mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-24 15:54:37 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
mariadb/sql/item_jsonfunc.h

944 lines
28 KiB
C
Raw Permalink Normal View History

MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
#ifndef ITEM_JSONFUNC_INCLUDED
#define ITEM_JSONFUNC_INCLUDED
MDEV-22640 fixup: clang -Winconsistent-missing-override
2021-06-29 15:02:10 +03:00
/* Copyright (c) 2016, 2021, MariaDB
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */
/* This file defines all JSON functions */
#include <json_lib.h>
#include "item_cmpfunc.h" // Item_bool_func
#include "item_strfunc.h" // Item_str_func
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
#include "item_sum.h"
MDEV-27018 IF and COALESCE lose "json" property Hybrid functions (IF, COALESCE, etc) did not preserve the JSON property from their arguments. The same problem was repeatable for single row subselects. The problem happened because the method Item::is_json_type() was inconsistently implemented across the Item hierarchy. For example, Item_hybrid_func and Item_singlerow_subselect did not override is_json_type(). Solution: - Removing Item::is_json_type() - Implementing specific JSON type handlers: Type_handler_string_json Type_handler_varchar_json Type_handler_tiny_blob_json Type_handler_blob_json Type_handler_medium_blob_json Type_handler_long_blob_json - Reusing the existing data type infrastructure to pass JSON type handlers across all item types, including classes Item_hybrid_func and Item_singlerow_subselect. Note, these two classes themselves do not need any changes! - Extending the data type infrastructure so data types can inherit their properties (e.g. aggregation rules) from their base data types. E.g. VARCHAR/JSON acts as VARCHAR, LONGTEXT/JSON acts as LONGTEXT when mixed to a non-JSON data type. This is done by: - adding virtual method Type_handler::type_handler_base() - adding a helper class Type_handler_pair - refactoring Type_handler_hybrid_field_type methods aggregate_for_result(), aggregate_for_min_max(), aggregate_for_num_op() to use Type_handler_pair. This change also fixes: MDEV-27361 Hybrid functions with JSON arguments do not send format metadata Also, adding mtr tests for JSON replication. It was not covered yet. And the current patch changes the replication code slightly.
2022-01-10 18:05:55 +04:00
#include "sql_type_json.h"
MDEV-27128: Implement JSON Schema Validation FUNCTION Implementation: Implementation is made according to json schema validation draft 2020 JSON schema basically has same structure as that of json object, consisting of key-value pairs. So it can be parsed in the same manner as any json object. However, none of the keywords are mandatory, so making guess about the json value type based only on the keywords would be incorrect. Hence we need separate objects denoting each keyword. So during create_object_and_handle_keyword() we create appropriate objects based on the keywords and validate each of them individually on the json document by calling respective validate() function if the type matches. If any of them fails, return false, else return true.
2022-10-28 13:03:13 +05:30
#include "json_schema.h"
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
class json_path_with_flags
{
public:
json_path_t p;
bool constant;
bool parsed;
json_path_step_t *cur_step;
void set_constant_flag(bool s_constant)
{
constant= s_constant;
parsed= FALSE;
}
};
MDEV-17399 Add support for JSON_TABLE. The specific table handler for the table functions was introduced, and used to implement JSON_TABLE.
2021-03-17 09:03:45 +04:00
void report_path_error_ex(const char *ps, json_path_t *p,
const char *fname, int n_param,
Sql_condition::enum_warning_level lv);
void report_json_error_ex(const char *js, json_engine_t *je,
const char *fname, int n_param,
Sql_condition::enum_warning_level lv);
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Json_engine_scan: public json_engine_t
{
public:
Json_engine_scan(CHARSET_INFO *i_cs, const uchar *str, const uchar *end)
{
json_scan_start(this, i_cs, str, end);
}
Json_engine_scan(const String &str)
:Json_engine_scan(str.charset(), (const uchar *) str.ptr(),
(const uchar *) str.end())
{ }
bool check_and_get_value_scalar(String *res, int *error);
MDEV-30145: JSON_TABLE: allow to retrieve the key when iterating on JSON objects Idea behind implementation: We get the json object specified by the json path. Then, transform it into key-value pairs by going over the json. Get each key-value pair one-by-one and return the result.
2023-05-25 15:45:43 +05:30
bool check_and_get_value_complex(String *res, int *error,
json_value_types cur_value_type=
JSON_VALUE_UNINITIALIZED);
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
};
class Json_path_extractor: public json_path_with_flags
{
protected:
String tmp_js, tmp_path;
virtual ~Json_path_extractor() { }
virtual bool check_and_get_value(Json_engine_scan *je,
String *to, int *error)=0;
bool extract(String *to, Item *js, Item *jp, CHARSET_INFO *cs);
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_valid: public Item_bool_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
String tmp_value;
public:
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_func_json_valid(THD *thd, Item *json) : Item_bool_func(thd, json) {}
MDEV-34123 CONCAT Function Returns Unexpected Empty Set in Query Search conditions were evaluated using val_int(), which was wrong. Fixing the code to use val_bool() instead. Details: - Adding a new item_base_t::IS_COND flag which marks Items used as <search condition> in WHERE, HAVING, JOIN ON, CASE WHEN clauses. The flag is at the parse time. These expressions must be evaluated using val_bool() rather than val_int(). Note, the optimizer creates more Items which are used as search conditions. Most of these items are not marked with IS_COND yet. This is OK for now, but eventually these Items can also be fixed to have the flag. - Adding a method Item::is_cond() which tests if the Item has the IS_COND flag. - Implementing Item_cache_bool. It evaluates the cached expression using val_bool() rather than val_int(). Overriding Type_handler_bool::Item_get_cache() to create Item_cache_bool. - Implementing Item::save_bool_in_field(). It uses val_bool() rather than val_int() to evaluate the expression. - Implementing Type_handler_bool::Item_save_in_field() using Item::save_bool_in_field(). - Fixing all Item_bool_func descendants to implement a virtual val_bool() rather than a virtual val_int(). - To find places where val_int() should be fixed to val_bool(), a few DBUG_ASSERT(!is_cond()) where added into val_int() implementations of selected (most frequent) classes: Item_field Item_str_func Item_datefunc Item_timefunc Item_datetimefunc Item_cache_bool Item_bool_func Item_func_hybrid_field_type Item_basic_constant descendants - Fixing all places where DBUG_ASSERT() happened during an "mtr" run to use val_bool() instead of val_int().
2024-05-14 09:19:34 +04:00
bool val_bool() override;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_valid") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
if (Item_bool_func::fix_length_and_dec(thd))
MDEV-11071: Assertion `thd->transaction.stmt.is_empty()' failed in Locked_tables_list::unlock_locked_table fix_length_and_dec now return result (error/OK)
2018-05-08 15:26:26 +02:00
return TRUE;
Split item->flags into base_flags and with_flags This was done to simplify copying of with_* flags Other things: - Changed Flags to C++ enums, which enables gdb to print out bit values for the flags. This also enables compiler errors if one tries to manipulate a non existing bit in a variable. - Added set_maybe_null() as a shortcut as setting the MAYBE_NULL flags was used in a LOT of places. - Renamed PARAM flag to SP_VAR to ensure it's not confused with persistent statement parameters.
