndexExpr and assignment LHS=Index; postfix parse LBRACK chain - rust/builder: lower arr/map index to BoxCall get/set; annotate array/map literals; Fail‑Fast for unsupported types - CURRENT_TASK: mark Rust side done; add Hako tasks checklist Note: files disappeared likely due to branch FF to a lineage without lang/src/compiler; no explicit delete commit found. Added anchor checks and suggested CI guard in follow-up.
This commit is contained in:
nyash-codex
239
lang/src/compiler/parser/parser_box.hako
Normal file
239
lang/src/compiler/parser/parser_box.hako
Normal file
@ -0,0 +1,239 @@
|
||||
// Moved from apps/selfhost-compiler/boxes/parser/parser_box.hako
|
||||
// ParserBox — Stage‑1 JSON v0 generator (coordinator, delegates to specialized boxes)
|
||||
// Responsibility: Coordinate parsing, manage state, delegate to specialized boxes
|
||||
// API: parse_program2(src) -> JSON
|
||||
|
||||
using lang.compiler.parser.scan.parser_string_utils_box
|
||||
using lang.compiler.parser.scan.parser_ident_scan_box
|
||||
using lang.compiler.parser.scan.parser_string_scan_box
|
||||
using lang.compiler.parser.using.using_collector_box
|
||||
using lang.compiler.parser.expr.parser_expr_box
|
||||
using lang.compiler.parser.stmt.parser_stmt_box
|
||||
using lang.compiler.parser.stmt.parser_control_box
|
||||
|
||||
box ParserBox {
|
||||
gpos
|
||||
usings_json
|
||||
stage3
|
||||
|
||||
birth() {
|
||||
me.gpos = 0
|
||||
me.usings_json = "[]"
|
||||
me.stage3 = 0
|
||||
return 0
|
||||
}
|
||||
|
||||
stage3_enable(flag) {
|
||||
if flag == null { flag = 0 }
|
||||
if flag == 0 { me.stage3 = 0 } else { me.stage3 = 1 }
|
||||
return 0
|
||||
}
|
||||
|
||||
stage3_enabled() {
|
||||
if me.stage3 == 1 { return 1 }
|
||||
return 0
|
||||
}
|
||||
|
||||
// === State management ===
|
||||
gpos_set(i) { me.gpos = i return 0 }
|
||||
gpos_get() { return me.gpos }
|
||||
|
||||
// === JSON utilities ===
|
||||
esc_json(s) {
|
||||
local out = ""
|
||||
local i = 0
|
||||
local n = s.size()
|
||||
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
|
||||
}
|
||||
|
||||
// === Delegation to ParserStringUtilsBox ===
|
||||
is_digit(ch) { return ParserStringUtilsBox.is_digit(ch) }
|
||||
|
||||
is_space(ch) { return ParserStringUtilsBox.is_space(ch) }
|
||||
|
||||
is_alpha(ch) { return ParserStringUtilsBox.is_alpha(ch) }
|
||||
|
||||
starts_with(src, i, pat) { return ParserStringUtilsBox.starts_with(src, i, pat) }
|
||||
|
||||
index_of(src, i, pat) { return ParserStringUtilsBox.index_of(src, i, pat) }
|
||||
|
||||
trim(s) { return ParserStringUtilsBox.trim(s) }
|
||||
|
||||
starts_with_kw(src, i, kw) { return ParserStringUtilsBox.starts_with_kw(src, i, kw) }
|
||||
|
||||
i2s(v) { return ParserStringUtilsBox.i2s(v) }
|
||||
|
||||
to_int(s) { return ParserStringUtilsBox.to_int(s) }
|
||||
|
||||
skip_ws(src, i) { return ParserStringUtilsBox.skip_ws(src, i) }
|
||||
|
||||
// === Delegation to scanner boxes ===
|
||||
read_ident2(src, i) { return ParserIdentScanBox.scan_ident(src, i) }
|
||||
|
||||
read_string_lit(src, i) {
|
||||
local pair = ParserStringScanBox.scan(src, i)
|
||||
local at = pair.lastIndexOf("@")
|
||||
local content = pair.substring(0, at)
|
||||
local pos = 0
|
||||
if at >= 0 { pos = me.to_int(pair.substring(at+1, pair.size())) }
|
||||
else { pos = i }
|
||||
me.gpos_set(pos)
|
||||
return content
|
||||
}
|
||||
|
||||
// === using system ===
|
||||
add_using(kind, target, alias) {
|
||||
local cur = me.usings_json
|
||||
