mysql-server/sql/sql_lex.cc
2025-03-05 14:31:37 +07:00

5281 lines
171 KiB
C++

/*
Copyright (c) 2000, 2024, Oracle and/or its affiliates.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License, version 2.0,
as published by the Free Software Foundation.
This program is designed to work with certain software (including
but not limited to OpenSSL) that is licensed under separate terms,
as designated in a particular file or component or in included license
documentation. The authors of MySQL hereby grant you an additional
permission to link the program and your derivative works with the
separately licensed software that they have either included with
the program or referenced in the documentation.
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, version 2.0, 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 */
/* A lexical scanner on a temporary buffer with a yacc interface */
#include "sql/sql_lex.h"
#include <algorithm> // find_if, iter_swap, reverse
#include <climits>
#include <cstdlib>
#include <initializer_list>
#include <memory>
#include "field_types.h"
#include "m_string.h"
#include "my_alloc.h"
#include "my_dbug.h"
#include "mysql/mysql_lex_string.h"
#include "mysql/service_mysql_alloc.h"
#include "mysql/strings/collations.h"
#include "mysql/strings/m_ctype.h"
#include "mysql_version.h" // MYSQL_VERSION_ID
#include "mysqld_error.h"
#include "prealloced_array.h" // Prealloced_array
#include "sql/current_thd.h"
#include "sql/derror.h"
#include "sql/item_func.h"
#include "sql/item_subselect.h"
#include "sql/lex_symbol.h"
#include "sql/lexer_yystype.h"
#include "sql/mysqld.h" // table_alias_charset
#include "sql/nested_join.h"
#include "sql/opt_hints.h"
#include "sql/parse_location.h"
#include "sql/parse_tree_nodes.h" // PT_with_clause
#include "sql/protocol.h"
#include "sql/select_lex_visitor.h"
#include "sql/sp_head.h" // sp_head
#include "sql/sql_admin.h"
#include "sql/sql_base.h"
#include "sql/sql_class.h" // THD
#include "sql/sql_cmd.h"
#include "sql/sql_digest_stream.h"
#include "sql/sql_error.h"
#include "sql/sql_insert.h" // Sql_cmd_insert_base
#include "sql/sql_lex_hash.h"
#include "sql/sql_lex_hints.h"
#include "sql/sql_optimizer.h" // JOIN
#include "sql/sql_parse.h" // add_to_list
#include "sql/sql_plugin.h" // plugin_unlock_list
#include "sql/sql_profile.h"
#include "sql/sql_show.h" // append_identifier
#include "sql/sql_table.h" // primary_key_name
#include "sql/sql_yacc.h"
#include "sql/system_variables.h"
#include "sql/table_function.h"
#include "sql/window.h"
#include "sql_update.h" // Sql_cmd_update
#include "string_with_len.h"
#include "template_utils.h"
class PT_hint_list;
extern int my_hint_parser_parse(THD *thd, Hint_scanner *scanner,
PT_hint_list **ret);
static int lex_one_token(Lexer_yystype *yylval, THD *thd);
/**
LEX_STRING constant for null-string to be used in parser and other places.
*/
const LEX_STRING null_lex_str = {nullptr, 0};
/**
Mapping from enum values in enum_binlog_stmt_unsafe to error codes.
@note The order of the elements of this array must correspond to
the order of elements in enum_binlog_stmt_unsafe.
Todo/fixme Bug#22860121 ER_BINLOG_UNSAFE_* FAMILY OF ERROR CODES IS UNUSED
suggests to turn ER_BINLOG_UNSAFE* to private consts/messages.
*/
const int
Query_tables_list::binlog_stmt_unsafe_errcode[BINLOG_STMT_UNSAFE_COUNT] = {
ER_BINLOG_UNSAFE_LIMIT,
ER_BINLOG_UNSAFE_SYSTEM_TABLE,
ER_BINLOG_UNSAFE_AUTOINC_COLUMNS,
ER_BINLOG_UNSAFE_UDF,
ER_BINLOG_UNSAFE_SYSTEM_VARIABLE,
ER_BINLOG_UNSAFE_SYSTEM_FUNCTION,
ER_BINLOG_UNSAFE_NONTRANS_AFTER_TRANS,
ER_BINLOG_UNSAFE_MULTIPLE_ENGINES_AND_SELF_LOGGING_ENGINE,
ER_BINLOG_UNSAFE_MIXED_STATEMENT,
ER_BINLOG_UNSAFE_INSERT_IGNORE_SELECT,
ER_BINLOG_UNSAFE_INSERT_SELECT_UPDATE,
ER_BINLOG_UNSAFE_WRITE_AUTOINC_SELECT,
ER_BINLOG_UNSAFE_REPLACE_SELECT,
ER_BINLOG_UNSAFE_CREATE_IGNORE_SELECT,
ER_BINLOG_UNSAFE_CREATE_REPLACE_SELECT,
ER_BINLOG_UNSAFE_CREATE_SELECT_AUTOINC,
ER_BINLOG_UNSAFE_UPDATE_IGNORE,
ER_BINLOG_UNSAFE_INSERT_TWO_KEYS,
ER_BINLOG_UNSAFE_AUTOINC_NOT_FIRST,
ER_BINLOG_UNSAFE_FULLTEXT_PLUGIN,
ER_BINLOG_UNSAFE_SKIP_LOCKED,
ER_BINLOG_UNSAFE_NOWAIT,
ER_BINLOG_UNSAFE_XA,
ER_BINLOG_UNSAFE_DEFAULT_EXPRESSION_IN_SUBSTATEMENT,
ER_BINLOG_UNSAFE_ACL_TABLE_READ_IN_DML_DDL,
ER_CREATE_SELECT_WITH_GIPK_DISALLOWED_IN_SBR};
/*
Names of the index hints (for error messages). Keep in sync with
index_hint_type
*/
const char *index_hint_type_name[] = {"IGNORE INDEX", "USE INDEX",
"FORCE INDEX"};
Prepare_error_tracker::~Prepare_error_tracker() {
if (unlikely(thd->is_error())) thd->lex->mark_broken();
}
/**
@note The order of the elements of this array must correspond to
the order of elements in enum_explain_type.
