Skip to content

parser: fix idents with uescape #553

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
kodiakhq merged 1 commit into from
Jun 20, 2025
Merged

parser: fix idents with uescape #553

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
86 changes: 51 additions & 35 deletions crates/squawk_parser/src/grammar.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1485,11 +1485,7 @@ fn opt_name(p: &mut Parser<'_>) -> Option<CompletedMarker> {
return None;
}
let m = p.start();
if p.eat(IDENT) {
if p.eat(UESCAPE_KW) {
p.expect(STRING);
}
} else {
if !opt_ident(p) {
p.bump_any();
}
Some(m.complete(p, NAME))
Expand Down Expand Up @@ -1894,7 +1890,9 @@ fn name_ref_(p: &mut Parser<'_>) -> Option<CompletedMarker> {
INTERVAL_TYPE
}
_ => {
p.bump_any();
if !opt_ident(p) {
p.bump_any();
}
NAME_REF
}
};
Expand Down Expand Up @@ -3384,7 +3382,8 @@ fn opt_include_columns(p: &mut Parser<'_>) -> Option<CompletedMarker> {

// [ WITH ( storage_parameter [= value] [, ... ] ) ]
fn opt_with_params(p: &mut Parser<'_>) -> Option<CompletedMarker> {
if p.at(WITH_KW) {
// check for both in case someone forgot a semi after `create table`
if p.at(WITH_KW) && p.nth_at(1, L_PAREN) {
let m = p.start();
p.bump(WITH_KW);
p.expect(L_PAREN);
Expand Down Expand Up @@ -4082,7 +4081,9 @@ fn opt_compression_method(p: &mut Parser<'_>) -> Option<CompletedMarker> {
let m = p.start();
// [ COMPRESSION compression_method ]
if p.eat(COMPRESSION_KW) && (p.at(DEFAULT_KW) || p.at(IDENT)) {
p.bump_any();
if !opt_ident(p) && !p.eat(DEFAULT_KW) {
p.error("expected default or identifier");
}
Some(m.complete(p, COMPRESSION_METHOD))
} else {
m.abandon(p);
Expand All @@ -4094,7 +4095,9 @@ fn opt_storage(p: &mut Parser<'_>) -> Option<CompletedMarker> {
let m = p.start();
// [ STORAGE { PLAIN | EXTERNAL | EXTENDED | MAIN | DEFAULT } ]
if p.eat(STORAGE_KW) && (p.at(DEFAULT_KW) || p.at(EXTERNAL_KW) || p.at(IDENT)) {
p.bump_any();
if !opt_ident(p) {
p.bump_any();
}
Some(m.complete(p, STORAGE))
} else {
m.abandon(p);
Expand Down Expand Up @@ -4290,7 +4293,7 @@ fn window_definition(p: &mut Parser<'_>) -> Option<CompletedMarker> {
return None;
}
let m = p.start();
p.eat(IDENT);
opt_ident(p);
if p.eat(PARTITION_KW) {
p.expect(BY_KW);
if expr(p).is_none() {
Expand Down Expand Up @@ -4656,7 +4659,7 @@ fn part_elem(p: &mut Parser<'_>, allow_extra_params: bool) -> bool {
}
opt_collate(p);
// [ opclass ]
p.eat(IDENT);
opt_ident(p);
if allow_extra_params {
// [ ( opclass_parameter = value [, ... ] ) ]
if p.eat(L_PAREN) {
Expand Down Expand Up @@ -4716,10 +4719,10 @@ fn partition_option(p: &mut Parser<'_>) {
// FOR VALUES WITH (modulus 5, remainder 0)
if p.eat(WITH_KW) {
p.expect(L_PAREN);
p.expect(IDENT);
ident(p);
p.expect(INT_NUMBER);
p.expect(COMMA);
p.expect(IDENT);
ident(p);
p.expect(INT_NUMBER);
p.expect(R_PAREN);
// FOR VALUES IN '(' expr_list ')'
Expand Down Expand Up @@ -4951,7 +4954,9 @@ fn opt_partition_by(p: &mut Parser<'_>) -> Option<CompletedMarker> {
p.expect(BY_KW);
// name
if p.at_ts(TYPE_KEYWORDS) || p.at(IDENT) {
p.bump_any();
if !opt_ident(p) {
p.bump_any();
}
}
// (
// { column_name | ( expression ) }
