Skip to content

feat(query): support Latin-1 binary input format #20238

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

Draft
dbsid wants to merge 1 commit into
base: main
Choose a base branch
from
+175 −1
Draft

feat(query): support Latin-1 binary input format #20238

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
27 changes: 26 additions & 1 deletion src/query/functions/src/scalars/binary.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,12 +172,16 @@ pub fn register(registry: &mut FunctionRegistry) {
match format.to_ascii_lowercase().as_str() {
"hex" => eval_unhex(val, ctx),
"base64" => eval_from_base64(val, ctx),
"latin1" | "latin-1" | "iso-8859-1" => eval_latin1_bytes(val, ctx),
"utf-8" => match val {
Value::Scalar(val) => Value::Scalar(val.as_bytes().to_vec()),
Value::Column(col) => Value::Column(col.into()),
},
_ => {
ctx.set_error(0, "The format option only supports hex, base64, and utf-8");
ctx.set_error(
0,
"The format option only supports hex, base64, latin1, and utf-8",
);
Value::Scalar(Vec::new())
}
}
Expand Down Expand Up @@ -206,6 +210,7 @@ pub fn register(registry: &mut FunctionRegistry) {
match format.to_ascii_lowercase().as_str() {
"hex" => error_to_null(eval_unhex)(val, ctx),
"base64" => error_to_null(eval_from_base64)(val, ctx),
"latin1" | "latin-1" | "iso-8859-1" => error_to_null(eval_latin1_bytes)(val, ctx),
"utf-8" => match val {
Value::Scalar(val) => Value::Scalar(Some(val.as_bytes().to_vec())),
Value::Column(col) => {
Expand Down Expand Up @@ -347,6 +352,26 @@ fn eval_utf8_bytes_nullable(val: Value<StringType>) -> Value<NullableType<Binary
}
}

fn eval_latin1_bytes(val: Value<StringType>, ctx: &mut EvalContext) -> Value<BinaryType> {
vectorize_string_to_binary(
|col| col.total_bytes_len(),
|val, output, ctx| {
for ch in val.chars() {
let code_point = u32::from(ch);
if code_point > u32::from(u8::MAX) {
ctx.set_error(
output.len(),
format!("character U+{code_point:04X} cannot be encoded as latin1"),
);
break;
}
output.data.push(code_point as u8);
}
output.commit_row();
},
)(val, ctx)
}

fn eval_unhex(val: Value<StringType>, ctx: &mut EvalContext) -> Value<BinaryType> {
vectorize_string_to_binary(
|col| col.total_bytes_len() / 2,
Expand Down
23 changes: 23 additions & 0 deletions src/query/functions/tests/it/scalars/binary.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,6 +110,29 @@ fn test_to_jsonb_binary(file: &mut impl Write) {
fn test_to_binary(file: &mut impl Write, is_try: bool) {
let prefix = if is_try { "TRY_" } else { "" };

run_ast(
file,
format!("to_hex({prefix}to_binary(char(0, 65, 127, 128, 255), 'latin1'))"),
&[],
);
run_ast(
file,
format!("to_hex({prefix}to_binary(char(128, 255), 'latin-1'))"),
&[],
);
run_ast(
file,
format!("to_hex({prefix}to_binary(char(128, 255), 'iso-8859-1'))"),
&[],
);
run_ast(file, format!("{prefix}to_binary(char(256), 'latin1')"), &[]);
run_ast(file, format!("to_hex({prefix}to_binary(s, 'latin1'))"), &[
(
"s",
StringType::from_data(vec!["\u{80}\u{ff}", "\u{100}", "plain"]),
),
]);

run_ast(file, format!("{prefix}to_binary(to_bitmap('1,2,3'))"), &[]);
run_ast(
file,
Expand Down
101 changes: 101 additions & 0 deletions src/query/functions/tests/it/scalars/testdata/binary.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -248,6 +248,49 @@ evaluation (internal):
+--------+--------------------------------------------------------+


ast : to_hex(to_binary(char(0, 65, 127, 128, 255), 'latin1'))
raw expr : to_hex(to_binary(char(0, 65, 127, 128, 255), 'latin1'))
checked expr : to_hex<Binary>(to_binary<String, String>(char<Int64, Int64, Int64, Int64, Int64>(CAST<UInt8>(0_u8 AS Int64), CAST<UInt8>(65_u8 AS Int64), CAST<UInt8>(127_u8 AS Int64), CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "latin1"))
optimized expr : "00417f80ff"
output type : String
output domain : {"00417f80ff"..="00417f80ff"}
output : '00417f80ff'


ast : to_hex(to_binary(char(128, 255), 'latin-1'))
raw expr : to_hex(to_binary(char(128, 255), 'latin-1'))
checked expr : to_hex<Binary>(to_binary<String, String>(char<Int64, Int64>(CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "latin-1"))
optimized expr : "80ff"
output type : String
output domain : {"80ff"..="80ff"}
output : '80ff'


ast : to_hex(to_binary(char(128, 255), 'iso-8859-1'))
raw expr : to_hex(to_binary(char(128, 255), 'iso-8859-1'))
checked expr : to_hex<Binary>(to_binary<String, String>(char<Int64, Int64>(CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "iso-8859-1"))
optimized expr : "80ff"
output type : String
output domain : {"80ff"..="80ff"}
output : '80ff'