*/
const char *Query_block::type_str[static_cast<int>(
enum_explain_type::EXPLAIN_total)] = {"NONE", "PRIMARY",
"SIMPLE", "DERIVED",
"SUBQUERY", "UNION",
"INTERSECT", "EXCEPT",
"UNION RESULT", "INTERSECT RESULT",
"EXCEPT RESULT", "RESULT" /*unary*/,
"MATERIALIZED"};
Table_ident::Table_ident(Protocol *protocol, const LEX_CSTRING &db_arg,
const LEX_CSTRING &table_arg, bool force)
: table(table_arg), sel(nullptr), table_function(nullptr) {
if (!force && protocol->has_client_capability(CLIENT_NO_SCHEMA))
db = NULL_CSTR;
else
db = db_arg;
}
void lex_free() { // Call this when daemon ends
DBUG_TRACE;
}
void st_parsing_options::reset() {
allows_variable = true;
allows_select_into = true;
}
/**
Cleans slave connection info.
*/
void struct_replica_connection::reset() {
user = nullptr;
password = nullptr;
plugin_auth = nullptr;
plugin_dir = nullptr;
}
/**
Perform initialization of Lex_input_stream instance.
Basically, a buffer for a pre-processed query. This buffer should be large
enough to keep a multi-statement query. The allocation is done once in
Lex_input_stream::init() in order to prevent memory pollution when
the server is processing large multi-statement queries.
*/
bool Lex_input_stream::init(THD *thd, const char *buff, size_t length) {
DBUG_EXECUTE_IF("bug42064_simulate_oom",
DBUG_SET("+d,simulate_out_of_memory"););
query_charset = thd->charset();
m_cpp_buf = (char *)thd->alloc(length + 1);
DBUG_EXECUTE_IF("bug42064_simulate_oom",
DBUG_SET("-d,bug42064_simulate_oom"););
if (m_cpp_buf == nullptr) return true;
m_thd = thd;
reset(buff, length);
return false;
}
/**
Prepare Lex_input_stream instance state for use for handling next SQL
statement.
It should be called between two statements in a multi-statement query.
The operation resets the input stream to the beginning-of-parse state,
but does not reallocate m_cpp_buf.
*/
void Lex_input_stream::reset(const char *buffer, size_t length) {
yylineno = 1;
yytoklen = 0;
yylval = nullptr;
lookahead_token = grammar_selector_token;
static Lexer_yystype dummy_yylval;
lookahead_yylval = &dummy_yylval;
skip_digest = false;
/*
Lex_input_stream modifies the query string in one special case (sic!).
yyUnput() modifises the string when patching version comments.
This is done to prevent newer slaves from executing a different
statement than older masters.
For now, cast away const here. This means that e.g. SHOW PROCESSLIST
can see partially patched query strings. It would be better if we
could replicate the query string as is and have the slave take the
master version into account.
*/
m_ptr = const_cast<char *>(buffer);
m_tok_start = nullptr;
m_tok_end = nullptr;
m_end_of_query = buffer + length;
m_buf = buffer;
m_buf_length = length;
m_echo = true;
m_cpp_tok_start = nullptr;
m_cpp_tok_end = nullptr;
m_body_utf8 = nullptr;
m_cpp_utf8_processed_ptr = nullptr;
next_state = MY_LEX_START;
found_semicolon = nullptr;
ignore_space = m_thd->variables.sql_mode & MODE_IGNORE_SPACE;
stmt_prepare_mode = false;
multi_statements = true;
in_comment = NO_COMMENT;
m_underscore_cs = nullptr;
m_cpp_ptr = m_cpp_buf;
}
/**
The operation is called from the parser in order to
1) designate the intention to have utf8 body;
1) Indicate to the lexer that we will need a utf8 representation of this
statement;
2) Determine the beginning of the body.
@param thd Thread context.
@param begin_ptr Pointer to the start of the body in the pre-processed
buffer.
*/
void Lex_input_stream::body_utf8_start(THD *thd, const char *begin_ptr) {
assert(begin_ptr);
assert(m_cpp_buf <= begin_ptr && begin_ptr <= m_cpp_buf + m_buf_length);
const size_t body_utf8_length =
(m_buf_length / thd->variables.character_set_client->mbminlen) *
my_charset_utf8mb3_bin.mbmaxlen;
m_body_utf8 = (char *)thd->alloc(body_utf8_length + 1);
m_body_utf8_ptr = m_body_utf8;
*m_body_utf8_ptr = 0;
m_cpp_utf8_processed_ptr = begin_ptr;
}
/**
@brief The operation appends unprocessed part of pre-processed buffer till
the given pointer (ptr) and sets m_cpp_utf8_processed_ptr to end_ptr.
The idea is that some tokens in the pre-processed buffer (like character
set introducers) should be skipped.
Example:
CPP buffer: SELECT 'str1', _latin1 'str2';
m_cpp_utf8_processed_ptr -- points at the "SELECT ...";
In order to skip "_latin1", the following call should be made:
body_utf8_append(<pointer to "_latin1 ...">, <pointer to " 'str2'...">)
@param ptr Pointer in the pre-processed buffer, which specifies the
end of the chunk, which should be appended to the utf8
body.
@param end_ptr Pointer in the pre-processed buffer, to which
m_cpp_utf8_processed_ptr will be set in the end of the
operation.
*/
void Lex_input_stream::body_utf8_append(const char *ptr, const char *end_ptr) {
assert(m_cpp_buf <= ptr && ptr <= m_cpp_buf + m_buf_length);
assert(m_cpp_buf <= end_ptr && end_ptr <= m_cpp_buf + m_buf_length);
if (!m_body_utf8) return;
if (m_cpp_utf8_processed_ptr >= ptr) return;
const size_t bytes_to_copy = ptr - m_cpp_utf8_processed_ptr;
memcpy(m_body_utf8_ptr, m_cpp_utf8_processed_ptr, bytes_to_copy);
m_body_utf8_ptr += bytes_to_copy;
*m_body_utf8_ptr = 0;
m_cpp_utf8_processed_ptr = end_ptr;
}
/**
The operation appends unprocessed part of the pre-processed buffer till
the given pointer (ptr) and sets m_cpp_utf8_processed_ptr to ptr.
