workshopsoft/ludic
2
0
Fork 0
Code Issues Pull requests Projects Releases 26 Packages Wiki Activity Actions

Json.parse decodes an escaped string in one pass into one buffer - joined a character at a time, every shorter copy was kept, and a long escaped text took gigabytes (ui-preview: 6.5 GB in 3 s); protocol-v1.md states the @import path rule and the key a file override answers to

Browse source 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Orkun ÇAKILKAYA 2026-09-29 21:53:57 +03:00
parent ea51307eaa
commit 4fb1dc0ddc
3 changed files with 64 additions and 30 deletions
Download patch file Download diff file Expand all files Collapse all files

77
runtime/native/value.ludic
View file

@ -605,36 +605,67 @@ function jp_string(p: JP) -> string {
return p.s[a..p.n]
}
# the same with escapes in it, the rare case, a character at a time
# the same with escapes in it: decoded in one pass into one buffer, sized by the text up to the closing
# quote: an escape decodes to at most one and a half times its bytes (\uXXXX is six bytes and at most
# three out; a \u with bad hex is U+FFFD, three for two).
# Joining the result a character at a time kept every shorter copy - a long escaped text made
# gigabytes of them (ui-preview's model, 6.5 GB in three seconds).
function jp_string_esc(p: JP) -> string {
var out = ""
var end = p.i
while end < p.n and p.s[end] != '"' {
if p.s[end] == '\\' { end += 1 }
end += 1
}
if end > p.n { end = p.n }
let out = bytes((end - p.i) * 3 / 2 + 4) # a \u with bad hex is U+FFFD: three bytes for two
var w = 0
while p.i < p.n {
let c = p.s[p.i]
if c == '"' { p.i += 1; return out } # closing quote
if c == '"' { # closing quote
p.i += 1
out[w] = 0
let done: string = out
return done
}
if c == '\\' { # escape
p.i += 1
if p.i < p.n {
let e = p.s[p.i]
if e == 'n' { out += "\n" }
else if e == 't' { out += "\t" }
else if e == 'r' { out += "\r" }
else if e == 'b' or e == 'f' or e == 'u' {
var code = 8
if e == 'f' { code = 12 }
if e == 'u' { code = jp_code(p) } # leaves p.i on the escape's last hex digit
let u = jp_utf8(code)
out += u
free(u)
}
else { out += p.s[p.i..p.i + 1] } # \" \\ \/ -> the literal char
if e == 'n' { out[w] = 10; w += 1 }
else if e == 't' { out[w] = 9; w += 1 }
else if e == 'r' { out[w] = 13; w += 1 }
else if e == 'b' { out[w] = 8; w += 1 }
else if e == 'f' { out[w] = 12; w += 1 }
else if e == 'u' { w = jp_put_utf8(out, w, jp_code(p)) } # leaves p.i on the escape's last hex digit
else { out[w] = e; w += 1 } # \" \\ \/ -> the literal char
p.i += 1
}
} else {
out += p.s[p.i..p.i + 1]
out[w] = c
w += 1
p.i += 1
}
}
return out
out[w] = 0
let rest: string = out
return rest
}
# one code point's UTF-8 bytes into out at w; where the next one goes
function jp_put_utf8(out: bytes, w: int, c: int) -> int {
if c < 128 {
out[w] = c
return w + 1
}
if c < 2048 {
out[w] = 192 | (c >> 6); out[w + 1] = 128 | (c & 63)
return w + 2
}
if c < 65536 {
out[w] = 224 | (c >> 12); out[w + 1] = 128 | ((c >> 6) & 63); out[w + 2] = 128 | (c & 63)
return w + 3
}
out[w] = 240 | (c >> 18); out[w + 1] = 128 | ((c >> 12) & 63); out[w + 2] = 128 | ((c >> 6) & 63); out[w + 3] = 128 | (c & 63)
return w + 4
}
# \uXXXX, p.i on the 'u': the code point, a surrogate pair joined (\ud83d\ude00 is one emoji); a
@ -671,18 +702,6 @@ function jp_hex4(p: JP, at: int) -> int {
}
return v
}
# one code point as a UTF-8 string
function jp_utf8(c: int) -> string {
let out = bytes(5)
var n = 0
if c < 128 { out[0] = c; n = 1 }
else if c < 2048 { out[0] = 192 | (c >> 6); out[1] = 128 | (c & 63); n = 2 }
else if c < 65536 { out[0] = 224 | (c >> 12); out[1] = 128 | ((c >> 6) & 63); out[2] = 128 | (c & 63); n = 3 }
else { out[0] = 240 | (c >> 18); out[1] = 128 | ((c >> 12) & 63); out[2] = 128 | ((c >> 6) & 63); out[3] = 128 | (c & 63); n = 4 }
out[n] = 0
let t: string = out
return t
}
# read a number; a '.' makes it a fixed node, otherwise an int node.
function jp_number(p: JP) -> Val {