2020-09-02 03:13:32 +03:00
set_maybe_null();
MDEV-11071: Assertion `thd->transaction.stmt.is_empty()' failed in Locked_tables_list::unlock_locked_table fix_length_and_dec now return result (error/OK)
2018-05-08 15:26:26 +02:00
return FALSE;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
}
MDEV-17832 Protocol: extensions for Pluggable types and JSON, GEOMETRY
2020-03-10 16:05:42 +04:00
bool set_format_by_check_constraint(Send_field_extended_metadata *to) const
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
override
MDEV-17832 Protocol: extensions for Pluggable types and JSON, GEOMETRY
2020-03-10 16:05:42 +04:00
{
static const Lex_cstring fmt(STRING_WITH_LEN("json"));
return to->set_format_name(fmt);
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_valid>(thd, this); }
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
enum Functype functype() const override { return JSON_VALID_FUNC; }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-23143 Add JSON_EQUALS function This patch implements JSON_EQUALS SQL function. The function takes advantage of the json_normalize functionality and does the following: norm_a = json_normalize(a) norm_b = json_normalize(b) return strcmp(norm_a, norm_b) Co-authored-by: Vicențiu Ciorbaru <vicentiu@mariadb.org>
2021-07-02 08:20:37 +02:00
class Item_func_json_equals: public Item_bool_func
{
public:
Item_func_json_equals(THD *thd, Item *a, Item *b):
Item_bool_func(thd, a, b) {}
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_equals") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Merge branch '10.6' into 10.11
2024-07-20 08:16:24 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-23143 Add JSON_EQUALS function This patch implements JSON_EQUALS SQL function. The function takes advantage of the json_normalize functionality and does the following: norm_a = json_normalize(a) norm_b = json_normalize(b) return strcmp(norm_a, norm_b) Co-authored-by: Vicențiu Ciorbaru <vicentiu@mariadb.org>
2021-07-02 08:20:37 +02:00
{ return get_item_copy<Item_func_json_equals>(thd, this); }
Merge branch '10.6' into 10.11
2024-10-29 15:24:38 +01:00
bool val_bool() override;
MDEV-23143 Add JSON_EQUALS function This patch implements JSON_EQUALS SQL function. The function takes advantage of the json_normalize functionality and does the following: norm_a = json_normalize(a) norm_b = json_normalize(b) return strcmp(norm_a, norm_b) Co-authored-by: Vicențiu Ciorbaru <vicentiu@mariadb.org>
2021-07-02 08:20:37 +02:00
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_exists: public Item_bool_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
json_path_with_flags path;
String tmp_js, tmp_path;
public:
MDEV-11573 JSON_LENGTH returns incorrect results. Item_func_json_length fixed.
2016-12-24 11:40:31 +04:00
Item_func_json_exists(THD *thd, Item *js, Item *i_path):
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_bool_func(thd, js, i_path) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_exists") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_exists>(thd, this); }
MDEV-34123 CONCAT Function Returns Unexpected Empty Set in Query Search conditions were evaluated using val_int(), which was wrong. Fixing the code to use val_bool() instead. Details: - Adding a new item_base_t::IS_COND flag which marks Items used as <search condition> in WHERE, HAVING, JOIN ON, CASE WHEN clauses. The flag is at the parse time. These expressions must be evaluated using val_bool() rather than val_int(). Note, the optimizer creates more Items which are used as search conditions. Most of these items are not marked with IS_COND yet. This is OK for now, but eventually these Items can also be fixed to have the flag. - Adding a method Item::is_cond() which tests if the Item has the IS_COND flag. - Implementing Item_cache_bool. It evaluates the cached expression using val_bool() rather than val_int(). Overriding Type_handler_bool::Item_get_cache() to create Item_cache_bool. - Implementing Item::save_bool_in_field(). It uses val_bool() rather than val_int() to evaluate the expression. - Implementing Type_handler_bool::Item_save_in_field() using Item::save_bool_in_field(). - Fixing all Item_bool_func descendants to implement a virtual val_bool() rather than a virtual val_int(). - To find places where val_int() should be fixed to val_bool(), a few DBUG_ASSERT(!is_cond()) where added into val_int() implementations of selected (most frequent) classes: Item_field Item_str_func Item_datefunc Item_timefunc Item_datetimefunc Item_cache_bool Item_bool_func Item_func_hybrid_field_type Item_basic_constant descendants - Fixing all places where DBUG_ASSERT() happened during an "mtr" run to use val_bool() instead of val_int().
2024-05-14 09:19:34 +04:00
bool val_bool() override;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Item_json_func: public Item_str_func
{
public:
Item_json_func(THD *thd)
:Item_str_func(thd) { }
Item_json_func(THD *thd, Item *a)
:Item_str_func(thd, a) { }
Item_json_func(THD *thd, Item *a, Item *b)
:Item_str_func(thd, a, b) { }
Item_json_func(THD *thd, List<Item> &list)
:Item_str_func(thd, list) { }
MDEV-27018 IF and COALESCE lose "json" property Hybrid functions (IF, COALESCE, etc) did not preserve the JSON property from their arguments. The same problem was repeatable for single row subselects. The problem happened because the method Item::is_json_type() was inconsistently implemented across the Item hierarchy. For example, Item_hybrid_func and Item_singlerow_subselect did not override is_json_type(). Solution: - Removing Item::is_json_type() - Implementing specific JSON type handlers: Type_handler_string_json Type_handler_varchar_json Type_handler_tiny_blob_json Type_handler_blob_json Type_handler_medium_blob_json Type_handler_long_blob_json - Reusing the existing data type infrastructure to pass JSON type handlers across all item types, including classes Item_hybrid_func and Item_singlerow_subselect. Note, these two classes themselves do not need any changes! - Extending the data type infrastructure so data types can inherit their properties (e.g. aggregation rules) from their base data types. E.g. VARCHAR/JSON acts as VARCHAR, LONGTEXT/JSON acts as LONGTEXT when mixed to a non-JSON data type. This is done by: - adding virtual method Type_handler::type_handler_base() - adding a helper class Type_handler_pair - refactoring Type_handler_hybrid_field_type methods aggregate_for_result(), aggregate_for_min_max(), aggregate_for_num_op() to use Type_handler_pair. This change also fixes: MDEV-27361 Hybrid functions with JSON arguments do not send format metadata Also, adding mtr tests for JSON replication. It was not covered yet. And the current patch changes the replication code slightly.
2022-01-10 18:05:55 +04:00
const Type_handler *type_handler() const override
MDEV-17832 Protocol: extensions for Pluggable types and JSON, GEOMETRY
2020-03-10 16:05:42 +04:00
{
MDEV-27018 IF and COALESCE lose "json" property Hybrid functions (IF, COALESCE, etc) did not preserve the JSON property from their arguments. The same problem was repeatable for single row subselects. The problem happened because the method Item::is_json_type() was inconsistently implemented across the Item hierarchy. For example, Item_hybrid_func and Item_singlerow_subselect did not override is_json_type(). Solution: - Removing Item::is_json_type() - Implementing specific JSON type handlers: Type_handler_string_json Type_handler_varchar_json Type_handler_tiny_blob_json Type_handler_blob_json Type_handler_medium_blob_json Type_handler_long_blob_json - Reusing the existing data type infrastructure to pass JSON type handlers across all item types, including classes Item_hybrid_func and Item_singlerow_subselect. Note, these two classes themselves do not need any changes! - Extending the data type infrastructure so data types can inherit their properties (e.g. aggregation rules) from their base data types. E.g. VARCHAR/JSON acts as VARCHAR, LONGTEXT/JSON acts as LONGTEXT when mixed to a non-JSON data type. This is done by: - adding virtual method Type_handler::type_handler_base() - adding a helper class Type_handler_pair - refactoring Type_handler_hybrid_field_type methods aggregate_for_result(), aggregate_for_min_max(), aggregate_for_num_op() to use Type_handler_pair. This change also fixes: MDEV-27361 Hybrid functions with JSON arguments do not send format metadata Also, adding mtr tests for JSON replication. It was not covered yet. And the current patch changes the replication code slightly.
2022-01-10 18:05:55 +04:00
return Type_handler_json_common::json_type_handler(max_length);
MDEV-17832 Protocol: extensions for Pluggable types and JSON, GEOMETRY
2020-03-10 16:05:42 +04:00
}
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
};
class Item_func_json_value: public Item_str_func,
public Json_path_extractor
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
public:
MDEV-11573 JSON_LENGTH returns incorrect results. Item_func_json_length fixed.
2016-12-24 11:40:31 +04:00
Item_func_json_value(THD *thd, Item *js, Item *i_path):
Item_str_func(thd, js, i_path) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_value") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override ;
Remove the most annoying clang-cl warnings Fix "consistent override", and "unused private field" warnings in popular header files.
2019-06-29 23:53:09 +02:00
String *val_str(String *to) override
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
{
null_value= Json_path_extractor::extract(to, args[0], args[1],
collation.collation);
return null_value ? NULL : to;
}
bool check_and_get_value(Json_engine_scan *je,
String *res, int *error) override
{
return je->check_and_get_value_scalar(res, error);
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_value>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Item_func_json_query: public Item_json_func,
public Json_path_extractor
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
public:
MDEV-11573 JSON_LENGTH returns incorrect results. Item_func_json_length fixed.