@param ptr Pointer in the pre-processed buffer, which specifies the end
of the chunk, which should be appended to the utf8 body.
*/
void Lex_input_stream::body_utf8_append(const char *ptr) {
body_utf8_append(ptr, ptr);
}
/**
The operation converts the specified text literal to the utf8 and appends
the result to the utf8-body.
@param thd Thread context.
@param txt Text literal.
@param txt_cs Character set of the text literal.
@param end_ptr Pointer in the pre-processed buffer, to which
m_cpp_utf8_processed_ptr will be set in the end of the
operation.
*/
void Lex_input_stream::body_utf8_append_literal(THD *thd, const LEX_STRING *txt,
const CHARSET_INFO *txt_cs,
const char *end_ptr) {
if (!m_cpp_utf8_processed_ptr) return;
LEX_STRING utf_txt;
if (!my_charset_same(txt_cs, &my_charset_utf8mb3_general_ci)) {
thd->convert_string(&utf_txt, &my_charset_utf8mb3_general_ci, txt->str,
txt->length, txt_cs);
} else {
utf_txt.str = txt->str;
utf_txt.length = txt->length;
}
/* NOTE: utf_txt.length is in bytes, not in symbols. */
memcpy(m_body_utf8_ptr, utf_txt.str, utf_txt.length);
m_body_utf8_ptr += utf_txt.length;
*m_body_utf8_ptr = 0;
m_cpp_utf8_processed_ptr = end_ptr;
}
void Lex_input_stream::add_digest_token(uint token, Lexer_yystype *yylval) {
if (m_digest != nullptr) {
m_digest = digest_add_token(m_digest, token, yylval);
}
}
void Lex_input_stream::reduce_digest_token(uint token_left, uint token_right) {
if (m_digest != nullptr) {
m_digest = digest_reduce_token(m_digest, token_left, token_right);
}
}
void LEX::assert_ok_set_current_query_block() {
// (2) Only owning thread could change m_current_query_block
// (1) bypass for bootstrap and "new THD"
assert(!current_thd || !thd || //(1)
thd == current_thd); //(2)
}
LEX::~LEX() {
destroy_query_tables_list();
plugin_unlock_list(nullptr, plugins.begin(), plugins.size());
unit = nullptr; // Created in mem_root - no destructor
query_block = nullptr;
m_current_query_block = nullptr;
assert(m_secondary_engine_context == nullptr);
}
/**
Reset a LEX object so that it is ready for a new query preparation
and execution.
Pointers to query expression and query block objects are set to NULL.
This is correct, as they point into a mem_root that has been recycled.
*/
void LEX::reset() {
// CREATE VIEW
create_view_mode = enum_view_create_mode::VIEW_CREATE_NEW;
create_view_algorithm = VIEW_ALGORITHM_UNDEFINED;
create_view_suid = true;
context_stack.clear();
unit = nullptr;
query_block = nullptr;
m_current_query_block = nullptr;
all_query_blocks_list = nullptr;
bulk_insert_row_cnt = 0;
purge_value_list.clear();
kill_value_list.clear();
set_var_list.clear();
param_list.clear();
prepared_stmt_params.clear();
context_analysis_only = 0;
safe_to_cache_query = true;
insert_table = nullptr;
insert_table_leaf = nullptr;
parsing_options.reset();
alter_info = nullptr;
part_info = nullptr;
duplicates = DUP_ERROR;
ignore = false;
spname = nullptr;
sphead = nullptr;
set_sp_current_parsing_ctx(nullptr);
m_sql_cmd = nullptr;
query_tables = nullptr;
reset_query_tables_list(false);
expr_allows_subquery = true;
use_only_table_context = false;
contains_plaintext_password = false;
keep_diagnostics = DA_KEEP_NOTHING;
m_statement_options = 0;
next_binlog_file_nr = 0;
name.str = nullptr;
name.length = 0;
event_parse_data = nullptr;
profile_options = PROFILE_NONE;
select_number = 0;
allow_sum_func = 0;
m_deny_window_func = 0;
m_subquery_to_derived_is_impossible = false;
in_sum_func = nullptr;
create_info = nullptr;
server_options.reset();
explain_format = nullptr;
is_explain_analyze = false;
set_using_hypergraph_optimizer(false);
is_lex_started = true;
reset_replica_info.all = false;
mi.channel = nullptr;
wild = nullptr;
mark_broken(false);
reset_exec_started();
max_execution_time = 0;
reparse_common_table_expr_at = 0;
reparse_derived_table_condition = false;
opt_hints_global = nullptr;
binlog_need_explicit_defaults_ts = false;
m_extended_show = false;
option_type = OPT_DEFAULT;
check_opt = HA_CHECK_OPT();
clear_privileges();
grant_as.cleanup();
alter_user_attribute = enum_alter_user_attribute::ALTER_USER_COMMENT_NOT_USED;
m_was_replication_command_executed = false;
grant_if_exists = false;
ignore_unknown_user = false;
reset_rewrite_required();
set_execute_only_in_secondary_engine(
/*execute_only_in_secondary_engine_param=*/false, SUPPORTED_IN_PRIMARY);
set_execute_only_in_hypergraph_optimizer(
/*execute_in_hypergraph_optimizer_param=*/false,
SUPPORTED_IN_BOTH_OPTIMIZERS);
}
/**
Call lex_start() before every query that is to be prepared and executed.
Because of this, it's critical not to do too many things here. (We already
do too much)
The function creates a query_block and a query_block_query_expression object.
These objects should rather be created by the parser bottom-up.
*/
bool lex_start(THD *thd) {
DBUG_TRACE;
LEX *lex = thd->lex;
lex->thd = thd;
lex->reset();
// Initialize the cost model to be used for this query
thd->init_cost_model();
const bool status = lex->new_top_level_query();
assert(lex->current_query_block() == nullptr);
lex->m_current_query_block = lex->query_block;
assert(lex->m_IS_table_stats.is_valid() == false);
assert(lex->m_IS_tablespace_stats.is_valid() == false);
return status;
}
/**
Call this function after preparation and execution of a query.
