Skip to content

Update elm/bytes _Bytes_read_string to use array instead of string as accumulator #110

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Aug 23, 2025
Merged

Update elm/bytes _Bytes_read_string to use array instead of string as accumulator #110

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions eslint.config.mjs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ export default defineConfig([
{ files: ["lib/browser.js"], languageOptions: { globals: globals.browser } },
{ files: ["lib/node.js"], languageOptions: { globals: globals.node } },
{ files: ["try/**/*.{js,mjs,cjs}"], languageOptions: { globals: { ...globals.browser, ...globals.node } } },
{ files: ["scripts/*.js"], languageOptions: { globals: globals.node } },
{
files: ["**/*.{js,mjs,cjs}"],
plugins: { js },
Expand Down
1 change: 1 addition & 0 deletions scripts/build.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@ js="$filepath.js"
min="$filepath.min.js"

guida make --optimize --output=$js $elm_entry
node scripts/replace-bytes-read-string.js $js

uglifyjs $js --compress "pure_funcs=[F2,F3,F4,F5,F6,F7,F8,F9,A2,A3,A4,A5,A6,A7,A8,A9],pure_getters,keep_fargs=false,unsafe_comps,unsafe" | uglifyjs --mangle --output $min

Expand Down
82 changes: 82 additions & 0 deletions scripts/replace-bytes-read-string.js
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,82 @@
#!/usr/bin/env node

/* This change prevents V8 from retaining large "concatenated string" chains, which can cause OOMs.
Tested against `rtfeldman/elm-css` compilation.

See the related discussion for context: https://discourse.elm-lang.org/t/guida-compiler-was-there-are-3-elm-compilers-written-in-elm/10329/34
and issue: https://github.com/guida-lang/compiler/issues/107
*/

const fs = require('node:fs');

const argv = process.argv.slice(2);
const path = argv[0];

const data = fs
.readFileSync(path, { encoding: 'utf8', flag: 'r' })
.replace(`var _Bytes_read_string = F3(function(len, bytes, offset)
{
var string = '';
var end = offset + len;
for (; offset < end;)
{
var byte = bytes.getUint8(offset++);
string +=
(byte < 128)
? String.fromCharCode(byte)
:
((byte & 0xE0 /* 0b11100000 */) === 0xC0 /* 0b11000000 */)
? String.fromCharCode((byte & 0x1F /* 0b00011111 */) << 6 | bytes.getUint8(offset++) & 0x3F /* 0b00111111 */)
:
((byte & 0xF0 /* 0b11110000 */) === 0xE0 /* 0b11100000 */)
? String.fromCharCode(
(byte & 0xF /* 0b00001111 */) << 12
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 6
| bytes.getUint8(offset++) & 0x3F /* 0b00111111 */
)
:
(byte =
((byte & 0x7 /* 0b00000111 */) << 18
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 12
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 6
| bytes.getUint8(offset++) & 0x3F /* 0b00111111 */
) - 0x10000
, String.fromCharCode(Math.floor(byte / 0x400) + 0xD800, byte % 0x400 + 0xDC00)
);
}
return _Utils_Tuple2(offset, string);
});`, `var _Bytes_read_string = F3(function(len, bytes, offset)
{
var string = [];
var end = offset + len;
for (; offset < end;)
{
var byte = bytes.getUint8(offset++);
string.push(
(byte < 128)
? String.fromCharCode(byte)
:
((byte & 0xE0 /* 0b11100000 */) === 0xC0 /* 0b11000000 */)
? String.fromCharCode((byte & 0x1F /* 0b00011111 */) << 6 | bytes.getUint8(offset++) & 0x3F /* 0b00111111 */)
:
((byte & 0xF0 /* 0b11110000 */) === 0xE0 /* 0b11100000 */)
? String.fromCharCode(
(byte & 0xF /* 0b00001111 */) << 12
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 6
| bytes.getUint8(offset++) & 0x3F /* 0b00111111 */
)
:
(byte =
((byte & 0x7 /* 0b00000111 */) << 18
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 12
| (bytes.getUint8(offset++) & 0x3F /* 0b00111111 */) << 6
| bytes.getUint8(offset++) & 0x3F /* 0b00111111 */
) - 0x10000
, String.fromCharCode(Math.floor(byte / 0x400) + 0xD800, byte % 0x400 + 0xDC00)
)
);
}
return _Utils_Tuple2(offset, string.join(''));
});`);

fs.writeFileSync(path, data, { encoding: 'utf8', flag: 'w' });
11 changes: 2 additions & 9 deletions src/Compiler/Parse/Primitives.elm
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -483,24 +483,17 @@ snippetEncoder (Snippet { fptr, offset, length, offRow, offCol }) =

snippetDecoder : BD.Decoder Snippet
snippetDecoder =
-- This `String.toList` -> `String.fromList` round-trip prevents V8 from retaining
-- large "concatenated string" chains, which can cause OOMs.
-- Tested against `rtfeldman/elm-css` compilation.
-- The split call avoids elm-review flags for this pattern.
--
-- See the related discussion for context:
-- https://discourse.elm-lang.org/t/guida-compiler-was-there-are-3-elm-compilers-written-in-elm/10329/25
BD.map5
(\fptr offset length offRow offCol ->
Snippet
{ fptr = String.fromList fptr
{ fptr = fptr
, offset = offset
, length = length
, offRow = offRow
, offCol = offCol
}
)
(BD.map String.toList BD.string)
BD.string
BD.int
BD.int
BD.int
Expand Down
Loading