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.

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

Store raw blob data in value #1774

Merged
merged 5 commits into from
Jul 7, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
7 changes: 7 additions & 0 deletions src/c_api/value.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -231,6 +231,13 @@ char* kuzu_value_get_string(kuzu_value* value) {
return c_string;
}

uint8_t* kuzu_value_get_blob(kuzu_value* value) {
auto string_val = static_cast<Value*>(value->_value)->getValue<std::string>();
auto* c_blob = (uint8_t*)malloc(string_val.size() + 1);
strcpy((char*)c_blob, string_val.c_str());
return c_blob;
}

char* kuzu_value_to_string(kuzu_value* value) {
auto string_val = static_cast<Value*>(value->_value)->toString();
auto* c_string = (char*)malloc(string_val.size() + 1);
Expand Down
13 changes: 6 additions & 7 deletions src/common/types/blob.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -66,16 +66,15 @@ uint64_t Blob::fromString(const char* str, uint64_t length, uint8_t* resultBuffe
return resultPos;
}

std::string Blob::toString(blob_t& blob) {
std::string Blob::toString(const uint8_t* value, uint64_t len) {
std::string result;
auto blobData = (uint8_t*)blob.value.getData();
for (auto i = 0u; i < blob.value.len; i++) {
if (isRegularChar(blobData[i])) {
for (auto i = 0u; i < len; i++) {
if (isRegularChar(value[i])) {
// ascii characters are rendered as-is.
result += blobData[i];
result += value[i];
} else {
auto firstByte = blobData[i] >> HexFormatConstants::NUM_BYTES_TO_SHIFT_FOR_FIRST_BYTE;
auto secondByte = blobData[i] & HexFormatConstants::SECOND_BYTE_MASK;
auto firstByte = value[i] >> HexFormatConstants::NUM_BYTES_TO_SHIFT_FOR_FIRST_BYTE;
auto secondByte = value[i] & HexFormatConstants::SECOND_BYTE_MASK;
// non-ascii characters are rendered as hexadecimal (e.g. \x00).
result += '\\';
result += 'x';
Expand Down
3 changes: 2 additions & 1 deletion src/common/types/value.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -182,7 +182,7 @@ void Value::copyValueFrom(const uint8_t* value) {
val.internalIDVal = *((nodeID_t*)value);
} break;
case LogicalTypeID::BLOB: {
strVal = Blob::toString(*(blob_t*)value);
strVal = ((blob_t*)value)->value.getAsString();
} break;
case LogicalTypeID::STRING: {
strVal = ((ku_string_t*)value)->getAsString();
Expand Down Expand Up @@ -289,6 +289,7 @@ std::string Value::toString() const {
case LogicalTypeID::INTERNAL_ID:
return TypeUtils::toString(val.internalIDVal);
case LogicalTypeID::BLOB:
return Blob::toString(reinterpret_cast<const uint8_t*>(strVal.c_str()), strVal.length());
case LogicalTypeID::STRING:
return strVal;
case LogicalTypeID::MAP: {
Expand Down
8 changes: 7 additions & 1 deletion src/include/c_api/kuzu.h
Original file line number Diff line number Diff line change
Expand Up @@ -649,7 +649,7 @@ KUZU_C_API int16_t kuzu_value_get_int16(kuzu_value* value);
*/
KUZU_C_API int32_t kuzu_value_get_int32(kuzu_value* value);
/**
* @brief Returns the int64 value of the given value. The value must be of type INT64.
* @brief Returns the int64 value of the given value. The value must be of type INT64 or SERIAL.
* @param value The value to return.
*/
KUZU_C_API int64_t kuzu_value_get_int64(kuzu_value* value);
Expand Down Expand Up @@ -688,6 +688,12 @@ KUZU_C_API kuzu_interval_t kuzu_value_get_interval(kuzu_value* value);
* @param value The value to return.
*/
KUZU_C_API char* kuzu_value_get_string(kuzu_value* value);
/**
* @brief Returns the blob value of the given value. The returned buffer is null-terminated similar
* to a string. The value must be of type BLOB.
* @param value The value to return.
*/
KUZU_C_API uint8_t* kuzu_value_get_blob(kuzu_value* value);
/**
* @brief Converts the given value to string.
* @param value The value to convert.
Expand Down
6 changes: 5 additions & 1 deletion src/include/common/types/blob.h
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,11 @@ struct HexFormatConstants {
};

struct Blob {
static std::string toString(blob_t& blob);
static std::string toString(const uint8_t* value, uint64_t len);

static inline std::string toString(blob_t& blob) {
return toString(blob.value.getData(), blob.value.len);
}

static uint64_t getBlobSize(const ku_string_t& blob);