2016-12-24 11:40:31 +04:00
Item_func_json_query(THD *thd, Item *js, Item *i_path):
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
Item_json_func(thd, js, i_path) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_query") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Remove the most annoying clang-cl warnings Fix "consistent override", and "unused private field" warnings in popular header files.
2019-06-29 23:53:09 +02:00
String *val_str(String *to) override
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
{
null_value= Json_path_extractor::extract(to, args[0], args[1],
collation.collation);
return null_value ? NULL : to;
}
bool check_and_get_value(Json_engine_scan *je,
String *res, int *error) override
{
MDEV-30145: JSON_TABLE: allow to retrieve the key when iterating on JSON objects Idea behind implementation: We get the json object specified by the json path. Then, transform it into key-value pairs by going over the json. Get each key-value pair one-by-one and return the result.
2023-05-25 15:45:43 +05:30
return je->check_and_get_value_complex(res, error, JSON_VALUE_UNINITIALIZED);
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_query>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
class Item_func_json_quote: public Item_str_func
{
protected:
String tmp_s;
public:
Item_func_json_quote(THD *thd, Item *s): Item_str_func(thd, s) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_quote") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_quote>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
class Item_func_json_unquote: public Item_str_func
{
protected:
String tmp_s;
MDEV-12604 Comparison of JSON_EXTRACT result differs with Mysql. Comparison fixed to take the actual type of JSON value into account. Bug in escaping handling fixed.
2017-08-08 10:35:26 +04:00
String *read_json(json_engine_t *je);
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
public:
Item_func_json_unquote(THD *thd, Item *s): Item_str_func(thd, s) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_unquote") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_unquote>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Item_json_str_multipath: public Item_json_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
json_path_with_flags *paths;
String *tmp_paths;
MDEV-32733: Two JSON related tests running in PS mode fail on server built with -DWITH_PROTECT_STATEMENT_MEMROOT=YES The tests main.func_json and json.json_no_table fail on server built with the option -DWITH_PROTECT_STATEMENT_MEMROOT=YES by the reason that a memory is allocated on the statement's memory root on the second execution of a query that uses the function json_contains_path(). The reason that a memory is allocated on second execution of a prepared statement that ivokes the function json_contains_path() is that a memory allocated on every call of the method Item_json_str_multipath::fix_fields To fix the issue, memory allocation should be done only once on first call of the method Item_json_str_multipath::fix_fields. Simmilar issue take place inside the method Item_func_json_contains_path::fix_fields. Both methods are modified to make memory allocation only once on its first execution and later re-use the allocated memory. Before this patch the memory referenced by the pointers stored in the array tmp_paths were released by the method Item_func_json_contains_path::cleanup that is called on finishing execution of a prepared statement. Now that memory allocation performed inside the method Item_json_str_multipath::fix_fields is done only once, the item clean up has degenerate form and can be delegated to the cleanup() method of the base class and memory deallocation can be performed in the destructor.
2023-11-10 17:46:19 +07:00
private:
/**
Number of paths returned by calling virtual method get_n_paths() and
remembered inside fix_fields(). It is used by the virtual destructor
~Item_json_str_multipath() to iterate along allocated memory chunks stored
in the array tmp_paths and free every of them. The virtual method
get_n_paths() can't be used for this goal from within virtual destructor.
We could get rid of the virtual method get_n_paths() and store the number
of paths directly in the constructor of classes derived from the class
Item_json_str_multipath but presence of the method get_n_paths() allows
to check invariant that the number of arguments not changed between
sequential runs of the same prepared statement that seems to be useful.
*/
uint n_paths;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
public:
Item_json_str_multipath(THD *thd, List<Item> &list):
Merge branch '10.4' into 10.5
2023-12-01 13:43:58 +01:00
Item_json_func(thd, list), paths(NULL), tmp_paths(0), n_paths(0) {}
MDEV-32733: Two JSON related tests running in PS mode fail on server built with -DWITH_PROTECT_STATEMENT_MEMROOT=YES The tests main.func_json and json.json_no_table fail on server built with the option -DWITH_PROTECT_STATEMENT_MEMROOT=YES by the reason that a memory is allocated on the statement's memory root on the second execution of a query that uses the function json_contains_path(). The reason that a memory is allocated on second execution of a prepared statement that ivokes the function json_contains_path() is that a memory allocated on every call of the method Item_json_str_multipath::fix_fields To fix the issue, memory allocation should be done only once on first call of the method Item_json_str_multipath::fix_fields. Simmilar issue take place inside the method Item_func_json_contains_path::fix_fields. Both methods are modified to make memory allocation only once on its first execution and later re-use the allocated memory. Before this patch the memory referenced by the pointers stored in the array tmp_paths were released by the method Item_func_json_contains_path::cleanup that is called on finishing execution of a prepared statement. Now that memory allocation performed inside the method Item_json_str_multipath::fix_fields is done only once, the item clean up has degenerate form and can be delegated to the cleanup() method of the base class and memory deallocation can be performed in the destructor.
2023-11-10 17:46:19 +07:00
virtual ~Item_json_str_multipath();
MDEV-33746 Supply missing override markings Find and fix missing virtual override markings. Updates cmake maintainer flags to include -Wsuggest-override and -Winconsistent-missing-override.
2024-06-12 09:46:26 -04:00
bool fix_fields(THD *thd, Item **ref) override;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
virtual uint get_n_paths() const = 0;
};
class Item_func_json_extract: public Item_json_str_multipath
{
protected:
String tmp_js;
public:
MDEV-12604 Comparison of JSON_EXTRACT result differs with Mysql. JSON_EXTRACT behaves specifically in the comparison, so we have to implement specific method for that in Arg_comparator. Conflicts: sql/item_cmpfunc.cc
2017-08-11 00:50:29 +04:00
String *read_json(String *str, json_value_types *type,
char **out_val, int *value_len);
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
Item_func_json_extract(THD *thd, List<Item> &list):
Item_json_str_multipath(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_extract") };
return name;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
enum Functype functype() const override { return JSON_EXTRACT_FUNC; }
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
longlong val_int() override;
double val_real() override;
my_decimal *val_decimal(my_decimal *) override;
uint get_n_paths() const override { return arg_count - 1; }
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_extract>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_contains: public Item_bool_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
String tmp_js;
MDEV-11453 JSON_CONTAINS returns incorrect values. The weird logic of json_contains was implemented.
2016-12-11 01:12:33 +04:00
json_path_with_flags path;
String tmp_path;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
bool a2_constant, a2_parsed;
String tmp_val, *val;
public:
Item_func_json_contains(THD *thd, List<Item> &list):
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_bool_func(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_contains") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
MDEV-34123 CONCAT Function Returns Unexpected Empty Set in Query Search conditions were evaluated using val_int(), which was wrong. Fixing the code to use val_bool() instead. Details: - Adding a new item_base_t::IS_COND flag which marks Items used as <search condition> in WHERE, HAVING, JOIN ON, CASE WHEN clauses. The flag is at the parse time. These expressions must be evaluated using val_bool() rather than val_int(). Note, the optimizer creates more Items which are used as search conditions. Most of these items are not marked with IS_COND yet. This is OK for now, but eventually these Items can also be fixed to have the flag. - Adding a method Item::is_cond() which tests if the Item has the IS_COND flag. - Implementing Item_cache_bool. It evaluates the cached expression using val_bool() rather than val_int(). Overriding Type_handler_bool::Item_get_cache() to create Item_cache_bool. - Implementing Item::save_bool_in_field(). It uses val_bool() rather than val_int() to evaluate the expression. - Implementing Type_handler_bool::Item_save_in_field() using Item::save_bool_in_field(). - Fixing all Item_bool_func descendants to implement a virtual val_bool() rather than a virtual val_int(). - To find places where val_int() should be fixed to val_bool(), a few DBUG_ASSERT(!is_cond()) where added into val_int() implementations of selected (most frequent) classes: Item_field Item_str_func Item_datefunc Item_timefunc Item_datetimefunc Item_cache_bool Item_bool_func Item_func_hybrid_field_type Item_basic_constant descendants - Fixing all places where DBUG_ASSERT() happened during an "mtr" run to use val_bool() instead of val_int().
