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
355
lang/src/compiler/parser/expr/parser_expr_box.hako
Normal file
355
lang/src/compiler/parser/expr/parser_expr_box.hako
Normal file
@ -0,0 +1,355 @@
|
||||
// Moved from apps/selfhost-compiler/boxes/parser/expr/parser_expr_box.hako
|
||||
// ParserExprBox — expression parser coordinator
|
||||
// Responsibility: Parse expressions and delegate to specialized boxes
|
||||
// API: parse(src, i, ctx) -> JSON (delegates to parse_expr2)
|
||||
|
||||
using lang.compiler.parser.scan.parser_number_scan_box
|
||||
using lang.compiler.parser.expr.parser_peek_box
|
||||
using lang.compiler.parser.expr.parser_literal_box
|
||||
|
||||
static box ParserExprBox {
|
||||
parse_number2(src, i, ctx) {
|
||||
local pair = ParserNumberScanBox.scan_int(src, i)
|
||||
local at = pair.lastIndexOf("@")
|
||||
local json = pair.substring(0, at)
|
||||
local pos = i
|
||||
if at >= 0 { pos = ctx.to_int(pair.substring(at+1, pair.size())) }
|
||||
ctx.gpos_set(pos)
|
||||
return json
|
||||
}
|
||||
|
||||
parse_string2(src, i, ctx) {
|
||||
local n = src.size()
|
||||
local j = i + 1
|
||||
local out = ""
|
||||
local guard = 0
|
||||
local max = 200000
|
||||
|
||||
loop(j < n) {
|
||||
if guard > max { break }
|
||||
guard = guard + 1
|
||||
local ch = src.substring(j, j+1)
|
||||
|
||||
if ch == "\"" {
|
||||
j = j + 1
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Str\",\"value\":\"" + ctx.esc_json(out) + "\"}"
|
||||
}
|
||||
|
||||
if ch == "\\" && j + 1 < n {
|
||||
local nx = src.substring(j+1, j+2)
|
||||
if nx == "\"" { out = out + "\"" j = j + 2 }
|
||||
else { if nx == "\\" { out = out + "\\" j = j + 2 }
|
||||
else { if nx == "n" { out = out + "\n" j = j + 2 }
|
||||
else { if nx == "r" { out = out + "\r" j = j + 2 }
|
||||
else { if nx == "t" { out = out + "\t" j = j + 2 }
|
||||
else { if nx == "u" && j + 5 < n { out = out + src.substring(j, j+6) j = j + 6 }
|
||||
else { out = out + nx j = j + 2 } } } } } }
|
||||
} else {
|
||||
out = out + ch
|
||||
j = j + 1
|
||||
}
|
||||
}
|
||||
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Str\",\"value\":\"" + ctx.esc_json(out) + "\"}"
|
||||
}
|
||||
|
||||
parse_factor2(src, i, ctx) {
|
||||
local j = ctx.skip_ws(src, i)
|
||||
if j >= src.size() {
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Int\",\"value\":0}"
|
||||
}
|
||||
|
||||
if ctx.starts_with_kw(src, j, "true") == 1 {
|
||||
ctx.gpos_set(j + 4)
|
||||
return "{\"type\":\"Bool\",\"value\":true}"
|
||||
}
|
||||
|
||||
if ctx.starts_with_kw(src, j, "false") == 1 {
|
||||
ctx.gpos_set(j + 5)
|
||||
return "{\"type\":\"Bool\",\"value\":false}"
|
||||
}
|
||||
|
||||
if ctx.starts_with_kw(src, j, "null") == 1 {
|
||||
ctx.gpos_set(j + 4)
|
||||
return "{\"type\":\"Null\"}"
|
||||
}
|
||||
|
||||
// Peek expression: delegate to ParserPeekBox
|
||||
if ctx.starts_with_kw(src, j, "peek") == 1 {
|
||||
j = j + 4
|
||||
return ParserPeekBox.parse(src, j, ctx)
|
||||
}
|
||||
|
||||
local ch = src.substring(j, j+1)
|
||||
|
||||
// Parenthesized
|
||||
if ch == "(" {
|
||||
local inner = me.parse_expr2(src, j + 1, ctx)
|
||||
local k = ctx.gpos_get()
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == ")" { k = k + 1 }
|
||||
ctx.gpos_set(k)
|
||||
return inner
|
||||
}
|
||||
|
||||
// String literal
|
||||
if ch == "\"" {
|
||||
return me.parse_string2(src, j, ctx)
|
||||
}
|
||||
|
||||
// Map literal: delegate to ParserLiteralBox
|
||||
if ch == "{" {
|
||||
return ParserLiteralBox.parse_map(src, j, ctx)
|
||||
}
|
||||
|
||||
// Array literal: delegate to ParserLiteralBox
|
||||
if ch == "[" {
|
||||
return ParserLiteralBox.parse_array(src, j, ctx)