error:
--> SQL:1:1
|
1 | to_binary(char(256), 'latin1')
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ character U+0100 cannot be encoded as latin1 while evaluating function `to_binary('Ā', 'latin1')` in expr `to_binary(char(CAST(256 AS Int64)), 'latin1')`



error:
--> SQL:1:8
|
1 | to_hex(to_binary(s, 'latin1'))
| ^^^^^^^^^^^^^^^^^^^^^^ character U+0100 cannot be encoded as latin1 while evaluating function `to_binary('Ā', 'latin1')` in expr `to_binary(s, 'latin1')`, during run expr: `to_hex(to_binary(s, 'latin1'))`



ast : to_binary(to_bitmap('1,2,3'))
raw expr : to_binary(to_bitmap('1,2,3'))
checked expr : CAST<Bitmap>(CAST<String>("1,2,3" AS Bitmap) AS Binary)
Expand Down Expand Up @@ -381,6 +424,64 @@ evaluation (internal):
+--------+--------------------------------------------------------+


ast : to_hex(TRY_to_binary(char(0, 65, 127, 128, 255), 'latin1'))
raw expr : to_hex(TRY_to_binary(char(0, 65, 127, 128, 255), 'latin1'))
checked expr : to_hex<Binary NULL>(try_to_binary<String, String>(char<Int64, Int64, Int64, Int64, Int64>(CAST<UInt8>(0_u8 AS Int64), CAST<UInt8>(65_u8 AS Int64), CAST<UInt8>(127_u8 AS Int64), CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "latin1"))
optimized expr : "00417f80ff"
output type : String NULL
output domain : {"00417f80ff"..="00417f80ff"}
output : '00417f80ff'


ast : to_hex(TRY_to_binary(char(128, 255), 'latin-1'))
raw expr : to_hex(TRY_to_binary(char(128, 255), 'latin-1'))
checked expr : to_hex<Binary NULL>(try_to_binary<String, String>(char<Int64, Int64>(CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "latin-1"))
optimized expr : "80ff"
output type : String NULL
output domain : {"80ff"..="80ff"}
output : '80ff'


ast : to_hex(TRY_to_binary(char(128, 255), 'iso-8859-1'))
raw expr : to_hex(TRY_to_binary(char(128, 255), 'iso-8859-1'))
checked expr : to_hex<Binary NULL>(try_to_binary<String, String>(char<Int64, Int64>(CAST<UInt8>(128_u8 AS Int64), CAST<UInt8>(255_u8 AS Int64)), "iso-8859-1"))
optimized expr : "80ff"
output type : String NULL
output domain : {"80ff"..="80ff"}
output : '80ff'


ast : TRY_to_binary(char(256), 'latin1')
raw expr : TRY_to_binary(char(256), 'latin1')
checked expr : try_to_binary<String, String>(char<Int64>(CAST<UInt16>(256_u16 AS Int64)), "latin1")
optimized expr : NULL
output type : Binary NULL
output domain : {NULL}
output : NULL


ast : to_hex(TRY_to_binary(s, 'latin1'))
raw expr : to_hex(TRY_to_binary(s::String, 'latin1'))
checked expr : to_hex<Binary NULL>(try_to_binary<String, String>(s, "latin1"))
evaluation:
+--------+-----------------+-----------------+
| | s | Output |
+--------+-----------------+-----------------+
| Type | String | String NULL |
| Domain | {"plain"..="Ā"} | {""..} ∪ {NULL} |
| Row 0 | '€ÿ' | '80ff' |
| Row 1 | 'Ā' | NULL |
| Row 2 | 'plain' | '706c61696e' |
+--------+-----------------+-----------------+
evaluation (internal):
+--------+-------------------------------------------------------------------------------------+
| Column | Data |
+--------+-------------------------------------------------------------------------------------+
| s | Column(StringColumn[€ÿ, Ā, plain]) |
| Output | NullableColumn { column: StringColumn[80ff, , 706c61696e], validity: [0b_____101] } |
+--------+-------------------------------------------------------------------------------------+


ast : TRY_to_binary(to_bitmap('1,2,3'))
raw expr : TRY_to_binary(to_bitmap('1,2,3'))
checked expr : TRY_CAST<Bitmap>(CAST<String>("1,2,3" AS Bitmap) AS Binary NULL)
Expand Down
25 changes: 25 additions & 0 deletions tests/sqllogictests/suites/base/03_common/03_0041_insert_into_binary.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,31 @@ SELECT id, v FROM t1 order by id
6 616161
7 616161

query T
SELECT to_hex(to_binary(char(0, 65, 127, 128, 255), 'latin1'))
----
00417f80ff

query T
SELECT to_hex(to_binary(as_string(get(parse_json('{"v":"\u0000A\u007f\u0080\u00ff"}'), 'v')), 'latin1'))
----
00417f80ff

query T
SELECT to_hex(try_to_binary(char(128, 255), 'latin-1'))
----
80ff

query T
SELECT to_hex(try_to_binary(char(128, 255), 'iso-8859-1'))
----
80ff

query ?
SELECT try_to_binary(char(256), 'latin1')
----
NULL

statement ok
ALTER TABLE t1 MODIFY COLUMN v string

Expand Down
Loading