Expand Down
3 changes: 2 additions & 1 deletion src/include/common/types/value.h
Original file line number Diff line number Diff line change
Expand Up @@ -434,7 +434,8 @@ inline internalID_t Value::getValue() const {
*/
KUZU_API template<>
inline std::string Value::getValue() const {
assert(dataType.getLogicalTypeID() == LogicalTypeID::STRING);
assert(dataType.getLogicalTypeID() == LogicalTypeID::STRING ||
dataType.getLogicalTypeID() == LogicalTypeID::BLOB);
return strVal;
}

Expand Down
22 changes: 22 additions & 0 deletions test/c_api/value_test.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -588,6 +588,28 @@ TEST_F(CApiValueTest, GetString) {
kuzu_query_result_destroy(result);
}

TEST_F(CApiValueTest, GetBlob) {
auto connection = getConnection();
auto result =
kuzu_connection_query(connection, (char*)R"(RETURN BLOB('\\xAA\\xBB\\xCD\\x1A');)");
ASSERT_TRUE(kuzu_query_result_is_success(result));
ASSERT_TRUE(kuzu_query_result_has_next(result));
auto flatTuple = kuzu_query_result_get_next(result);
auto value = kuzu_flat_tuple_get_value(flatTuple, 0);
ASSERT_TRUE(value->_is_owned_by_cpp);
ASSERT_FALSE(kuzu_value_is_null(value));
auto blob = kuzu_value_get_blob(value);
ASSERT_EQ(blob[0], 0xAA);
ASSERT_EQ(blob[1], 0xBB);
ASSERT_EQ(blob[2], 0xCD);
ASSERT_EQ(blob[3], 0x1A);
ASSERT_EQ(blob[4], 0x00);
free(blob);
kuzu_value_destroy(value);
kuzu_flat_tuple_destroy(flatTuple);
kuzu_query_result_destroy(result);
}

TEST_F(CApiValueTest, ToSting) {
auto connection = getConnection();
auto result = kuzu_connection_query(connection,
Expand Down
10 changes: 8 additions & 2 deletions tools/java_api/src/jni/kuzu_java.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -927,12 +927,18 @@ JNIEXPORT jobject JNICALL Java_com_kuzudb_KuzuNative_kuzu_1value_1get_1value(
jobject ret = env->NewObject(retClass, ctor, iid.tableID, iid.offset);
return ret;
}
case LogicalTypeID::STRING:
case LogicalTypeID::BLOB: {
case LogicalTypeID::STRING: {
std::string str = v->getValue<std::string>();
jstring ret = env->NewStringUTF(str.c_str());
return ret;
}
case LogicalTypeID::BLOB: {
auto str = v->getValue<std::string>();
auto byteBuffer = str.c_str();
auto ret = env->NewByteArray(str.size());
env->SetByteArrayRegion(ret, 0, str.size(), (jbyte*)byteBuffer);
return ret;
}
default: {
// Throw exception here?
return nullptr;
Expand Down
18 changes: 18 additions & 0 deletions tools/java_api/src/test/java/com/kuzudb/test/ValueTest.java
Original file line number Diff line number Diff line change
Expand Up @@ -529,6 +529,24 @@ void ValueGetString() throws KuzuObjectRefDestroyedException {
result.destroy();
}