|
||||
}
|
||||
|
||||
// new Class(args)
|
||||
if ctx.starts_with_kw(src, j, "new") == 1 {
|
||||
local p = ctx.skip_ws(src, j + 3)
|
||||
local idp = ctx.read_ident2(src, p)
|
||||
local at = idp.lastIndexOf("@")
|
||||
local cls = idp.substring(0, at)
|
||||
local k = ctx.to_int(idp.substring(at+1, idp.size()))
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == "(" { k = k + 1 }
|
||||
local args_and_pos = me.parse_args2(src, k, ctx)
|
||||
local at2 = args_and_pos.lastIndexOf("@")
|
||||
local args_json = args_and_pos.substring(0, at2)
|
||||
k = ctx.to_int(args_and_pos.substring(at2+1, args_and_pos.size()))
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == ")" { k = k + 1 }
|
||||
ctx.gpos_set(k)
|
||||
return "{\"type\":\"New\",\"class\":\"" + cls + "\",\"args\":" + args_json + "}"
|
||||
}
|
||||
|
||||
// Identifier / Call / Method chain
|
||||
if ctx.is_alpha(ch) {
|
||||
local idp = ctx.read_ident2(src, j)
|
||||
local at = idp.lastIndexOf("@")
|
||||
local name = idp.substring(0, at)
|
||||
local k = ctx.to_int(idp.substring(at+1, idp.size()))
|
||||
local node = "{\"type\":\"Var\",\"name\":\"" + name + "\"}"
|
||||
local cont2 = 1
|
||||
|
||||
loop(cont2 == 1) {
|
||||
k = ctx.skip_ws(src, k)
|
||||
local tch = src.substring(k, k+1)
|
||||
|
||||
if tch == "(" {
|
||||
k = k + 1
|
||||
local args_and_pos = me.parse_args2(src, k, ctx)
|
||||
local at2 = args_and_pos.lastIndexOf("@")
|
||||
local args_json = args_and_pos.substring(0, at2)
|
||||
k = ctx.to_int(args_and_pos.substring(at2+1, args_and_pos.size()))
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == ")" { k = k + 1 }
|
||||
node = "{\"type\":\"Call\",\"name\":\"" + name + "\",\"args\":" + args_json + "}"
|
||||
} else {
|
||||
if tch == "." {
|
||||
k = k + 1
|
||||
k = ctx.skip_ws(src, k)
|
||||
local midp = ctx.read_ident2(src, k)
|
||||
local at3 = midp.lastIndexOf("@")
|
||||
local mname = midp.substring(0, at3)
|
||||
k = ctx.to_int(midp.substring(at3+1, midp.size()))
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == "(" { k = k + 1 }
|
||||
local args2 = me.parse_args2(src, k, ctx)
|
||||
local at4 = args2.lastIndexOf("@")
|
||||
local args_json2 = args2.substring(0, at4)
|
||||
k = ctx.to_int(args2.substring(at4+1, args2.size()))
|
||||
k = ctx.skip_ws(src, k)
|
||||
if src.substring(k, k+1) == ")" { k = k + 1 }
|
||||
node = "{\"type\":\"Method\",\"recv\":" + node + ",\"method\":\"" + mname + "\",\"args\":" + args_json2 + "}"
|
||||
} else {
|
||||
cont2 = 0
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
ctx.gpos_set(k)
|
||||
return node
|
||||
}
|
||||
|
||||
// Fallback: number
|
||||
return me.parse_number2(src, j, ctx)
|
||||
}
|
||||
|
||||
parse_unary2(src, i, ctx) {
|
||||
local j = ctx.skip_ws(src, i)
|
||||
if src.substring(j, j+1) == "-" {
|
||||
local rhs = me.parse_factor2(src, j + 1, ctx)
|
||||
j = ctx.gpos_get()
|
||||
local zero = "{\"type\":\"Int\",\"value\":0}"
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Binary\",\"op\":\"-\",\"lhs\":" + zero + ",\"rhs\":" + rhs + "}"
|
||||
}
|
||||
return me.parse_factor2(src, j, ctx)
|
||||
}
|
||||
|
||||
parse_term2(src, i, ctx) {
|
||||
local lhs = me.parse_unary2(src, i, ctx)
|
||||
local j = ctx.gpos_get()
|
||||
local cont = 1
|
||||
|
||||
loop(cont == 1) {
|
||||
j = ctx.skip_ws(src, j)
|
||||
if j >= src.size() {
|
||||
cont = 0
|
||||
} else {
|
||||
local op = src.substring(j, j+1)
|
||||
if op != "*" && op != "/" {
|
||||
cont = 0
|
||||
} else {
|
||||
local rhs = me.parse_unary2(src, j+1, ctx)
|
||||
j = ctx.gpos_get()
|
||||
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
ctx.gpos_set(j)