*/
void lex_end(LEX *lex) {
DBUG_TRACE;
DBUG_PRINT("enter", ("lex: %p", lex));
/* release used plugins */
lex->release_plugins();
sp_head::destroy(lex->sphead);
lex->sphead = nullptr;
}
void LEX::release_plugins() {
if (!plugins.empty()) /* No function call and no mutex if no plugins. */
{
plugin_unlock_list(nullptr, plugins.begin(), plugins.size());
plugins.clear();
}
}
/**
Clear execution state for a statement after it has been prepared or executed,
and before it is (re-)executed.
*/
void LEX::clear_execution() {
// Clear execution state for all query expressions:
for (Query_block *sl = all_query_blocks_list; sl;
sl = sl->next_select_in_list())
sl->master_query_expression()->clear_execution();
set_current_query_block(query_block);
reset_exec_started();
/*
m_view_ctx_list contains all the view tables view_ctx objects and must
be emptied now since it's going to be re-populated below as we reiterate
over all query_tables and call Table_ref::prepare_security().
*/
thd->m_view_ctx_list.clear();
// Reset all table references so that they can be bound with new TABLEs
/*
NOTE: We should reset whole table list here including all tables added
by prelocking algorithm (it is not a problem for substatements since
they have their own table list).
Another note: this loop uses query_tables so does not see TABLE_LISTs
which represent join nests.
*/
for (Table_ref *tr = query_tables; tr != nullptr; tr = tr->next_global)
tr->reset();
}
Query_block *LEX::new_empty_query_block() {
Query_block *select =
new (thd->mem_root) Query_block(thd->mem_root, nullptr, nullptr);
if (select == nullptr) return nullptr; /* purecov: inspected */
select->parent_lex = this;
return select;
}
Query_expression *LEX::create_query_expr_and_block(
THD *thd, Query_block *current_query_block, Item *where, Item *having,
enum_parsing_context ctx) {
if (current_query_block != nullptr &&
current_query_block->nest_level >= MAX_SELECT_NESTING) {
my_error(ER_TOO_HIGH_LEVEL_OF_NESTING_FOR_SELECT, MYF(0));
return nullptr;
}
auto *const new_expression = new (thd->mem_root) Query_expression(ctx);
if (new_expression == nullptr) return nullptr;
auto *const new_query_block =
new (thd->mem_root) Query_block(thd->mem_root, where, having);
if (new_query_block == nullptr) return nullptr;
// Link the new query expression below the current query block, if any
if (current_query_block != nullptr)
new_expression->include_down(this, current_query_block);
new_query_block->include_down(this, new_expression);
new_query_block->parent_lex = this;
new_query_block->include_in_global(&this->all_query_blocks_list);
// Set root query_term a priori. It is usually set later by
// Parse_context::finalize_query_expression. This is necessary since it
// doesn't always happen during server bootstrap of the dictionary, e.g. in
// View_metadata_updater_context which creates a top level query whose
// Query_expression/Query_block is not parsed/contextualized, so we never
// call finalize_query_expression in the usual way, after contextualization
// of PT_query_expression.
new_expression->set_query_term(new_query_block);
return new_expression;
}
/**
Create new query_block_query_expression and query_block objects for a query
block, which can be either a top-level query or a subquery. For the second and
subsequent query block of a UNION query, use LEX::new_set_operation_query()
instead.
Set the new query_block as the current query_block of the LEX object.
@param curr_query_block current query block, NULL if an outer-most
query block should be created.
@return new query specification if successful, NULL if error
*/
Query_block *LEX::new_query(Query_block *curr_query_block) {
DBUG_TRACE;
Name_resolution_context *outer_context = current_context();
const enum_parsing_context parsing_place =
curr_query_block != nullptr ? curr_query_block->parsing_place : CTX_NONE;
Query_expression *const new_query_expression = create_query_expr_and_block(
thd, curr_query_block, nullptr, nullptr, parsing_place);
if (new_query_expression == nullptr) return nullptr;
Query_block *const new_query_block =
new_query_expression->first_query_block();
if (new_query_block->set_context(nullptr)) return nullptr;
/*
Assume that a subquery has an outer name resolution context
(even a non-lateral derived table may have outer references).
When we come here for a view, it's when we parse the view (in
open_tables()): we parse it as a standalone query, where parsing_place
is CTX_NONE, so the outer context is set to nullptr. Then we'll resolve the
view's query (thus, using no outer context). Later we may merge the
view's query, but that happens after resolution, so there's no chance that
a view "looks outside" (uses outer references). An assertion in
resolve_derived() checks this.
*/
if (parsing_place == CTX_NONE) // Outer-most query block
{
} else if (parsing_place == CTX_INSERT_UPDATE &&
curr_query_block->master_query_expression()->is_set_operation()) {
/*
Outer references are not allowed for
subqueries in INSERT ... ON DUPLICATE KEY UPDATE clauses,
when the outer query expression is a UNION.
*/
assert(new_query_block->context.outer_context == nullptr);
} else {
new_query_block->context.outer_context = outer_context;
}
/*
in subquery is SELECT query and we allow resolution of names in SELECT
list
*/
new_query_block->context.resolve_in_select_list = true;
DBUG_PRINT("outer_field", ("ctx %p <-> SL# %d", &new_query_block->context,
new_query_block->select_number));
return new_query_block;
}
/**
Create new query_block object for all branches of a UNION, EXCEPT or INTERSECT
except the left-most one.
Set the new query_block as the current query_block of the LEX object.
@param curr_query_block current query specification
@return new query specification if successful, NULL if an error occurred.
*/
Query_block *LEX::new_set_operation_query(Query_block *curr_query_block) {
DBUG_TRACE;
assert(unit != nullptr && query_block != nullptr);
// Is this the outer-most query expression?
bool const outer_most = curr_query_block->master_query_expression() == unit;
/*
Only the last SELECT can have INTO. Since the grammar won't allow INTO in
a nested SELECT, we make this check only when creating a query block on
the outer-most level:
*/
if (outer_most && result) {
my_error(ER_MISPLACED_INTO, MYF(0));
return nullptr;
}
Query_block *const select = new_empty_query_block();
if (!select) return nullptr; /* purecov: inspected */
select->include_neighbour(this, curr_query_block);
Query_expression *const sel_query_expression =
select->master_query_expression();
if (select->set_context(
sel_query_expression->first_query_block()->context.outer_context))
return nullptr; /* purecov: inspected */
select->include_in_global(&all_query_blocks_list);
/*
By default we assume that this is a regular subquery, in which resolution
of names in SELECT list is allowed.