Expand Down Expand Up @@ -8435,7 +8440,7 @@ fn create_event_trigger(p: &mut Parser<'_>) -> CompletedMarker {
p.expect(TRIGGER_KW);
name(p);
p.expect(ON_KW);
p.expect(IDENT);
ident(p);
if p.eat(WHEN_KW) {
event_trigger_when(p);
while !p.at(EOF) && p.eat(AND_KW) {
Expand Down Expand Up @@ -8860,7 +8865,7 @@ fn create_policy(p: &mut Parser<'_>) -> CompletedMarker {
p.expect(ON_KW);
path_name_ref(p);
if p.eat(AS_KW) {
p.expect(IDENT);
ident(p);
}
if p.eat(FOR_KW) {
let _ = p.eat(ALL_KW)
Expand Down Expand Up @@ -10148,7 +10153,7 @@ fn explain_option(p: &mut Parser<'_>) {
return;
}
// { TEXT | XML | JSON | YAML }
if p.eat(TEXT_KW) || p.eat(XML_KW) || p.eat(JSON_KW) || p.eat(IDENT) {
if p.eat(TEXT_KW) || p.eat(XML_KW) || p.eat(JSON_KW) || opt_ident(p) {
return;
}
}
Expand Down Expand Up @@ -10954,7 +10959,7 @@ fn set_role(p: &mut Parser<'_>) -> CompletedMarker {
p.bump(SET_KW);
let _ = p.eat(SESSION_KW) || p.eat(LOCAL_KW);
p.expect(ROLE_KW);
if !p.eat(NONE_KW) && !p.eat(IDENT) && opt_string_literal(p).is_none() {
if !p.eat(NONE_KW) && !opt_ident(p) && opt_string_literal(p).is_none() {
p.error("expected NONE or role_name");
}
}
Expand Down Expand Up @@ -12161,7 +12166,7 @@ fn insert(p: &mut Parser<'_>, m: Option<Marker>) -> CompletedMarker {
}
opt_collate(p);
// [ opclass ]
p.eat(IDENT);
opt_ident(p);
// [, ...]
if !p.eat(COMMA) {
break;
Expand Down Expand Up @@ -12336,11 +12341,7 @@ fn delete(p: &mut Parser<'_>, m: Option<Marker>) -> CompletedMarker {
p.bump(DELETE_KW);
p.expect(FROM_KW);
relation_name(p);
if p.eat(AS_KW) {
p.expect(IDENT);
} else {
p.eat(IDENT);
}
opt_as_alias(p);
{
let m = p.start();
if p.eat(USING_KW) {
Expand Down Expand Up @@ -12371,7 +12372,7 @@ fn opt_where_current_of(p: &mut Parser<'_>) {
if p.eat(WHERE_KW) {
if p.eat(CURRENT_KW) {
p.expect(OF_KW);
p.expect(IDENT);
ident(p);
}
}
}
Expand Down Expand Up @@ -12746,7 +12747,9 @@ fn opt_function_option(p: &mut Parser<'_>) -> bool {
// string for language is deprecated but let's support it
if opt_string_literal(p).is_none() {
if p.at_ts(UNRESERVED_KEYWORDS) || p.at(IDENT) {
p.bump_any();
if !opt_ident(p) {
p.bump_any();
}
} else {
p.error(format!("expected a language name, got {:?}", p.current()));
}
Expand Down Expand Up @@ -12817,7 +12820,7 @@ fn opt_function_option(p: &mut Parser<'_>) -> bool {
// PARALLEL { UNSAFE | RESTRICTED | SAFE }
PARALLEL_KW => {
p.bump(PARALLEL_KW);
p.expect(IDENT);
ident(p);
PARALLEL_FUNC_OPTION
}
// COST execution_cost
Expand Down Expand Up @@ -13157,17 +13160,34 @@ fn create_extension(p: &mut Parser<'_>) -> CompletedMarker {
name(p);
p.eat(WITH_KW);
if p.eat(SCHEMA_KW) {
p.expect(IDENT);
name_ref(p);
}
if p.eat(VERSION_KW) {
if opt_string_literal(p).is_none() && !p.eat(IDENT) {
if opt_string_literal(p).is_none() && !opt_ident(p) {
p.error("expected string literal or IDENT");
}
}
p.eat(CASCADE_KW);
m.complete(p, CREATE_EXTENSION)
}

fn opt_ident(p: &mut Parser<'_>) -> bool {
if p.eat(IDENT) {
if p.eat(UESCAPE_KW) {
p.expect(STRING);
}
true
} else {
false
}
}

fn ident(p: &mut Parser<'_>) {
if !opt_ident(p) {
p.error("expected identifier");
}
}