|
||||
return lhs
|
||||
}
|
||||
|
||||
parse_sum2(src, i, ctx) {
|
||||
local lhs = me.parse_term2(src, i, ctx)
|
||||
local j = ctx.gpos_get()
|
||||
local cont = 1
|
||||
|
||||
loop(cont == 1) {
|
||||
j = ctx.skip_ws(src, j)
|
||||
if j >= src.size() {
|
||||
cont = 0
|
||||
} else {
|
||||
local op = src.substring(j, j+1)
|
||||
if op != "+" && op != "-" {
|
||||
cont = 0
|
||||
} else {
|
||||
local rhs = me.parse_term2(src, j+1, ctx)
|
||||
j = ctx.gpos_get()
|
||||
lhs = "{\"type\":\"Binary\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
ctx.gpos_set(j)
|
||||
return lhs
|
||||
}
|
||||
|
||||
parse_compare2(src, i, ctx) {
|
||||
local lhs = me.parse_sum2(src, i, ctx)
|
||||
local j = ctx.gpos_get()
|
||||
j = ctx.skip_ws(src, j)
|
||||
local two = src.substring(j, j+2)
|
||||
local one = src.substring(j, j+1)
|
||||
local op = ""
|
||||
|
||||
if two == "==" || two == "!=" || two == "<=" || two == ">=" {
|
||||
op = two
|
||||
j = j + 2
|
||||
} else {
|
||||
if one == "<" || one == ">" {
|
||||
op = one
|
||||
j = j + 1
|
||||
}
|
||||
}
|
||||
|
||||
if op == "" {
|
||||
ctx.gpos_set(j)
|
||||
return lhs
|
||||
}
|
||||
|
||||
local rhs = me.parse_sum2(src, j, ctx)
|
||||
j = ctx.gpos_get()
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Compare\",\"op\":\"" + op + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
|
||||
}
|
||||
|
||||
parse_expr2(src, i, ctx) {
|
||||
local lhs = me.parse_compare2(src, i, ctx)
|
||||
local j = ctx.gpos_get()
|
||||
local cont = 1
|
||||
|
||||
loop(cont == 1) {
|
||||
j = ctx.skip_ws(src, j)
|
||||
local two = src.substring(j, j+2)
|
||||
if two != "&&" && two != "||" {
|
||||
cont = 0
|
||||
} else {
|
||||
local rhs = me.parse_compare2(src, j+2, ctx)
|
||||
j = ctx.gpos_get()
|
||||
lhs = "{\"type\":\"Logical\",\"op\":\"" + two + "\",\"lhs\":" + lhs + ",\"rhs\":" + rhs + "}"
|
||||
}
|
||||
}
|
||||
|
||||
j = ctx.skip_ws(src, j)
|
||||
if src.substring(j, j+1) == "?" {
|
||||
j = j + 1
|
||||
j = ctx.skip_ws(src, j)
|
||||
local then_expr = me.parse_expr2(src, j, ctx)
|
||||
j = ctx.gpos_get()
|
||||
j = ctx.skip_ws(src, j)
|
||||
if src.substring(j, j+1) == ":" { j = j + 1 }
|
||||
j = ctx.skip_ws(src, j)
|
||||
local else_expr = me.parse_expr2(src, j, ctx)
|
||||
j = ctx.gpos_get()
|
||||
if else_expr.size() == 0 { else_expr = "{\"type\":\"Int\",\"value\":0}" }
|
||||
ctx.gpos_set(j)
|
||||
return "{\"type\":\"Ternary\",\"cond\":" + lhs + ",\"then\":" + then_expr + ",\"else\":" + else_expr + "}"
|
||||
}
|
||||
|
||||
ctx.gpos_set(j)
|
||||
return lhs
|
||||
}
|
||||
|
||||
parse_args2(src, i, ctx) {
|
||||
local j = ctx.skip_ws(src, i)
|
||||
local n = src.size()
|
||||
local out = "["
|
||||
j = ctx.skip_ws(src, j)
|
||||
|
||||
if j < n && src.substring(j, j+1) == ")" {
|
||||
return "[]@" + ctx.i2s(j)
|
||||
}
|
||||
|
||||
// first argument
|
||||
local e = me.parse_expr2(src, j, ctx)
|
||||
j = ctx.gpos_get()
|
||||
out = out + e
|
||||
|
||||
// subsequent arguments with guard
|
||||
local cont_args = 1
|
||||
local guard = 0
|
||||
local max = 100000
|
||||
|
||||
loop(cont_args == 1) {
|
||||
if guard > max { cont_args = 0 } else { guard = guard + 1 }
|
||||
local before = j
|
||||
j = ctx.skip_ws(src, j)
|
||||
|
||||
if j < n && src.substring(j, j+1) == "," {
|
||||
j = j + 1
|
||||
j = ctx.skip_ws(src, j)
|
||||
e = me.parse_expr2(src, j, ctx)
|
||||
j = ctx.gpos_get()
|
||||
out = out + "," + e
|
||||
} else {
|
||||
cont_args = 0
|
||||
}
|
||||
|
||||
if j == before { cont_args = 0 }
|
||||
}
|
||||
|
||||
out = out + "]"
|
||||
return out + "@" + ctx.i2s(j)
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user