2024-05-14 09:19:34 +04:00
bool val_bool() override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_contains>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_contains_path: public Item_bool_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
String tmp_js;
json_path_with_flags *paths;
String *tmp_paths;
bool mode_one;
bool ooa_constant, ooa_parsed;
MDEV-11557 port MySQL-5.7 JSON tests to MariaDB. json_no_table.test ported.
2017-01-24 17:34:44 +04:00
bool *p_found;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
public:
Item_func_json_contains_path(THD *thd, List<Item> &list):
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_bool_func(thd, list), tmp_paths(0) {}
MDEV-32733: Two JSON related tests running in PS mode fail on server built with -DWITH_PROTECT_STATEMENT_MEMROOT=YES The tests main.func_json and json.json_no_table fail on server built with the option -DWITH_PROTECT_STATEMENT_MEMROOT=YES by the reason that a memory is allocated on the statement's memory root on the second execution of a query that uses the function json_contains_path(). The reason that a memory is allocated on second execution of a prepared statement that ivokes the function json_contains_path() is that a memory allocated on every call of the method Item_json_str_multipath::fix_fields To fix the issue, memory allocation should be done only once on first call of the method Item_json_str_multipath::fix_fields. Simmilar issue take place inside the method Item_func_json_contains_path::fix_fields. Both methods are modified to make memory allocation only once on its first execution and later re-use the allocated memory. Before this patch the memory referenced by the pointers stored in the array tmp_paths were released by the method Item_func_json_contains_path::cleanup that is called on finishing execution of a prepared statement. Now that memory allocation performed inside the method Item_json_str_multipath::fix_fields is done only once, the item clean up has degenerate form and can be delegated to the cleanup() method of the base class and memory deallocation can be performed in the destructor.
2023-11-10 17:46:19 +07:00
virtual ~Item_func_json_contains_path();
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_contains_path") };
return name;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool fix_fields(THD *thd, Item **ref) override;
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
MDEV-34123 CONCAT Function Returns Unexpected Empty Set in Query Search conditions were evaluated using val_int(), which was wrong. Fixing the code to use val_bool() instead. Details: - Adding a new item_base_t::IS_COND flag which marks Items used as <search condition> in WHERE, HAVING, JOIN ON, CASE WHEN clauses. The flag is at the parse time. These expressions must be evaluated using val_bool() rather than val_int(). Note, the optimizer creates more Items which are used as search conditions. Most of these items are not marked with IS_COND yet. This is OK for now, but eventually these Items can also be fixed to have the flag. - Adding a method Item::is_cond() which tests if the Item has the IS_COND flag. - Implementing Item_cache_bool. It evaluates the cached expression using val_bool() rather than val_int(). Overriding Type_handler_bool::Item_get_cache() to create Item_cache_bool. - Implementing Item::save_bool_in_field(). It uses val_bool() rather than val_int() to evaluate the expression. - Implementing Type_handler_bool::Item_save_in_field() using Item::save_bool_in_field(). - Fixing all Item_bool_func descendants to implement a virtual val_bool() rather than a virtual val_int(). - To find places where val_int() should be fixed to val_bool(), a few DBUG_ASSERT(!is_cond()) where added into val_int() implementations of selected (most frequent) classes: Item_field Item_str_func Item_datefunc Item_timefunc Item_datetimefunc Item_cache_bool Item_bool_func Item_func_hybrid_field_type Item_basic_constant descendants - Fixing all places where DBUG_ASSERT() happened during an "mtr" run to use val_bool() instead of val_int().
2024-05-14 09:19:34 +04:00
bool val_bool() override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_contains_path>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Item_func_json_array: public Item_json_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
protected:
String tmp_val;
MDEV-11833 JSON functions don't seem to respect max_allowed_packet. Now let's check JSON length to fit the max_allowed packet.
2017-03-14 17:31:14 +04:00
ulong result_limit;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
public:
Item_func_json_array(THD *thd):
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
Item_json_func(thd) {}
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
Item_func_json_array(THD *thd, List<Item> &list):
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
Item_json_func(thd, list) {}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_array") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_array>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
class Item_func_json_array_append: public Item_json_str_multipath
{
protected:
String tmp_js;
String tmp_val;
public:
Item_func_json_array_append(THD *thd, List<Item> &list):
Item_json_str_multipath(thd, list) {}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
uint get_n_paths() const override { return arg_count/2; }
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_array_append") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_array_append>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
class Item_func_json_array_insert: public Item_func_json_array_append
{
public:
Item_func_json_array_insert(THD *thd, List<Item> &list):
Item_func_json_array_append(thd, list) {}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_array_insert") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_array_insert>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
class Item_func_json_object: public Item_func_json_array
{
public:
Item_func_json_object(THD *thd):
Item_func_json_array(thd) {}
Item_func_json_object(THD *thd, List<Item> &list):
Item_func_json_array(thd, list) {}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_object") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_object>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
class Item_func_json_merge: public Item_func_json_array
{
protected:
MDEV-11433 JSON_MERGE returns a non-NULL result with a NULL argument. Item_func_json_merge fixed.
2016-12-06 01:32:13 +04:00
String tmp_js1, tmp_js2;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
public:
Item_func_json_merge(THD *thd, List<Item> &list):
Item_func_json_array(thd, list) {}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_merge_preserve") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_merge>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-13992 Implement JSON_MERGE_PATCH. JSON_MERGE_PATCH implemented. Added JSON_MERGE_PRESERVE as a synonim for the JSON_MERGE.
2019-05-17 11:53:58 +04:00
class Item_func_json_merge_patch: public Item_func_json_merge
{
public:
Item_func_json_merge_patch(THD *thd, List<Item> &list):
Item_func_json_merge(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_merge_patch") };
return name;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Merge branch '10.2' into 10.3
2019-05-17 17:23:01 +02:00
{ return get_item_copy<Item_func_json_merge_patch>(thd, this); }
MDEV-13992 Implement JSON_MERGE_PATCH. JSON_MERGE_PATCH implemented. Added JSON_MERGE_PRESERVE as a synonim for the JSON_MERGE.
2019-05-17 11:53:58 +04:00
};
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
MDEV-16375 Function to normalize a json value This patch implements JSON_NORMALIZE SQL function. Co-authored-by: Vicențiu Ciorbaru <vicentiu@mariadb.org>
2021-07-02 08:18:16 +02:00
class Item_func_json_normalize: public Item_json_func
{
public:
Item_func_json_normalize(THD *thd, Item *a):
Item_json_func(thd, a) {}
String *val_str(String *) override;
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_normalize") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Merge branch '10.6' into 10.11
2024-07-20 08:16:24 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-16375 Function to normalize a json value This patch implements JSON_NORMALIZE SQL function. Co-authored-by: Vicențiu Ciorbaru <vicentiu@mariadb.org>
2021-07-02 08:18:16 +02:00
{ return get_item_copy<Item_func_json_normalize>(thd, this); }
};
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
class Item_func_json_object_to_array: public Item_json_func
{
protected:
String tmp;
public:
Item_func_json_object_to_array(THD *thd, Item *a):
Item_json_func(thd, a) {}
String *val_str(String *) override;
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_object_to_array") };
return name;
}
bool fix_length_and_dec(THD *thd) override;
Merge branch '11.1' into 11.2
2024-08-03 09:50:16 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
{ return get_item_copy<Item_func_json_object_to_array>(thd, this); }
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_length: public Item_long_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool check_arguments() const override
MDEV-13967 Parameter data type control for Item_long_func - Implementing stricter data type control for Item_long_func descendants - Cleanup: renaming Type_handler::can_return_str_ascii() to can_return_text() (a better name).