*/
select->context.resolve_in_select_list = true;
return select;
}
Query_block *Query_expression::create_post_processing_block(
Query_term_set_op *term) {
Query_block *first_qb = first_query_block();
DBUG_TRACE;
Query_block *const qb = first_qb->parent_lex->new_empty_query_block();
if (qb == nullptr) return nullptr; /* purecov: inspected */
qb->include_standalone(this);
qb->select_number = INT_MAX;
qb->linkage = GLOBAL_OPTIONS_TYPE;
qb->select_limit = nullptr;
qb->set_context(first_qb->context.outer_context);
/* allow item list resolving in fake select for ORDER BY */
qb->context.resolve_in_select_list = true;
qb->no_table_names_allowed =
!term->is_unary() || first_qb->is_ordered() || first_qb->has_limit();
first_qb->parent_lex->pop_context();
return qb;
}
bool Query_expression::is_leaf_block(Query_block *qb) {
for (Query_block *q = first_query_block(); q != nullptr;
q = q->next_query_block()) {
if (q == qb) return true;
}
return false;
}
/**
Given a LEX object, create a query expression object
(query_block_query_expression) and a query block object (query_block).
@return false if successful, true if error
*/
bool LEX::new_top_level_query() {
DBUG_TRACE;
// Assure that the LEX does not contain any query expression already
assert(unit == nullptr && query_block == nullptr);
// Check for the special situation when using INTO OUTFILE and LOAD DATA.
assert(result == nullptr);
query_block = new_query(nullptr);
if (query_block == nullptr) return true; /* purecov: inspected */
unit = query_block->master_query_expression();
return false;
}
/**
Initialize a LEX object, a query expression object
(query_block_query_expression) and a query block object (query_block). All
objects are passed as pointers so they can be stack-allocated. The purpose of
this structure is for short-lived procedures that need a LEX and a query block
object.
Do not extend the struct with more query objects after creation.
The struct can be abandoned after use, no cleanup is needed.
@param sel_query_expression Pointer to the query expression object
@param select Pointer to the query block object
*/
void LEX::new_static_query(Query_expression *sel_query_expression,
Query_block *select)
{
DBUG_TRACE;
reset();
assert(unit == nullptr && query_block == nullptr &&
current_query_block() == nullptr);
select->parent_lex = this;
select->include_down(this, sel_query_expression);
select->include_in_global(&all_query_blocks_list);
(void)select->set_context(nullptr);
query_block = select;
unit = sel_query_expression;
set_current_query_block(select);
select->context.resolve_in_select_list = true;
}
/**
Check whether preparation state for prepared statement is invalid.
Preparation state amy become invalid if a repreparation is forced,
e.g because of invalid metadata, and that repreparation fails.
@returns true if preparation state is invalid, false otherwise.
*/
bool LEX::check_preparation_invalid(THD *thd_arg) {
DBUG_TRACE;
if (unlikely(is_broken())) {
// Force a Reprepare, to get a fresh LEX
if (ask_to_reprepare(thd_arg)) return true;
}
return false;
}
Yacc_state::~Yacc_state() {
if (yacc_yyss) {
my_free(yacc_yyss);
my_free(yacc_yyvs);
my_free(yacc_yyls);
}
}
static bool consume_optimizer_hints(Lex_input_stream *lip) {
const my_lex_states *state_map = lip->query_charset->state_maps->main_map;
int whitespace = 0;
uchar c = lip->yyPeek();
size_t newlines = 0;
for (; state_map[c] == MY_LEX_SKIP;
whitespace++, c = lip->yyPeekn(whitespace)) {
if (c == '\n') newlines++;
}
if (lip->yyPeekn(whitespace) == '/' && lip->yyPeekn(whitespace + 1) == '*' &&
lip->yyPeekn(whitespace + 2) == '+') {
lip->yylineno += newlines;
lip->yySkipn(whitespace); // skip whitespace
Hint_scanner hint_scanner(lip->m_thd, lip->yylineno, lip->get_ptr(),
lip->get_end_of_query() - lip->get_ptr(),
lip->m_digest);
PT_hint_list *hint_list = nullptr;
const int rc = my_hint_parser_parse(lip->m_thd, &hint_scanner, &hint_list);
if (rc == 2) return true; // Bison's internal OOM error
if (rc == 1) {
/*
This branch is for 2 cases:
1. YYABORT in the hint parser grammar (we use it to process OOM errors),
2. open commentary error.
*/
lip->start_token(); // adjust error message text pointer to "/*+"
return true;
}
lip->yylineno = hint_scanner.get_lineno();
lip->yySkipn(hint_scanner.get_ptr() - lip->get_ptr());
lip->yylval->optimizer_hints = hint_list; // NULL in case of syntax error
lip->m_digest = hint_scanner.get_digest(); // NULL is digest buf. is full.
