Skip to content
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

fix(parser): emit lexer error on escaped surrogate to avoid panic #658

Merged
merged 2 commits into from
Sep 28, 2023
Merged

fix(parser): emit lexer error on escaped surrogate to avoid panic #658

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ef6d7c6
fix(parser): emit lexer error on escaped surrogate to avoid panic
SimonSapin Sep 25, 2023
644c423
Changelog
SimonSapin Sep 28, 2023
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
2 changes: 2 additions & 0 deletions crates/apollo-parser/src/ast/node_ext.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,6 +167,8 @@ fn unescape_string(input: &str) -> String {
break;
};

// TODO: https://github.com/apollographql/apollo-rs/issues/657 needs
// changes both here and in `lexer/mod.rs`
let mut unicode = || {
// 1. Let value be the 16-bit hexadecimal value represented
// by the sequence of hexadecimal digits within EscapedUnicode.
Expand Down
4 changes: 2 additions & 2 deletions crates/apollo-parser/src/lexer/cursor.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,8 +6,8 @@ use crate::Error;
#[derive(Debug, Clone)]
pub(crate) struct Cursor<'a> {
index: usize,
offset: usize,
source: &'a str,
pub(super) offset: usize,
pub(super) source: &'a str,
chars: CharIndices<'a>,
pending: Option<char>,
pub(crate) err: Option<Error>,
Expand Down
19 changes: 18 additions & 1 deletion crates/apollo-parser/src/lexer/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -326,7 +326,24 @@ impl<'a> Cursor<'a> {
_ => {
if remaining <= 1 {
state = State::StringLiteral;

let hex_end = self.offset + 1;
let hex_start = hex_end - 4;
let hex = &self.source[hex_start..hex_end];
// `is_ascii_hexdigit()` checks in previous iterations ensures
// this `unwrap()` does not panic:
let code_point = u32::from_str_radix(hex, 16).unwrap();
if char::from_u32(code_point).is_none() {
// TODO: https://github.com/apollographql/apollo-rs/issues/657 needs
// changes both here and in `ast/node_ext.rs`
let escape_sequence_start = hex_start - 2; // include "\u"
let escape_sequence = &self.source[escape_sequence_start..hex_end];
self.add_err(Error::new(
"surrogate code point is invalid in unicode escape sequence \
(paired surrogate not supported yet: \
https://github.com/apollographql/apollo-rs/issues/657)",
escape_sequence.to_owned(),
));
}
continue;
}

Expand Down
12 changes: 12 additions & 0 deletions crates/apollo-parser/test_data/lexer/err/0030_escaped_surrogate.graphql
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
# TODO: move these cases back to ok/0004_string_value.graphql when
# https://github.com/apollographql/apollo-rs/issues/657 is implemented
"string with unicode surrogate pair escape \uD83D\uDE00"
"string with minimal surrogate pair escape \uD800\uDC00"
"string with maximal surrogate pair escape \uDBFF\uDFFF"

# TODO: emit two errors: https://github.com/apollographql/apollo-rs/issues/319
"split pair \uD83D \uDE00"

"Backwards pair \uDE00\uD83D"
"Lone lead surrogate \uD83E"
"Lone trail surrogate \uDD80"
21 changes: 21 additions & 0 deletions crates/apollo-parser/test_data/lexer/err/0030_escaped_surrogate.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
COMMENT@0:66 "# TODO: move these cases back to ok/0004_string_value.graphql when"
WHITESPACE@66:67 "\n"
COMMENT@67:137 "# https://github.com/apollographql/apollo-rs/issues/657 is implemented"
WHITESPACE@137:138 "\n"
ERROR@138:194 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "string with unicode surrogate pair escape \uD83D\uDE00"
WHITESPACE@194:195 "\n"
ERROR@195:251 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "string with minimal surrogate pair escape \uD800\uDC00"
WHITESPACE@251:252 "\n"
ERROR@252:308 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "string with maximal surrogate pair escape \uDBFF\uDFFF"
WHITESPACE@308:310 "\n\n"
COMMENT@310:388 "# TODO: emit two errors: https://github.com/apollographql/apollo-rs/issues/319"
WHITESPACE@388:389 "\n"
ERROR@389:415 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "split pair \uD83D \uDE00"
WHITESPACE@415:417 "\n\n"
ERROR@417:446 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "Backwards pair \uDE00\uD83D"
WHITESPACE@446:447 "\n"
ERROR@447:475 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "Lone lead surrogate \uD83E"
WHITESPACE@475:476 "\n"
ERROR@476:505 "surrogate code point is invalid in unicode escape sequence (paired surrogate not supported yet: https://github.com/apollographql/apollo-rs/issues/657)" "Lone trail surrogate \uDD80"
WHITESPACE@505:506 "\n"
EOF@506:506
3 changes: 0 additions & 3 deletions crates/apollo-parser/test_data/lexer/ok/0004_string_value.graphql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,9 +2,6 @@
"simple"
" white space "
"unicode \u1234\u5678\u90AB\uCDEF"
"string with unicode surrogate pair escape \uD83D\uDE00"
"string with minimal surrogate pair escape \uD800\uDC00"
"string with maximal surrogate pair escape \uDBFF\uDFFF"
"string with \"escaped\" characters"
"string with multiple languages котя, 猫, ねこ, قطة"
"""
Expand Down
20 changes: 7 additions & 13 deletions crates/apollo-parser/test_data/lexer/ok/0004_string_value.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,16 +6,10 @@ STRING_VALUE@12:27 "\" white space \""
WHITESPACE@27:28 "\n"
STRING_VALUE@28:62 "\"unicode \\u1234\\u5678\\u90AB\\uCDEF\""
WHITESPACE@62:63 "\n"
STRING_VALUE@63:119 "\"string with unicode surrogate pair escape \\uD83D\\uDE00\""
WHITESPACE@119:120 "\n"
STRING_VALUE@120:176 "\"string with minimal surrogate pair escape \\uD800\\uDC00\""
WHITESPACE@176:177 "\n"
STRING_VALUE@177:233 "\"string with maximal surrogate pair escape \\uDBFF\\uDFFF\""
WHITESPACE@233:234 "\n"
STRING_VALUE@234:270 "\"string with \\\"escaped\\\" characters\""
WHITESPACE@270:271 "\n"
STRING_VALUE@271:333 "\"string with multiple languages котя, 猫, ねこ, قطة\""
WHITESPACE@333:334 "\n"
STRING_VALUE@334:421 "\"\"\"\nblock string with unusual whitespaces\na b c\nd\n\ne\tf\ng\u{2028}\u{2029}h\ni\u{b}j\u{c}k\u{feff}l\u{85}\u{200e}\u{200f}m\n\"\"\""
WHITESPACE@421:422 "\n"
EOF@422:422
STRING_VALUE@63:99 "\"string with \\\"escaped\\\" characters\""
WHITESPACE@99:100 "\n"
STRING_VALUE@100:162 "\"string with multiple languages котя, 猫, ねこ, قطة\""
WHITESPACE@162:163 "\n"
STRING_VALUE@163:250 "\"\"\"\nblock string with unusual whitespaces\na b c\nd\n\ne\tf\ng\u{2028}\u{2029}h\ni\u{b}j\u{c}k\u{feff}l\u{85}\u{200e}\u{200f}m\n\"\"\""
WHITESPACE@250:251 "\n"
EOF@251:251