2017-10-01 00:30:58 +04:00
{
Merge 10.5 into 10.6
2023-04-11 16:15:19 +03:00
const LEX_CSTRING name= func_name_cstring();
MDEV-24538: JSON_LENGTH does not return error upon wrong number of parameters
2023-02-12 16:40:40 +01:00
if (arg_count == 0 || arg_count > 2)
{
Merge 10.5 into 10.6
2023-04-11 16:15:19 +03:00
my_error(ER_WRONG_PARAMCOUNT_TO_NATIVE_FCT, MYF(0), name.str);
MDEV-24538: JSON_LENGTH does not return error upon wrong number of parameters
2023-02-12 16:40:40 +01:00
return true;
}
Merge 10.5 into 10.6
2023-04-11 16:15:19 +03:00
return args[0]->check_type_can_return_text(name) ||
(arg_count > 1 && args[1]->check_type_general_purpose_string(name));
MDEV-13967 Parameter data type control for Item_long_func - Implementing stricter data type control for Item_long_func descendants - Cleanup: renaming Type_handler::can_return_str_ascii() to can_return_text() (a better name).
2017-10-01 00:30:58 +04:00
}
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
protected:
MDEV-11573 JSON_LENGTH returns incorrect results. Item_func_json_length fixed.
2016-12-24 11:40:31 +04:00
json_path_with_flags path;
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
String tmp_js;
String tmp_path;
public:
Item_func_json_length(THD *thd, List<Item> &list):
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_long_func(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_length") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
longlong val_int() override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_length>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
class Item_func_json_depth: public Item_long_func
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
{
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool check_arguments() const override
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
{ return args[0]->check_type_can_return_text(func_name_cstring()); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
protected:
String tmp_js;
public:
MDEV-12854 Synchronize CREATE..SELECT data type and result set metadata data type for INT functions
2017-05-20 16:29:11 +04:00
Item_func_json_depth(THD *thd, Item *js): Item_long_func(thd, js) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_depth") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override { max_length= 10; return FALSE; }
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
longlong val_int() override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_depth>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
class Item_func_json_type: public Item_str_func
{
protected:
String tmp_js;
public:
Item_func_json_type(THD *thd, Item *js): Item_str_func(thd, js) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_type") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_type>(thd, this); }
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
};
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
class Item_func_json_insert: public Item_json_str_multipath
{
protected:
String tmp_js;
String tmp_val;
bool mode_insert, mode_replace;
public:
Item_func_json_insert(bool i_mode, bool r_mode, THD *thd, List<Item> &list):
Item_json_str_multipath(thd, list),
mode_insert(i_mode), mode_replace(r_mode) {}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
uint get_n_paths() const override { return arg_count/2; }
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
{
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
static LEX_CSTRING json_set= {STRING_WITH_LEN("json_set") };
static LEX_CSTRING json_insert= {STRING_WITH_LEN("json_insert") };
Merge 10.5 to 10.6
2021-08-19 12:25:00 +03:00
static LEX_CSTRING json_replace= {STRING_WITH_LEN("json_replace") };
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
return (mode_insert ?
Merge 10.5 to 10.6
2021-08-19 12:25:00 +03:00
(mode_replace ? json_set : json_insert) : json_replace);
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_insert>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
class Item_func_json_remove: public Item_json_str_multipath
{
protected:
String tmp_js;
public:
Item_func_json_remove(THD *thd, List<Item> &list):
Item_json_str_multipath(thd, list) {}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
uint get_n_paths() const override { return arg_count - 1; }
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_remove") };
return name;
}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_remove>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
class Item_func_json_keys: public Item_str_func
{
protected:
json_path_with_flags path;
String tmp_js, tmp_path;
public:
Item_func_json_keys(THD *thd, List<Item> &list):
Item_str_func(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_keys") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_keys>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
class Item_func_json_search: public Item_json_str_multipath
{
protected:
MDEV-17018 JSON_SEARCH and User-Defined Variables. Item_func_json_search needs a member to store the path.
2018-09-12 21:57:28 +04:00
String tmp_js, tmp_path, esc_value;
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
bool mode_one;
bool ooa_constant, ooa_parsed;
int escape;
int n_path_found;
json_path_t sav_path;
int compare_json_value_wild(json_engine_t *je, const String *cmp_str);
public:
Item_func_json_search(THD *thd, List<Item> &list):
Item_json_str_multipath(thd, list) {}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_search") };
return name;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool fix_fields(THD *thd, Item **ref) override;
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *) override;
uint get_n_paths() const override { return arg_count > 4 ? arg_count - 4 : 0; }
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_search>(thd, this); }
MDEV-9143 JSON_xxx functions. The rest of mysql/json functions implemented. CAST AS JSON implemented.
2016-11-15 17:04:31 +04:00
};
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
class Item_func_json_format: public Item_json_func
MDEV-11557 port MySQL-5.7 JSON tests to MariaDB. Fixes for issues found.
2017-02-06 06:47:48 +04:00
{
public:
enum formats
{
NONE,
COMPACT,
LOOSE,
DETAILED
};
protected:
formats fmt;
String tmp_js;
public:
Item_func_json_format(THD *thd, Item *js, formats format):
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
Item_json_func(thd, js), fmt(format) {}
MDEV-11439 No data type JSON, but CAST(something AS JSON) pretends to work. json_detailed() fixed
2017-02-14 17:51:03 +04:00
Item_func_json_format(THD *thd, List<Item> &list):
MDEV-19888 Add abstract class Item_json_func
2019-06-28 12:35:02 +04:00
Item_json_func(thd, list), fmt(DETAILED) {}
MDEV-11439 No data type JSON, but CAST(something AS JSON) pretends to work. json_detailed() fixed
2017-02-14 17:51:03 +04:00
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override;
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String *val_str(String *str) override;
String *val_json(String *str) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
Remove not used mem_root argument from build_clone(), get_copy() and get_item_copy() TODO: - Make get_thd_memroot() inline - To do this, we need to reduce dependence of include files, especially so that sql_class.h is not depending in item.h
2017-11-22 08:01:07 +02:00
{ return get_item_copy<Item_func_json_format>(thd, this); }
MDEV-11557 port MySQL-5.7 JSON tests to MariaDB. Fixes for issues found.
2017-02-06 06:47:48 +04:00
};
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
class Item_func_json_arrayagg : public Item_func_group_concat
{
MDEV-22640, MDEV-22449, MDEV-21528 JSON_ARRAYAGG crashes with NULL values. We have to include NULL in the result which the GOUP_CONCAT doesn't always do. Also converting should be done into another String instance as these can be same.
2020-06-04 10:00:56 +04:00
protected:
/*
Overrides Item_func_group_concat::skip_nulls()
NULL-s should be added to the result as JSON null value.
*/
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool skip_nulls() const override { return false; }
String *get_str_from_item(Item *i, String *tmp) override;
MDEV-21914 JSON_ARRAYAGG doesn't reject ORDER BY clause, but doesn't work either. ORDER BY fixed for JSON_ARRAYAGG.
2020-06-04 13:53:14 +04:00
String *get_str_from_field(Item *i, Field *f, String *tmp,
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
const uchar *key, size_t offset) override;
MDEV-22844 JSON_ARRAYAGG is limited by group_concat_max_len. Warning message and function result fixed
2020-06-15 22:51:21 +04:00
void cut_max_length(String *result,
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
uint old_length, uint max_length) const override;
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
public:
MDEV-21914 JSON_ARRAYAGG doesn't reject ORDER BY clause, but doesn't work either. ORDER BY fixed for JSON_ARRAYAGG.
2020-06-04 13:53:14 +04:00
String m_tmp_json; /* Used in get_str_from_*.. */
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
Item_func_json_arrayagg(THD *thd, Name_resolution_context *context_arg,
bool is_distinct, List<Item> *is_select,
const SQL_I_List<ORDER> &is_order, String *is_separator,
bool limit_clause, Item *row_limit, Item *offset_limit):
Item_func_group_concat(thd, context_arg, is_distinct, is_select, is_order,
is_separator, limit_clause, row_limit, offset_limit)
{
}
MDEV-23004 When using GROUP BY with JSON_ARRAYAGG with joint table, the square brackets are not included. Item_func_json_arrayagg::copy_or_same() should be implemented.