// { value | 'value' | DEFAULT }
// where value can be specified as string constants, identifiers, numbers, or
// comma-separated lists of these
Expand All @@ -13180,7 +13200,7 @@ fn config_value(p: &mut Parser<'_>) -> bool {
while !p.at(EOF) {
if opt_string_literal(p).is_none()
&& opt_numeric_literal(p).is_none()
&& !p.eat(IDENT)
&& !opt_ident(p)
&& !opt_bool_literal(p)
{
if p.at_ts(BARE_LABEL_KEYWORDS) {
Expand Down Expand Up @@ -13838,11 +13858,7 @@ fn opt_alter_table_action(p: &mut Parser<'_>) -> Option<SyntaxKind> {
fn opt_col_label(p: &mut Parser<'_>) -> bool {
if p.at_ts(COL_LABEL_FIRST) {
let m = p.start();
if p.eat(IDENT) {
if p.eat(UESCAPE_KW) {
p.expect(STRING);
}
} else {
if !opt_ident(p) {
p.bump_any();
}
m.complete(p, NAME);
Expand Down
10 changes: 10 additions & 0 deletions crates/squawk_parser/tests/data/err/create_table.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,3 +45,13 @@ create unlogged table t (
cycle
)
);

create table z (
a int
)
-- ^ missing semi

with t as (
select 1
)
select * from t;
4 changes: 4 additions & 0 deletions crates/squawk_parser/tests/data/ok/select.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -498,6 +498,10 @@ WHERE ts >= DATE '2023-12-21' AND ts < DATE '2023-12-22'
GROUP BY sensor_id, DATE_TRUNC('day', ts)
ORDER BY sensor_id, day;

-- select with uescape;
select U&"d!0061t!+000061" UESCAPE '!';
SELECT U&' \' UESCAPE '!';

-- select_from_user_table
select * from user;

Expand Down
9 changes: 6 additions & 3 deletions crates/squawk_parser/tests/snapshots/tests__create_ext_ok.snap
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,8 @@ SOURCE_FILE
WHITESPACE " "
SCHEMA_KW "schema"
WHITESPACE " "
IDENT "addons"
NAME_REF
IDENT "addons"
SEMICOLON ";"
WHITESPACE "\n\n"
CREATE_EXTENSION
Expand All @@ -55,7 +56,8 @@ SOURCE_FILE
WHITESPACE " "
SCHEMA_KW "schema"
WHITESPACE " "
IDENT "bar"
NAME_REF
IDENT "bar"
WHITESPACE "\n "
VERSION_KW "version"
WHITESPACE " "
Expand All @@ -74,7 +76,8 @@ SOURCE_FILE
WHITESPACE "\n "
SCHEMA_KW "schema"
WHITESPACE " "
IDENT "bar"
NAME_REF
IDENT "bar"
WHITESPACE "\n "
VERSION_KW "version"
WHITESPACE " "
Expand Down
66 changes: 66 additions & 0 deletions crates/squawk_parser/tests/snapshots/tests__create_table_err.snap
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -406,8 +406,74 @@ SOURCE_FILE
WHITESPACE "\n"
R_PAREN ")"
SEMICOLON ";"
WHITESPACE "\n\n"
CREATE_TABLE
CREATE_KW "create"
WHITESPACE " "
TABLE_KW "table"
WHITESPACE " "
PATH
PATH_SEGMENT
NAME
IDENT "z"
WHITESPACE " "
TABLE_ARG_LIST
L_PAREN "("
WHITESPACE "\n "
COLUMN
NAME
IDENT "a"
WHITESPACE " "
PATH_TYPE
PATH
PATH_SEGMENT
NAME_REF
INT_KW "int"
WHITESPACE "\n"
R_PAREN ")"
WHITESPACE "\n"
COMMENT "-- ^ missing semi"
WHITESPACE "\n\n"
SELECT
WITH_CLAUSE
WITH_KW "with"
WHITESPACE " "
WITH_TABLE
NAME
IDENT "t"
WHITESPACE " "
AS_KW "as"
WHITESPACE " "
L_PAREN "("
WHITESPACE "\n "
SELECT
SELECT_CLAUSE
SELECT_KW "select"
WHITESPACE " "
TARGET_LIST
TARGET
LITERAL
INT_NUMBER "1"
WHITESPACE "\n"
R_PAREN ")"
WHITESPACE "\n"
SELECT_CLAUSE
SELECT_KW "select"
WHITESPACE " "
TARGET_LIST
TARGET
STAR "*"
WHITESPACE " "
FROM_CLAUSE
FROM_KW "from"
WHITESPACE " "
FROM_ITEM
NAME_REF
IDENT "t"
SEMICOLON ";"
WHITESPACE "\n"
---
ERROR@39: expected path name
ERROR@143: unexpected trailing comma
ERROR@201: unexpected trailing comma
ERROR@947: expected SEMICOLON
Loading
Loading