return false;
} else
return false;
}
static int find_keyword(Lex_input_stream *lip, uint len, bool function) {
const char *tok = lip->get_tok_start();
const SYMBOL *symbol =
function ? Lex_hash::sql_keywords_and_funcs.get_hash_symbol(tok, len)
: Lex_hash::sql_keywords.get_hash_symbol(tok, len);
if (symbol) {
lip->yylval->keyword.symbol = symbol;
lip->yylval->keyword.str = const_cast<char *>(tok);
lip->yylval->keyword.length = len;
if ((symbol->tok == NOT_SYM) &&
(lip->m_thd->variables.sql_mode & MODE_HIGH_NOT_PRECEDENCE))
return NOT2_SYM;
if ((symbol->tok == OR_OR_SYM) &&
!(lip->m_thd->variables.sql_mode & MODE_PIPES_AS_CONCAT)) {
push_deprecated_warn(lip->m_thd, "|| as a synonym for OR", "OR");
return OR2_SYM;
}
lip->yylval->optimizer_hints = nullptr;
if (symbol->group & SG_HINTABLE_KEYWORDS) {
lip->add_digest_token(symbol->tok, lip->yylval);
if (consume_optimizer_hints(lip)) return ABORT_SYM;
lip->skip_digest = true;
}
return symbol->tok;
}
return 0;
}
/*
Check if name is a keyword
SYNOPSIS
is_keyword()
name checked name (must not be empty)
len length of checked name
RETURN VALUES
0 name is a keyword
1 name isn't a keyword
*/
bool is_keyword(const char *name, size_t len) {
assert(len != 0);
return Lex_hash::sql_keywords.get_hash_symbol(name, len) != nullptr;
}
/**
Check if name is a sql function
@param name checked name
@return is this a native function or not
@retval 0 name is a function
@retval 1 name isn't a function
*/
bool is_lex_native_function(const LEX_STRING *name) {
assert(name != nullptr);
return Lex_hash::sql_keywords_and_funcs.get_hash_symbol(
name->str, (uint)name->length) != nullptr;
}
/* make a copy of token before ptr and set yytoklen */
static LEX_STRING get_token(Lex_input_stream *lip, uint skip, uint length) {
LEX_STRING tmp;
lip->yyUnget(); // ptr points now after last token char
tmp.length = lip->yytoklen = length;
tmp.str = lip->m_thd->strmake(lip->get_tok_start() + skip, tmp.length);
lip->m_cpp_text_start = lip->get_cpp_tok_start() + skip;
lip->m_cpp_text_end = lip->m_cpp_text_start + tmp.length;
return tmp;
}
/*
todo:
There are no dangerous charsets in mysql for function
get_quoted_token yet. But it should be fixed in the
future to operate multichar strings (like ucs2)
*/
static LEX_STRING get_quoted_token(Lex_input_stream *lip, uint skip,
uint length, char quote) {
LEX_STRING tmp;
const char *from, *end;
char *to;
lip->yyUnget(); // ptr points now after last token char
tmp.length = lip->yytoklen = length;
tmp.str = (char *)lip->m_thd->alloc(tmp.length + 1);
from = lip->get_tok_start() + skip;
to = tmp.str;
end = to + length;
lip->m_cpp_text_start = lip->get_cpp_tok_start() + skip;
lip->m_cpp_text_end = lip->m_cpp_text_start + length;
for (; to != end;) {
if ((*to++ = *from++) == quote) {
from++; // Skip double quotes
lip->m_cpp_text_start++;
}
}
*to = 0; // End null for safety
return tmp;
}
/*
Return an unescaped text literal without quotes
Fix sometimes to do only one scan of the string
*/
static char *get_text(Lex_input_stream *lip, int pre_skip, int post_skip) {
uchar c, sep;
uint found_escape = 0;
const CHARSET_INFO *cs = lip->m_thd->charset();
lip->tok_bitmap = 0;
sep = lip->yyGetLast(); // String should end with this
while (!lip->eof()) {
c = lip->yyGet();
lip->tok_bitmap |= c;
{
int l;
if (use_mb(cs) &&
(l = my_ismbchar(cs, lip->get_ptr() - 1, lip->get_end_of_query()))) {
lip->skip_binary(l - 1);
continue;
}
}
if (c == '\\' && !(lip->m_thd->variables.sql_mode &
MODE_NO_BACKSLASH_ESCAPES)) { // Escaped character
found_escape = 1;
if (lip->eof()) return nullptr;
lip->yySkip();
} else if (c == sep) {
if (c == lip->yyGet()) // Check if two separators in a row
{
found_escape = 1; // duplicate. Remember for delete
continue;
} else
lip->yyUnget();
/* Found end. Unescape and return string */
const char *str, *end;
char *start;
str = lip->get_tok_start();
end = lip->get_ptr();
/* Extract the text from the token */
str += pre_skip;
end -= post_skip;
assert(end >= str);
if (!(start =
static_cast<char *>(lip->m_thd->alloc((uint)(end - str) + 1))))
return const_cast<char *>(""); // MEM_ROOT has set error flag
lip->m_cpp_text_start = lip->get_cpp_tok_start() + pre_skip;
lip->m_cpp_text_end = lip->get_cpp_ptr() - post_skip;
if (!found_escape) {
lip->yytoklen = (uint)(end - str);
memcpy(start, str, lip->yytoklen);
start[lip->yytoklen] = 0;
} else {
char *to;
for (to = start; str != end; str++) {
int l;
if (use_mb(cs) && (l = my_ismbchar(cs, str, end))) {
while (l--) *to++ = *str++;
str--;
continue;
}
if (!(lip->m_thd->variables.sql_mode & MODE_NO_BACKSLASH_ESCAPES) &&
*str == '\\' && str + 1 != end) {
switch (*++str) {
case 'n':
*to++ = '\n';
break;
case 't':
*to++ = '\t';
break;
case 'r':
*to++ = '\r';
break;
case 'b':
*to++ = '\b';
break;
case '0':
*to++ = 0; // Ascii null
break;
case 'Z': // ^Z must be escaped on Win32
*to++ = '\032';
break;
case '_':
case '%':
*to++ = '\\'; // remember prefix for wildcard
[[fallthrough]];
default:
*to++ = *str;
break;
}
} else if (*str == sep)
*to++ = *str++; // Two ' or "
else
*to++ = *str;
}
*to = 0;
lip->yytoklen = (uint)(to - start);
}
return start;
}
}
return nullptr; // unexpected end of query
}
/**
Get the text literal between dollar quotes.
A dollar quote is of the form $tag$, where tag is zero or more characters.
The same characters that are permitted for unquoted identifiers,
except dollar, may be used for the tag.
That is, basic ASCII letters, digits 0-9, underscore,
and any multibyte UTF8 characters.