if cur == null || cur.size() == 0 { cur = "[]" }
|
||||
|
||||
local name = ""
|
||||
local path = null
|
||||
|
||||
if kind == "path" {
|
||||
path = target
|
||||
if alias != null {
|
||||
name = alias
|
||||
} else {
|
||||
local p = target
|
||||
local idx = -1
|
||||
local t = 0
|
||||
loop(t < p.size()) {
|
||||
if p.substring(t,t+1) == "/" { idx = t }
|
||||
t = t + 1
|
||||
}
|
||||
if idx >= 0 { p = p.substring(idx+1, p.size()) }
|
||||
|
||||
if p.size() > 5 && me.starts_with(p, p.size()-5, ".hako") == 1 {
|
||||
p = p.substring(0, p.size()-5)
|
||||
} else {
|
||||
if p.size() > 6 && me.starts_with(p, p.size()-6, ".nyash") == 1 {
|
||||
p = p.substring(0, p.size()-6)
|
||||
}
|
||||
}
|
||||
name = p
|
||||
}
|
||||
} else {
|
||||
name = target
|
||||
if alias != null { name = alias }
|
||||
}
|
||||
|
||||
local entry = "{\"name\":\"" + me.esc_json(name) + "\""
|
||||
if path != null { entry = entry + ",\"path\":\"" + me.esc_json(path) + "\"" }
|
||||
entry = entry + "}"
|
||||
|
||||
if cur == "[]" {
|
||||
me.usings_json = "[" + entry + "]"
|
||||
return 0
|
||||
}
|
||||
|
||||
local pos = cur.lastIndexOf("]")
|
||||
if pos < 0 {
|
||||
me.usings_json = "[" + entry + "]"
|
||||
return 0
|
||||
}
|
||||
|
||||
me.usings_json = cur.substring(0, pos) + "," + entry + "]"
|
||||
return 0
|
||||
}
|
||||
|
||||
extract_usings(src) {
|
||||
me.usings_json = UsingCollectorBox.collect(src)
|
||||
return 0
|
||||
}
|
||||
|
||||
get_usings_json() {
|
||||
return me.usings_json
|
||||
}
|
||||
|
||||
// === Delegation to ParserExprBox ===
|
||||
parse_expr2(src, i) {
|
||||
local expr = new ParserExprBox()
|
||||
return expr.parse_expr2(src, i, me)
|
||||
}
|
||||
|
||||
// === Delegation to ParserStmtBox ===
|
||||
parse_stmt2(src, i) {
|
||||
local stmt = new ParserStmtBox()
|
||||
return stmt.parse(src, i, me)
|
||||
}
|
||||
|
||||
// === Delegation to ParserControlBox ===
|
||||
parse_block2(src, i) {
|
||||
local ctrl = new ParserControlBox()
|
||||
return ctrl.parse_block(src, i, me)
|
||||
}
|
||||
|
||||
// === Top-level program parser ===
|
||||
parse_program2(src) {
|
||||
local i = me.skip_ws(src, 0)
|
||||
local body = "["
|
||||
local first = 1
|
||||
local cont_prog = 1
|
||||
|
||||
loop(cont_prog == 1) {
|
||||
i = me.skip_ws(src, i)
|
||||
|
||||
if i >= src.size() {
|
||||
cont_prog = 0
|
||||
} else {
|
||||
local start_i = i
|
||||
local s = me.parse_stmt2(src, i)
|
||||
i = me.gpos_get()
|
||||
|
||||
// Progress guard
|
||||
if i <= start_i {
|
||||
if i < src.size() { i = i + 1 }
|
||||
else { i = src.size() }
|
||||
me.gpos_set(i)
|
||||
}
|
||||
|
||||
// consume optional semicolons
|
||||
local done2 = 0
|
||||
local guard2 = 0
|
||||
local max2 = 100000
|
||||
|
||||
loop(done2 == 0) {
|
||||
if guard2 > max2 { done2 = 1 }
|
||||
else { guard2 = guard2 + 1 }
|
||||
|
||||
local before2 = i
|
||||
i = me.skip_ws(src, i)
|
||||
|
||||
if i < src.size() && src.substring(i, i+1) == ";" {
|
||||
i = i + 1
|
||||
} else {
|
||||
done2 = 1
|
||||
}
|
||||
|
||||
if i == before2 { done2 = 1 }
|
||||
}
|
||||
|
||||
if s.size() > 0 {
|
||||
if first == 1 {
|
||||
body = body + s
|
||||
first = 0
|
||||
} else {
|
||||
body = body + "," + s
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
body = body + "]"
|
||||
return "{\"version\":0,\"kind\":\"Program\",\"body\":" + body + "}"
|
||||
}
|
||||
}
|
||||
|
||||
static box ParserStub {
|
||||
main(args) {
|
||||
return 0
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user