2021-06-24 13:28:28 +04:00
Item_func_json_arrayagg(THD *thd, Item_func_json_arrayagg *item) :
Item_func_group_concat(thd, item) {}
MDEV-27018 IF and COALESCE lose "json" property Hybrid functions (IF, COALESCE, etc) did not preserve the JSON property from their arguments. The same problem was repeatable for single row subselects. The problem happened because the method Item::is_json_type() was inconsistently implemented across the Item hierarchy. For example, Item_hybrid_func and Item_singlerow_subselect did not override is_json_type(). Solution: - Removing Item::is_json_type() - Implementing specific JSON type handlers: Type_handler_string_json Type_handler_varchar_json Type_handler_tiny_blob_json Type_handler_blob_json Type_handler_medium_blob_json Type_handler_long_blob_json - Reusing the existing data type infrastructure to pass JSON type handlers across all item types, including classes Item_hybrid_func and Item_singlerow_subselect. Note, these two classes themselves do not need any changes! - Extending the data type infrastructure so data types can inherit their properties (e.g. aggregation rules) from their base data types. E.g. VARCHAR/JSON acts as VARCHAR, LONGTEXT/JSON acts as LONGTEXT when mixed to a non-JSON data type. This is done by: - adding virtual method Type_handler::type_handler_base() - adding a helper class Type_handler_pair - refactoring Type_handler_hybrid_field_type methods aggregate_for_result(), aggregate_for_min_max(), aggregate_for_num_op() to use Type_handler_pair. This change also fixes: MDEV-27361 Hybrid functions with JSON arguments do not send format metadata Also, adding mtr tests for JSON replication. It was not covered yet. And the current patch changes the replication code slightly.
2022-01-10 18:05:55 +04:00
const Type_handler *type_handler() const override
{
return Type_handler_json_common::json_type_handler_sum(this);
}
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_arrayagg(") };
return name;
}
MDEV-24784 JSON_ARRAYAGG charset issue set the proper charset for a temporary result
2023-11-02 14:57:37 +01:00
bool fix_fields(THD *thd, Item **ref) override;
MDEV-22640 fixup: clang -Winconsistent-missing-override
2021-06-29 15:02:10 +03:00
enum Sumfunctype sum_func() const override { return JSON_ARRAYAGG_FUNC; }
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
String* val_str(String *str) override;
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
MDEV-23004 When using GROUP BY with JSON_ARRAYAGG with joint table, the square brackets are not included. Item_func_json_arrayagg::copy_or_same() should be implemented.
2021-06-24 13:28:28 +04:00
Item *copy_or_same(THD* thd) override;
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
{ return get_item_copy<Item_func_json_arrayagg>(thd, this); }
};
class Item_func_json_objectagg : public Item_sum
{
String result;
public:
Item_func_json_objectagg(THD *thd, Item *key, Item *value) :
Item_sum(thd, key, value)
{
MDEV-23029: JSON_OBJECTAGG returns NULL when used together with GROUP BY Quick grouping is not supported for JSON_OBJECTAGG. The same for GROUP_CONCAT too so make sure that Item::quick_group is set to FALSE. We need to make sure that in the case of JSON_OBJECTAGG we don't create an index over grouping fields of the temp table and update the result after each iteration. Instead we should first sort the result in accordance to the GROUP BY fields and then perform the grouping and write the result to the temp table.
2020-06-29 20:06:28 +05:30
quick_group= FALSE;
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
result.append('{');
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
}
Item_func_json_objectagg(THD *thd, Item_func_json_objectagg *item);
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
void cleanup() override;
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
enum Sumfunctype sum_func () const override { return JSON_OBJECTAGG_FUNC;}
Reduce usage of strlen() Changes: - To detect automatic strlen() I removed the methods in String that uses 'const char *' without a length: - String::append(const char*) - Binary_string(const char *str) - String(const char *str, CHARSET_INFO *cs) - append_for_single_quote(const char *) All usage of append(const char*) is changed to either use String::append(char), String::append(const char*, size_t length) or String::append(LEX_CSTRING) - Added STRING_WITH_LEN() around constant string arguments to String::append() - Added overflow argument to escape_string_for_mysql() and escape_quotes_for_mysql() instead of returning (size_t) -1 on overflow. This was needed as most usage of the above functions never tested the result for -1 and would have given wrong results or crashes in case of overflows. - Added Item_func_or_sum::func_name_cstring(), which returns LEX_CSTRING. Changed all Item_func::func_name()'s to func_name_cstring()'s. The old Item_func_or_sum::func_name() is now an inline function that returns func_name_cstring().str. - Changed Item::mode_name() and Item::func_name_ext() to return LEX_CSTRING. - Changed for some functions the name argument from const char * to to const LEX_CSTRING &: - Item::Item_func_fix_attributes() - Item::check_type_...() - Type_std_attributes::agg_item_collations() - Type_std_attributes::agg_item_set_converter() - Type_std_attributes::agg_arg_charsets...() - Type_handler_hybrid_field_type::aggregate_for_result() - Type_handler_geometry::check_type_geom_or_binary() - Type_handler::Item_func_or_sum_illegal_param() - Predicant_to_list_comparator::add_value_skip_null() - Predicant_to_list_comparator::add_value() - cmp_item_row::prepare_comparators() - cmp_item_row::aggregate_row_elements_for_comparison() - Cursor_ref::print_func() - Removes String_space() as it was only used in one cases and that could be simplified to not use String_space(), thanks to the fixed my_vsnprintf(). - Added some const LEX_CSTRING's for common strings: - NULL_clex_str, DATA_clex_str, INDEX_clex_str. - Changed primary_key_name to a LEX_CSTRING - Renamed String::set_quick() to String::set_buffer_if_not_allocated() to clarify what the function really does. - Rename of protocol function: bool store(const char *from, CHARSET_INFO *cs) to bool store_string_or_null(const char *from, CHARSET_INFO *cs). This was done to both clarify the difference between this 'store' function and also to make it easier to find unoptimal usage of store() calls. - Added Protocol::store(const LEX_CSTRING*, CHARSET_INFO*) - Changed some 'const char*' arrays to instead be of type LEX_CSTRING. - class Item_func_units now used LEX_CSTRING for name. Other things: - Fixed a bug in mysql.cc:construct_prompt() where a wrong escape character in the prompt would cause some part of the prompt to be duplicated. - Fixed a lot of instances where the length of the argument to append is known or easily obtain but was not used. - Removed some not needed 'virtual' definition for functions that was inherited from the parent. I added override to these. - Fixed Ordered_key::print() to preallocate needed buffer. Old code could case memory overruns. - Simplified some loops when adding char * to a String with delimiters.
2020-08-12 20:29:55 +03:00
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_objectagg") };
return name;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
const Type_handler *type_handler() const override
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
{
MDEV-27018 IF and COALESCE lose "json" property Hybrid functions (IF, COALESCE, etc) did not preserve the JSON property from their arguments. The same problem was repeatable for single row subselects. The problem happened because the method Item::is_json_type() was inconsistently implemented across the Item hierarchy. For example, Item_hybrid_func and Item_singlerow_subselect did not override is_json_type(). Solution: - Removing Item::is_json_type() - Implementing specific JSON type handlers: Type_handler_string_json Type_handler_varchar_json Type_handler_tiny_blob_json Type_handler_blob_json Type_handler_medium_blob_json Type_handler_long_blob_json - Reusing the existing data type infrastructure to pass JSON type handlers across all item types, including classes Item_hybrid_func and Item_singlerow_subselect. Note, these two classes themselves do not need any changes! - Extending the data type infrastructure so data types can inherit their properties (e.g. aggregation rules) from their base data types. E.g. VARCHAR/JSON acts as VARCHAR, LONGTEXT/JSON acts as LONGTEXT when mixed to a non-JSON data type. This is done by: - adding virtual method Type_handler::type_handler_base() - adding a helper class Type_handler_pair - refactoring Type_handler_hybrid_field_type methods aggregate_for_result(), aggregate_for_min_max(), aggregate_for_num_op() to use Type_handler_pair. This change also fixes: MDEV-27361 Hybrid functions with JSON arguments do not send format metadata Also, adding mtr tests for JSON replication. It was not covered yet. And the current patch changes the replication code slightly.
