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

Jump to bottom

[bug](function)fix json_replace check return type error #37014

Merged
merged 3 commits into from
Jul 5, 2024
Merged

[bug](function)fix json_replace check return type error #37014

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4286df2
[bug](function)fix json_replace check return type error
zhangstar333 Jun 28, 2024
0ca062b
update review2
zhangstar333 Jul 1, 2024
c9e6208
update2
zhangstar333 Jul 1, 2024
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
75 changes: 46 additions & 29 deletions be/src/vec/functions/function_json.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1275,11 +1275,13 @@ class FunctionJsonModifyImpl : public IFunction {

Status get_parsed_path_columns(std::vector<std::vector<std::vector<JsonPath>>>& json_paths,
const std::vector<const ColumnString*>& data_columns,
size_t input_rows_count) const {
size_t input_rows_count,
std::vector<bool>& column_is_consts) const {
for (auto col = 1; col + 1 < data_columns.size() - 1; col += 2) {
json_paths.emplace_back(std::vector<std::vector<JsonPath>>());
for (auto row = 0; row < input_rows_count; row++) {
const auto path = data_columns[col]->get_data_at(row);
const auto path = data_columns[col]->get_data_at(
index_check_const(row, column_is_consts[col]));
std::string_view path_string(path.data, path.size);
std::vector<JsonPath> parsed_paths;

Expand Down Expand Up @@ -1313,7 +1315,7 @@ class FunctionJsonModifyImpl : public IFunction {
DataTypePtr get_return_type_impl(const DataTypes& arguments) const override {
bool is_nullable = false;
// arguments: (json_str, path, val[, path, val...], type_flag)
for (auto col = 2; col < arguments.size() - 1; col += 2) {
for (auto col = 0; col < arguments.size() - 1; col += 1) {
if (arguments[col]->is_nullable()) {
is_nullable = true;
break;
Expand All @@ -1327,36 +1329,42 @@ class FunctionJsonModifyImpl : public IFunction {
size_t result, size_t input_rows_count) const override {
auto result_column = ColumnString::create();
bool is_nullable = false;
auto ret_null_map = ColumnUInt8::create(0, 0);
ColumnUInt8::MutablePtr ret_null_map = nullptr;
ColumnUInt8::Container* ret_null_map_data = nullptr;

std::vector<ColumnPtr> column_ptrs; // prevent converted column destruct
std::vector<const ColumnString*> data_columns;
std::vector<const ColumnUInt8*> nullmaps;
std::vector<bool> column_is_consts;
for (int i = 0; i < arguments.size(); i++) {
auto column = block.get_by_position(arguments[i]).column;
column_ptrs.push_back(column->convert_to_full_column_if_const());
const ColumnNullable* col_nullable =
check_and_get_column<ColumnNullable>(column_ptrs.back().get());
ColumnPtr arg_col;
bool arg_const;
std::tie(arg_col, arg_const) =
unpack_if_const(block.get_by_position(arguments[i]).column);
const auto* col_nullable = check_and_get_column<ColumnNullable>(arg_col.get());
column_is_consts.push_back(arg_const);
if (col_nullable) {
if (!is_nullable) {
is_nullable = true;
ret_null_map = ColumnUInt8::create(input_rows_count, 0);
}
const ColumnUInt8* col_nullmap = check_and_get_column<ColumnUInt8>(
const ColumnUInt8* col_nullmap = assert_cast<const ColumnUInt8*>(
col_nullable->get_null_map_column_ptr().get());
nullmaps.push_back(col_nullmap);
const ColumnString* col = check_and_get_column<ColumnString>(
const ColumnString* col = assert_cast<const ColumnString*>(
col_nullable->get_nested_column_ptr().get());
data_columns.push_back(col);
} else {
nullmaps.push_back(nullptr);
data_columns.push_back(assert_cast<const ColumnString*>(column_ptrs.back().get()));
data_columns.push_back(assert_cast<const ColumnString*>(arg_col.get()));
}
}

//*assert_cast<ColumnUInt8*>(ret_null_map.get())
if (is_nullable) {
ret_null_map = ColumnUInt8::create(input_rows_count, 0);
ret_null_map_data = &(ret_null_map->get_data());
}
RETURN_IF_ERROR(execute_process(
data_columns, *assert_cast<ColumnString*>(result_column.get()), input_rows_count,
nullmaps, is_nullable, *assert_cast<ColumnUInt8*>(ret_null_map.get())));
nullmaps, is_nullable, ret_null_map_data, column_is_consts));

if (is_nullable) {
block.replace_by_position(result, ColumnNullable::create(std::move(result_column),
zhangstar333 marked this conversation as resolved.
Show resolved Hide resolved
Expand All @@ -1370,13 +1378,15 @@ class FunctionJsonModifyImpl : public IFunction {
Status execute_process(const std::vector<const ColumnString*>& data_columns,
ColumnString& result_column, size_t input_rows_count,
const std::vector<const ColumnUInt8*> nullmaps, bool is_nullable,
ColumnUInt8& ret_null_map) const {
ColumnUInt8::Container* ret_null_map_data,
std::vector<bool>& column_is_consts) const {
std::string type_flags = data_columns.back()->get_data_at(0).to_string();

std::vector<rapidjson::Document> objects;
for (auto row = 0; row < input_rows_count; row++) {
objects.emplace_back(rapidjson::kNullType);
const auto json_doc = data_columns[0]->get_data_at(row);
const auto json_doc =
data_columns[0]->get_data_at(index_check_const(row, column_is_consts[0]));
std::string_view json_str(json_doc.data, json_doc.size);
objects[row].Parse(json_str.data(), json_str.size());
if (UNLIKELY(objects[row].HasParseError())) {
Expand All @@ -1386,9 +1396,10 @@ class FunctionJsonModifyImpl : public IFunction {
}