@param lip The input stream. When called, the current position
is right after the initial dollar character
@param tag_len The length of the tag
@returns The text literal without dollar quotes
*/
static LEX_CSTRING get_dollar_quoted_text(Lex_input_stream *lip, int tag_len) {
assert(lip->yyGetLast() == '$' && !lip->eof());
const CHARSET_INFO *cs = lip->m_thd->charset();
const uchar *left_delim = pointer_cast<const uchar *>(lip->get_tok_start());
lip->yySkipn(tag_len + 1); // skip beyond the 2nd '$'
enum { TEXT_BODY, RIGHT_DELIMITER } state = TEXT_BODY;
const char *body_start = lip->get_cpp_ptr();
int delim_pos = 0; // Position during delimiter matching
while (!lip->eof()) {
const uchar c = lip->yyGet();
switch (state) {
case TEXT_BODY:
if (use_mb(cs)) {
int mb_len =
my_ismbchar(cs, lip->get_ptr() - 1, lip->get_end_of_query());
if (mb_len > 1) {
lip->skip_binary(mb_len - 1);
break;
}
}
if (c == '$') {
state = RIGHT_DELIMITER;
}
break;
case RIGHT_DELIMITER:
if (c == left_delim[++delim_pos]) {
if (delim_pos == tag_len + 1) {
size_t length = lip->get_cpp_ptr() - body_start - tag_len - 2;
return LEX_CSTRING{body_start, length};
}
} else { // Not a right delimiter after all
state = TEXT_BODY;
delim_pos = 0;
}
break;
}
}
assert(lip->eof());
return LEX_CSTRING{}; // error: unterminated text
}
uint Lex_input_stream::get_lineno(const char *raw_ptr) const {
assert(m_buf <= raw_ptr && raw_ptr <= m_end_of_query);
if (!(m_buf <= raw_ptr && raw_ptr <= m_end_of_query)) return 1;
uint ret = 1;
const CHARSET_INFO *cs = m_thd->charset();
for (const char *c = m_buf; c < raw_ptr; c++) {
uint mb_char_len;
if (use_mb(cs) && (mb_char_len = my_ismbchar(cs, c, m_end_of_query))) {
c += mb_char_len - 1; // skip the rest of the multibyte character
continue; // we don't expect '\n' there
}
if (*c == '\n') ret++;
}
return ret;
}
Partition_expr_parser_state::Partition_expr_parser_state()
: Parser_state(GRAMMAR_SELECTOR_PART), result(nullptr) {}
Gcol_expr_parser_state::Gcol_expr_parser_state()
: Parser_state(GRAMMAR_SELECTOR_GCOL), result(nullptr) {}
Expression_parser_state::Expression_parser_state()
: Parser_state(GRAMMAR_SELECTOR_EXPR), result(nullptr) {}
Common_table_expr_parser_state::Common_table_expr_parser_state()
: Parser_state(GRAMMAR_SELECTOR_CTE), result(nullptr) {}
Derived_expr_parser_state::Derived_expr_parser_state()
: Parser_state(GRAMMAR_SELECTOR_DERIVED_EXPR), result(nullptr) {}
/*
** Calc type of integer; long integer, longlong integer or real.
** Returns smallest type that match the string.
** When using unsigned long long values the result is converted to a real
** because else they will be unexpected sign changes because all calculation
** is done with longlong or double.
*/
static const char *long_str = "2147483647";
static const uint long_len = 10;
static const char *signed_long_str = "-2147483648";
static const char *longlong_str = "9223372036854775807";
static const uint longlong_len = 19;
static const char *signed_longlong_str = "-9223372036854775808";
static const uint signed_longlong_len = 19;
static const char *unsigned_longlong_str = "18446744073709551615";
static const uint unsigned_longlong_len = 20;
static inline uint int_token(const char *str, uint length) {
if (length < long_len) // quick normal case
return NUM;
bool neg = false;
if (*str == '+') // Remove sign and pre-zeros
{
str++;
length--;
} else if (*str == '-') {
str++;
length--;
neg = true;
}
while (*str == '0' && length) {
str++;
length--;
}
if (length < long_len) return NUM;
uint smaller, bigger;
const char *cmp;
if (neg) {
if (length == long_len) {
cmp = signed_long_str + 1;
smaller = NUM; // If <= signed_long_str
bigger = LONG_NUM; // If >= signed_long_str
} else if (length < signed_longlong_len)
return LONG_NUM;
else if (length > signed_longlong_len)
return DECIMAL_NUM;
else {
cmp = signed_longlong_str + 1;
smaller = LONG_NUM; // If <= signed_longlong_str
bigger = DECIMAL_NUM;
}
} else {
if (length == long_len) {
cmp = long_str;
smaller = NUM;
bigger = LONG_NUM;
} else if (length < longlong_len)
return LONG_NUM;
else if (length > longlong_len) {
if (length > unsigned_longlong_len) return DECIMAL_NUM;
cmp = unsigned_longlong_str;
smaller = ULONGLONG_NUM;
bigger = DECIMAL_NUM;
} else {
cmp = longlong_str;
smaller = LONG_NUM;
bigger = ULONGLONG_NUM;
}
}
while (*cmp && *cmp++ == *str++)
;
return ((uchar)str[-1] <= (uchar)cmp[-1]) ? smaller : bigger;
}
/**
Given a stream that is advanced to the first contained character in
an open comment, consume the comment. Optionally, if we are allowed,
recurse so that we understand comments within this current comment.
At this level, we do not support version-condition comments. We might
have been called with having just passed one in the stream, though. In
that case, we probably want to tolerate mundane comments inside. Thus,
the case for recursion.
@retval Whether EOF reached before comment is closed.
*/
static bool consume_comment(Lex_input_stream *lip,
int remaining_recursions_permitted) {
// only one level of nested comments are allowed
assert(remaining_recursions_permitted == 0 ||
remaining_recursions_permitted == 1);
uchar c;
while (!lip->eof()) {
c = lip->yyGet();
if (remaining_recursions_permitted == 1) {
if ((c == '/') && (lip->yyPeek() == '*')) {
push_warning(
lip->m_thd, Sql_condition::SL_WARNING,
ER_WARN_DEPRECATED_SYNTAX_NO_REPLACEMENT,
ER_THD(lip->m_thd, ER_WARN_DEPRECATED_NESTED_COMMENT_SYNTAX));
lip->yyUnput('('); // Replace nested "/*..." with "(*..."
lip->yySkip(); // and skip "("
lip->yySkip(); /* Eat asterisk */
if (consume_comment(lip, 0)) return true;
lip->yyUnput(')'); // Replace "...*/" with "...*)"
lip->yySkip(); // and skip ")"
continue;
}
}
if (c == '*') {
if (lip->yyPeek() == '/') {
lip->yySkip(); /* Eat slash */
return false;
}
}
if (c == '\n') lip->yylineno++;
}
return true;
}
/**
yylex() function implementation for the main parser
@param [out] yacc_yylval semantic value of the token being parsed (yylval)
@param [out] yylloc "location" of the token being parsed (yylloc)
@param thd THD
@return token number
@note
my_sql_parser_lex remember the following states from the
following my_sql_parser_lex():
- MY_LEX_END Found end of query
*/
int my_sql_parser_lex(MY_SQL_PARSER_STYPE *yacc_yylval, POS *yylloc, THD *thd) {
auto *yylval = reinterpret_cast<Lexer_yystype *>(yacc_yylval);
Lex_input_stream *lip = &thd->m_parser_state->m_lip;
int token;
if (thd->is_error()) {
if (thd->get_parser_da()->has_sql_condition(ER_CAPACITY_EXCEEDED))
return ABORT_SYM;
}
if (lip->lookahead_token >= 0) {
/*
The next token was already parsed in advance,
return it.