2022-01-10 18:05:55 +04:00
return Type_handler_json_common::json_type_handler_sum(this);
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
void clear() override;
bool add() override;
void reset_field() override { DBUG_ASSERT(0); } // not used
void update_field() override { DBUG_ASSERT(0); } // not used
bool fix_fields(THD *,Item **) override;
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
MDEV-22640 fixup: clang -Winconsistent-missing-override
2021-06-29 15:02:10 +03:00
double val_real() override { return 0.0; }
longlong val_int() override { return 0; }
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
my_decimal *val_decimal(my_decimal *decimal_value) override
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
{
my_decimal_set_zero(decimal_value);
return decimal_value;
}
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
bool get_date(THD *thd, MYSQL_TIME *ltime, date_mode_t fuzzydate) override
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
{
return get_date_from_string(thd, ltime, fuzzydate);
}
MDEV-22640 fixup: clang -Winconsistent-missing-override
2021-06-29 15:02:10 +03:00
String* val_str(String* str) override;
Added override to all releveant methods in Item (and a few other classes) Other things: - Remove inline and virtual for methods that are overrides - Added a 'final' to some Item classes
2020-08-19 02:53:22 +03:00
Item *copy_or_same(THD* thd) override;
void no_rows_in_result() override {}
MDEV-34490 get_copy() and build_clone() may return an instance of an ancestor class instead of a copy/clone The `Item` class methods `get_copy()`, `build_clone()`, and `clone_item()` face an issue where they may be defined in a descendant class (e.g., `Item_func`) but not in a further descendant (e.g., `Item_func_child`). This can lead to scenarios where `build_clone()`, when operating on an instance of `Item_func_child` with a pointer to the base class (`Item`), returns an instance of `Item_func` instead of `Item_func_child`. Since this limitation cannot be resolved at compile time, this commit introduces runtime type checks for the copy/clone operations. A debug assertion will now trigger in case of a type mismatch. `get_copy()`, `build_clone()`, and `clone_item()` are no more virtual, but virtual `do_get_copy()`, `do_build_clone()`, and `do_clone_item()` are added to the protected section of the class `Item`. Additionally, const qualifiers have been added to certain methods to enhance code reliability. Reviewer: Oleksandr Byelkin <sanja@mariadb.com>
2024-07-03 15:42:21 +07:00
Item *do_get_copy(THD *thd) const override
MDEV-16620 JSON_ARRAYAGG and JSON_OBJECTAGG. Ison_objectagg implemented.
2019-10-14 14:24:22 +04:00
{ return get_item_copy<Item_func_json_objectagg>(thd, this); }
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
};
Merge branch '10.5' into 10.6
2022-02-03 17:01:31 +01:00
extern bool is_json_type(const Item *item);
MDEV-16620: Add JSON_ARRAYAGG function The JSON_ARRAYAGG function extends the GROUP_CONCAT function and provides a method of aggregating JSON results. The current implementation supports DISTINCT and LIMIT but not ORDER BY (Oracle supports GROUP BY). Adding GROUP BY support is possible but it requires some extra work as the grouping appears to be done inside a temporary table that complicates matters. Added test cases that covert aggregation of all JSON types and JSON validation for the generated results.
2019-07-04 13:12:08 +03:00
MDEV-27677: Implement JSON_OVERLAPS() 1) When at least one of the two json documents is of scalar type: 1.a) If value and json document both are scalar, then return true if they have same type and value. 1.b) If json document is scalar but other is array (or vice versa), then return true if array has at least one element of same type and value as scalar. 1.c) If one is scalar and other is object, then return false because it can't be compared. 2) When both arguments are of non-scalar type and below conditons are satisfied then return true: 2.a) When both arguments are arrays: Iterate over the value and json document. If there exists at least one element in other array of same type and value as that of element in value. 2.b) If both arguments are objects: Iterate over value and json document and if there exists at least one key-value pair common between two objects. 2.c) If either of json document or value is array and other is object: Iterate over the array, if an element of type object is found, then compare it with the object (which is the other arguemnt). If the entire object matches i.e all they key value pairs match.
2022-01-30 15:45:25 +05:30
class Item_func_json_overlaps: public Item_bool_func
{
String tmp_js;
bool a2_constant, a2_parsed;
String tmp_val, *val;
public:
Item_func_json_overlaps(THD *thd, Item *a, Item *b):
MDEV-27128: Implement JSON Schema Validation FUNCTION Implementation: Implementation is made according to json schema validation draft 2020 JSON schema basically has same structure as that of json object, consisting of key-value pairs. So it can be parsed in the same manner as any json object. However, none of the keywords are mandatory, so making guess about the json value type based only on the keywords would be incorrect. Hence we need separate objects denoting each keyword. So during create_object_and_handle_keyword() we create appropriate objects based on the keywords and validate each of them individually on the json document by calling respective validate() function if the type matches. If any of them fails, return false, else return true.
2022-10-28 13:03:13 +05:30
Item_bool_func(thd, a, b)
{}
MDEV-27677: Implement JSON_OVERLAPS() 1) When at least one of the two json documents is of scalar type: 1.a) If value and json document both are scalar, then return true if they have same type and value. 1.b) If json document is scalar but other is array (or vice versa), then return true if array has at least one element of same type and value as scalar. 1.c) If one is scalar and other is object, then return false because it can't be compared. 2) When both arguments are of non-scalar type and below conditons are satisfied then return true: 2.a) When both arguments are arrays: Iterate over the value and json document. If there exists at least one element in other array of same type and value as that of element in value. 2.b) If both arguments are objects: Iterate over value and json document and if there exists at least one key-value pair common between two objects. 2.c) If either of json document or value is array and other is object: Iterate over the array, if an element of type object is found, then compare it with the object (which is the other arguemnt). If the entire object matches i.e all they key value pairs match.
2022-01-30 15:45:25 +05:30
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_overlaps") };
return name;
}
MDEV-23479: Add a THD* argument to Item_func_or_sum::fix_length_and_dec() Fix: Added THD *thd argument in Item_func_or_sum::fix_length_and_dec() and in fix_length_and_dec() for all derived classes of Item_func_or_sum.
2022-01-26 03:02:45 +05:30
bool fix_length_and_dec(THD *thd) override;
Merge branch '10.6' into 10.11
2024-10-29 15:24:38 +01:00
bool val_bool() override;
Merge branch '10.6' into 10.11
2024-07-20 08:16:24 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-27677: Implement JSON_OVERLAPS() 1) When at least one of the two json documents is of scalar type: 1.a) If value and json document both are scalar, then return true if they have same type and value. 1.b) If json document is scalar but other is array (or vice versa), then return true if array has at least one element of same type and value as scalar. 1.c) If one is scalar and other is object, then return false because it can't be compared. 2) When both arguments are of non-scalar type and below conditons are satisfied then return true: 2.a) When both arguments are arrays: Iterate over the value and json document. If there exists at least one element in other array of same type and value as that of element in value. 2.b) If both arguments are objects: Iterate over value and json document and if there exists at least one key-value pair common between two objects. 2.c) If either of json document or value is array and other is object: Iterate over the array, if an element of type object is found, then compare it with the object (which is the other arguemnt). If the entire object matches i.e all they key value pairs match.
2022-01-30 15:45:25 +05:30
{ return get_item_copy<Item_func_json_overlaps>(thd, this); }
};
MDEV-27128: Implement JSON Schema Validation FUNCTION Implementation: Implementation is made according to json schema validation draft 2020 JSON schema basically has same structure as that of json object, consisting of key-value pairs. So it can be parsed in the same manner as any json object. However, none of the keywords are mandatory, so making guess about the json value type based only on the keywords would be incorrect. Hence we need separate objects denoting each keyword. So during create_object_and_handle_keyword() we create appropriate objects based on the keywords and validate each of them individually on the json document by calling respective validate() function if the type matches. If any of them fails, return false, else return true.