@Test
void ValueGetBlob() throws KuzuObjectRefDestroyedException {
KuzuQueryResult result = conn.query("RETURN BLOB('\\\\xAA\\\\xBB\\\\xCD\\\\x1A');");
assertTrue(result.isSuccess());
assertTrue(result.hasNext());
KuzuFlatTuple flatTuple = result.getNext();
KuzuValue value = flatTuple.getValue(0);
assertTrue(value.isOwnedByCPP());
assertFalse(value.isNull());

byte[] bytes = value.getValue();
assertTrue(bytes.length == 4);
assertTrue(bytes[0] == (byte) 0xAA);
assertTrue(bytes[1] == (byte) 0xBB);
assertTrue(bytes[2] == (byte) 0xCD);
assertTrue(bytes[3] == (byte) 0x1A);
}

@Test
void ValueToString() throws KuzuObjectRefDestroyedException {
KuzuQueryResult result = conn.query("MATCH (a:person) RETURN a.fName, a.isStudent, a.workedHours");
Expand Down
7 changes: 5 additions & 2 deletions tools/nodejs_api/src_cpp/node_util.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -27,10 +27,13 @@ Napi::Value Util::ConvertToNapiObject(const Value& value, Napi::Env env) {
case LogicalTypeID::DOUBLE: {
return Napi::Number::New(env, value.getValue<double>());
}
case LogicalTypeID::STRING:
case LogicalTypeID::BLOB: {
case LogicalTypeID::STRING: {
return Napi::String::New(env, value.getValue<std::string>());
}
case LogicalTypeID::BLOB: {
auto blobVal = value.getValue<std::string>();
return Napi::Buffer<char>::Copy(env, blobVal.c_str(), blobVal.size());
}
case LogicalTypeID::DATE: {
auto dateVal = value.getValue<date_t>();
// Javascript Date type contains both date and time information. This returns the Date at
Expand Down
18 changes: 18 additions & 0 deletions tools/nodejs_api/test/test_data_type.js
Original file line number Diff line number Diff line change
Expand Up @@ -113,6 +113,24 @@ describe("STRING", function () {
});
});

describe("BLOB", function () {
it("should convert BLOB type", async function () {
const queryResult = await conn.query(
"RETURN BLOB('\\\\xAA\\\\xBB\\\\xCD\\\\x1A');"
);
const result = await queryResult.getAll();
assert.equal(result.length, 1);
assert.equal(Object.keys(result[0]).length, 1);
const value = Object.values(result[0])[0];
assert.isTrue(value instanceof Uint8Array);
assert.equal(value.length, 4);
assert.equal(value[0], 0xaa);
assert.equal(value[1], 0xbb);
assert.equal(value[2], 0xcd);
assert.equal(value[3], 0x1a);
});
});

describe("DATE", function () {
it("should convert DATE type", async function () {
const queryResult = await conn.query(
Expand Down
8 changes: 6 additions & 2 deletions tools/python_api/src_cpp/py_query_result.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -90,10 +90,14 @@ py::object PyQueryResult::convertValueToPyObject(const Value& value) {
case LogicalTypeID::DOUBLE: {
return py::cast(value.getValue<double>());
}
case LogicalTypeID::STRING:
case LogicalTypeID::BLOB:{
case LogicalTypeID::STRING: {
return py::cast(value.getValue<std::string>());
}
case LogicalTypeID::BLOB: {
auto blobStr = value.getValue<std::string>();
auto blobBytesArray = blobStr.c_str();
return py::bytes(blobBytesArray, blobStr.size());
}
case LogicalTypeID::DATE: {
auto dateVal = value.getValue<date_t>();
int32_t year, month, day;
Expand Down
15 changes: 15 additions & 0 deletions tools/python_api/test/test_datatype.py
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,21 @@ def test_string(establish_connection):
assert not result.has_next()
result.close()


def test_blob(establish_connection):
conn, db = establish_connection
result = conn.execute("RETURN BLOB('\\\\xAA\\\\xBB\\\\xCD\\\\x1A')")
assert result.has_next()
result_blob = result.get_next()[0]
assert len(result_blob) == 4
assert result_blob[0] == 0xAA
assert result_blob[1] == 0xBB
assert result_blob[2] == 0xCD
assert result_blob[3] == 0x1A
assert not result.has_next()
result.close()


def test_date(establish_connection):
conn, db = establish_connection
result = conn.execute(
Expand Down
Loading