std::vector<std::vector<std::vector<JsonPath>>> json_paths;
RETURN_IF_ERROR(get_parsed_path_columns(json_paths, data_columns, input_rows_count));
RETURN_IF_ERROR(get_parsed_path_columns(json_paths, data_columns, input_rows_count,
column_is_consts));

execute_parse(type_flags, data_columns, objects, json_paths, nullmaps);
execute_parse(type_flags, data_columns, objects, json_paths, nullmaps, column_is_consts);

rapidjson::StringBuffer buf;
rapidjson::Writer<rapidjson::StringBuffer> writer(buf);
Expand All @@ -1398,7 +1409,7 @@ class FunctionJsonModifyImpl : public IFunction {
writer.Reset(buf);
objects[i].Accept(writer);
if (is_nullable && objects[i].IsNull()) {
ret_null_map.get_data()[i] = 1;
(*ret_null_map_data)[i] = 1;
}
result_column.insert_data(buf.GetString(), buf.GetSize());
}
Expand All @@ -1412,11 +1423,12 @@ class FunctionJsonModifyImpl : public IFunction {
const std::vector<const ColumnString*>& data_columns,
std::vector<rapidjson::Document>& objects,
std::vector<std::vector<std::vector<JsonPath>>>& json_paths,
const std::vector<const ColumnUInt8*>& nullmaps) {
const std::vector<const ColumnUInt8*>& nullmaps,
std::vector<bool>& column_is_consts) {
for (auto col = 1; col + 1 < data_columns.size() - 1; col += 2) {
constexpr_int_match<'0', '6', Reducer>::run(type_flags[col + 1], objects,
json_paths[col / 2], data_columns[col + 1],
nullmaps[col + 1]);
constexpr_int_match<'0', '6', Reducer>::run(
type_flags[col + 1], objects, json_paths[col / 2], data_columns[col + 1],
nullmaps[col + 1], column_is_consts[col + 1]);
}
}

Expand Down Expand Up @@ -1498,18 +1510,23 @@ class FunctionJsonModifyImpl : public IFunction {
template <typename TypeImpl>
static void execute_type(std::vector<rapidjson::Document>& objects,
std::vector<std::vector<JsonPath>>& paths_column,
const ColumnString* value_column, const ColumnUInt8* nullmap) {
const ColumnString* value_column, const ColumnUInt8* nullmap,
bool column_is_const) {
StringParser::ParseResult result;
rapidjson::Value value;
for (auto row = 0; row < objects.size(); row++) {
std::vector<JsonPath>* parsed_paths = &paths_column[row];

if (nullmap != nullptr && nullmap->get_data()[row]) {
JsonParser<'0'>::update_value(result, value, value_column->get_data_at(row),
objects[row].GetAllocator());
JsonParser<'0'>::update_value(
result, value,
value_column->get_data_at(index_check_const(row, column_is_const)),
objects[row].GetAllocator());
} else {
TypeImpl::update_value(result, value, value_column->get_data_at(row),
objects[row].GetAllocator());
TypeImpl::update_value(
result, value,
value_column->get_data_at(index_check_const(row, column_is_const)),
objects[row].GetAllocator());
}

switch (Kind::modify_type) {
Expand Down
5 changes: 5 additions & 0 deletions regression-test/data/query_p0/sql_functions/json_function/test_query_json_replace.out
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,3 +11,8 @@ null
{"id":3,"time":null,"a1":[1,9],"a2":[1,2]}
{"id":4,"time":null,"a1":[1,null],"a2":[1,2]}

-- !sql2 --
{"id":1,"time":"2022-01-01 11:45:14","a1":[1,9],"a2":[1,2]}
{"id":2,"time":"2022-01-01 11:45:14","a1":[1,null],"a2":[1,2]}
{"id":3,"time":null,"a1":[1,9],"a2":[1,2]}

23 changes: 23 additions & 0 deletions regression-test/suites/query_p0/sql_functions/json_function/test_query_json_replace.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,4 +42,27 @@ suite("test_query_json_replace", "query") {
sql "insert into ${tableName} values(4,null,null);"
qt_sql1 "select json_replace('{\"id\": 0, \"time\": \"1970-01-01 00:00:00\", \"a1\": [1, 2], \"a2\": [1, 2]}', '\$.id', id, '\$.time', time, '\$.a1[1]', k, '\$.a2[3]', k) from ${tableName} order by id;"
sql "DROP TABLE ${tableName};"

sql "DROP TABLE IF EXISTS test_query_json_replace_nullable"
sql """
CREATE TABLE test_query_json_replace_nullable (
`id` int(11) null,
`time` datetime,
`k` int(11)
) ENGINE=OLAP
DUPLICATE KEY(`id`,`time`,`k`)
COMMENT "OLAP"
DISTRIBUTED BY HASH(`id`) BUCKETS 1
PROPERTIES (
"replication_allocation" = "tag.location.default: 1",
"in_memory" = "false",
"storage_format" = "V2"
);
"""
sql "insert into test_query_json_replace_nullable values(1,'2022-01-01 11:45:14',9);"
sql "insert into test_query_json_replace_nullable values(2,'2022-01-01 11:45:14',null);"
sql "insert into test_query_json_replace_nullable values(3,null,9);"
qt_sql2 "select json_replace('{\"id\": 0, \"time\": \"1970-01-01 00:00:00\", \"a1\": [1, 2], \"a2\": [1, 2]}', '\$.id', id, '\$.time', time, '\$.a1[1]', k, '\$.a2[3]', k) from test_query_json_replace_nullable order by id;"

sql "DROP TABLE test_query_json_replace_nullable;"
}
Loading