2022-10-28 13:03:13 +05:30
class Item_func_json_schema_valid: public Item_bool_func
{
String tmp_js;
bool schema_parsed;
String tmp_val, *val;
List<Json_schema_keyword> keyword_list;
List<Json_schema_keyword> all_keywords;
public:
Item_func_json_schema_valid(THD *thd, Item *a, Item *b):
Item_bool_func(thd, a, b)
{
val= NULL;
schema_parsed= false;
MDEV-30677: Incorrect result for "SELECT JSON_SCHEMA_VALID('{}', NULL)" Analysis: null_value is not set if any one of the arguments is NULL. So it returns 1. Fix: when either argument is NULL, set null_value to true, so that null can be returned
2023-05-02 16:34:07 +05:30
set_maybe_null();
MDEV-27128: Implement JSON Schema Validation FUNCTION Implementation: Implementation is made according to json schema validation draft 2020 JSON schema basically has same structure as that of json object, consisting of key-value pairs. So it can be parsed in the same manner as any json object. However, none of the keywords are mandatory, so making guess about the json value type based only on the keywords would be incorrect. Hence we need separate objects denoting each keyword. So during create_object_and_handle_keyword() we create appropriate objects based on the keywords and validate each of them individually on the json document by calling respective validate() function if the type matches. If any of them fails, return false, else return true.
2022-10-28 13:03:13 +05:30
}
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_schema_valid") };
return name;
}
bool fix_length_and_dec(THD *thd) override;
Merge branch '10.11' into 11.2
2024-10-29 16:39:47 +01:00
bool val_bool() override;
Merge branch '10.11' into 11.1
2024-08-03 09:32:42 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-27128: Implement JSON Schema Validation FUNCTION Implementation: Implementation is made according to json schema validation draft 2020 JSON schema basically has same structure as that of json object, consisting of key-value pairs. So it can be parsed in the same manner as any json object. However, none of the keywords are mandatory, so making guess about the json value type based only on the keywords would be incorrect. Hence we need separate objects denoting each keyword. So during create_object_and_handle_keyword() we create appropriate objects based on the keywords and validate each of them individually on the json document by calling respective validate() function if the type matches. If any of them fails, return false, else return true.
2022-10-28 13:03:13 +05:30
{ return get_item_copy<Item_func_json_schema_valid>(thd, this); }
void cleanup() override;
};
MDEV-30145: JSON_TABLE: allow to retrieve the key when iterating on JSON objects Idea behind implementation: We get the json object specified by the json path. Then, transform it into key-value pairs by going over the json. Get each key-value pair one-by-one and return the result.
2023-05-25 15:45:43 +05:30
class Item_func_json_key_value: public Item_json_func,
public Json_path_extractor
{
String tmp_str;
public:
Item_func_json_key_value(THD *thd, Item *js, Item *i_path):
Item_json_func(thd, js, i_path) {}
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_key_value") };
return name;
}
bool fix_length_and_dec(THD *thd) override;
String *val_str(String *to) override;
bool check_and_get_value(Json_engine_scan *je,
String *res, int *error) override
{
return je->check_and_get_value_complex(res, error, JSON_VALUE_OBJECT);
}
bool get_key_value(json_engine_t *je, String *str);
Merge branch '11.1' into 11.2
2024-08-03 09:50:16 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-30145: JSON_TABLE: allow to retrieve the key when iterating on JSON objects Idea behind implementation: We get the json object specified by the json path. Then, transform it into key-value pairs by going over the json. Get each key-value pair one-by-one and return the result.
2023-05-25 15:45:43 +05:30
{ return get_item_copy<Item_func_json_key_value>(thd, this); }
};
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
class Item_func_json_array_intersect: public Item_str_func
{
protected:
String tmp_js1, tmp_js2;
bool hash_inited, root_inited;
HASH items;
MEM_ROOT hash_root;
MDEV-31411: JSON_ARRAY_INTERSECT/JSON_OBJECT_FILTER_KEYS should fetch data from a table similar to other JSON functions Analysis: Since we are fetching values for every row ( because we are running SELECT for all rows of a table ), correct value can be only obtained at the time of calling val_int() because it is called to get value for each row. Fix: Set up hash for each row instead of doing it during fixing fields.
2023-07-04 14:39:57 +05:30
bool parse_for_each_row;
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
public:
Item_func_json_array_intersect(THD *thd, Item *a, Item *b):
MDEV-31411: JSON_ARRAY_INTERSECT/JSON_OBJECT_FILTER_KEYS should fetch data from a table similar to other JSON functions Analysis: Since we are fetching values for every row ( because we are running SELECT for all rows of a table ), correct value can be only obtained at the time of calling val_int() because it is called to get value for each row. Fix: Set up hash for each row instead of doing it during fixing fields.
2023-07-04 14:39:57 +05:30
Item_str_func(thd, a, b) { hash_inited= root_inited= parse_for_each_row= false; }
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
String *val_str(String *) override;
bool fix_length_and_dec(THD *thd) override;
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_array_intersect") };
return name;
}
Merge branch '11.1' into 11.2
2024-08-03 09:50:16 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
{ return get_item_copy<Item_func_json_array_intersect>(thd, this); }
void cleanup() override
{
MDEV-26182 fixes for --ps * invoke parent's cleanup() * don't reinit memroot, if already inited (causes memory leak) also move free_root() from destructor to cleanup() to not accumulate allocations from prepare and multiple executes
2023-06-17 19:00:56 +02:00
Item_str_func::cleanup();
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
if (hash_inited)
my_hash_free(&items);
if (root_inited)
free_root(&hash_root, MYF(0));
}
MDEV-31411: JSON_ARRAY_INTERSECT/JSON_OBJECT_FILTER_KEYS should fetch data from a table similar to other JSON functions Analysis: Since we are fetching values for every row ( because we are running SELECT for all rows of a table ), correct value can be only obtained at the time of calling val_int() because it is called to get value for each row. Fix: Set up hash for each row instead of doing it during fixing fields.
2023-07-04 14:39:57 +05:30
void prepare_json_and_create_hash(json_engine_t *je1, String *js);
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
};
class Item_func_json_object_filter_keys: public Item_str_func
{
protected:
String tmp_js1, tmp_js2;
bool hash_inited, root_inited;
HASH items;
MEM_ROOT hash_root;
public:
Item_func_json_object_filter_keys(THD *thd, Item *a, Item *b):
Item_str_func(thd, a, b) { hash_inited= root_inited= false; }
String *val_str(String *) override;
bool fix_length_and_dec(THD *thd) override;
LEX_CSTRING func_name_cstring() const override
{
static LEX_CSTRING name= {STRING_WITH_LEN("json_object_filter_keys") };
return name;
}
Merge branch '11.1' into 11.2
2024-08-03 09:50:16 +02:00
Item *do_get_copy(THD *thd) const override
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
{ return get_item_copy<Item_func_json_object_filter_keys>(thd, this); }
void cleanup() override
{
MDEV-26182 fixes for --ps * invoke parent's cleanup() * don't reinit memroot, if already inited (causes memory leak) also move free_root() from destructor to cleanup() to not accumulate allocations from prepare and multiple executes
2023-06-17 19:00:56 +02:00
Item_str_func::cleanup();
MDEV-26182: Implement JSON_INTERSECT() The idea is to have simple functions that the user can combine to produce the exact result one wants, whether the user wants JSON object that has common keys with another JSON object, or same key/value pair etc. So making simpler function helps here. We accomplish this by making three separate functions. 1) JSON_OBJECT_FILTER_KEYS(Obj, Arr_keys): Put keys ( which are basically strings ) in hash, go over the object and get key one by one. If the key is present in the hash, add the key-value pair to result. 2) JSON_OBJECT_TO_ARRAY(Obj) : Create a string variable, Go over the json object, and add each key value pair as an array into the result. 3) JSON_ARRAY_INTERSECT(arr1, arr2) : Go over one of the json and add each item of the array in hash (after normalizing each item). Go over the second array, search the normalized item one by one in the hash. If item is found, add it to the result. Implementation Idea: Holyfoot ( Alexey Botchkov) Author: tanruixiang and Rucha Deodhar
2022-06-16 15:05:35 +08:00
if (hash_inited)
my_hash_free(&items);
if (root_inited)
free_root(&hash_root, MYF(0));
}
};
MDEV-9143 JSON_xxx functions. strings/json_lib.c added as a JSON library. SQL frunction added with sql/item_jsonfunc.h/cc
2016-10-19 14:10:03 +04:00
#endif /* ITEM_JSONFUNC_INCLUDED */
Reference in a new issue Copy permalink