*/
token = lip->lookahead_token;
lip->lookahead_token = -1;
*yylval = *(lip->lookahead_yylval);
yylloc->cpp.start = lip->get_cpp_tok_start();
yylloc->cpp.end = lip->get_cpp_ptr();
yylloc->raw.start = lip->get_tok_start();
yylloc->raw.end = lip->get_ptr();
lip->lookahead_yylval = nullptr;
lip->add_digest_token(token, yylval);
return token;
}
token = lex_one_token(yylval, thd);
yylloc->cpp.start = lip->get_cpp_tok_start();
yylloc->raw.start = lip->get_tok_start();
switch (token) {
case WITH:
/*
Parsing 'WITH' 'ROLLUP' requires 2 look ups,
which makes the grammar LALR(2).
Replace by a single 'WITH_ROLLUP' token,
to transform the grammar into a LALR(1) grammar,
which sql_yacc.yy can process.
*/
token = lex_one_token(yylval, thd);
switch (token) {
case ROLLUP_SYM:
yylloc->cpp.end = lip->get_cpp_ptr();
yylloc->raw.end = lip->get_ptr();
lip->add_digest_token(WITH_ROLLUP_SYM, yylval);
return WITH_ROLLUP_SYM;
default:
/*
Save the token following 'WITH'
*/
lip->lookahead_yylval = lip->yylval;
lip->yylval = nullptr;
lip->lookahead_token = token;
yylloc->cpp.end = lip->get_cpp_ptr();
yylloc->raw.end = lip->get_ptr();
lip->add_digest_token(WITH, yylval);
return WITH;
}
break;
}
yylloc->cpp.end = lip->get_cpp_ptr();
yylloc->raw.end = lip->get_ptr();
if (!lip->skip_digest) lip->add_digest_token(token, yylval);
lip->skip_digest = false;
return token;
}
static int lex_one_token(Lexer_yystype *yylval, THD *thd) {
uchar c = 0;
bool comment_closed;
int tokval, result_state;
uint length;
enum my_lex_states state;
Lex_input_stream *lip = &thd->m_parser_state->m_lip;
const CHARSET_INFO *cs = thd->charset();
const my_lex_states *state_map = cs->state_maps->main_map;
const uchar *ident_map = cs->ident_map;
lip->yylval = yylval; // The global state
lip->start_token();
state = lip->next_state;
lip->next_state = MY_LEX_START;
for (;;) {
switch (state) {
case MY_LEX_START: // Start of token
// Skip starting whitespace
while (state_map[c = lip->yyPeek()] == MY_LEX_SKIP) {
if (c == '\n') lip->yylineno++;
lip->yySkip();
}
/* Start of real token */
lip->restart_token();
c = lip->yyGet();
state = state_map[c];
break;
case MY_LEX_CHAR: // Unknown or single char token
case MY_LEX_SKIP: // This should not happen
if (c == '-' && lip->yyPeek() == '-' &&
(my_isspace(cs, lip->yyPeekn(1)) ||
my_iscntrl(cs, lip->yyPeekn(1)))) {
state = MY_LEX_COMMENT;
break;
}
if (c == '-' && lip->yyPeek() == '>') // '->'
{
lip->yySkip();
lip->next_state = MY_LEX_START;
if (lip->yyPeek() == '>') {
lip->yySkip();
return JSON_UNQUOTED_SEPARATOR_SYM;
}
return JSON_SEPARATOR_SYM;
}
if (c != ')') lip->next_state = MY_LEX_START; // Allow signed numbers
/*
Check for a placeholder: it should not precede a possible identifier
because of binlogging: when a placeholder is replaced with its value
in a query for the binlog, the query must stay grammatically correct.
*/
if (c == '?' && lip->stmt_prepare_mode && !ident_map[lip->yyPeek()])
return (PARAM_MARKER);
return ((int)c);
case MY_LEX_IDENT_OR_NCHAR:
if (lip->yyPeek() != '\'') {
state = MY_LEX_IDENT;
break;
}
/* Found N'string' */
lip->yySkip(); // Skip '
if (!(yylval->lex_str.str = get_text(lip, 2, 1))) {
state = MY_LEX_CHAR; // Read char by char
break;
}
yylval->lex_str.length = lip->yytoklen;
return (NCHAR_STRING);
case MY_LEX_IDENT_OR_HEX:
if (lip->yyPeek() == '\'') { // Found x'hex-number'
state = MY_LEX_HEX_NUMBER;
break;
}
[[fallthrough]];
case MY_LEX_IDENT_OR_BIN:
if (lip->yyPeek() == '\'') { // Found b'bin-number'
state = MY_LEX_BIN_NUMBER;
break;
}
[[fallthrough]];
case MY_LEX_IDENT:
const char *start;
if (use_mb(cs)) {
result_state = IDENT_QUOTED;
switch (my_mbcharlen(cs, lip->yyGetLast())) {
case 1:
break;
case 0:
if (my_mbmaxlenlen(cs) < 2) break;
[[fallthrough]];
default:
const int l =
my_ismbchar(cs, lip->get_ptr() - 1, lip->get_end_of_query());
if (l == 0) {
state = MY_LEX_CHAR;
continue;
}
lip->skip_binary(l - 1);
}
while (ident_map[c = lip->yyGet()]) {
switch (my_mbcharlen(cs, c)) {
case 1:
break;
case 0:
if (my_mbmaxlenlen(cs) < 2) break;
[[fallthrough]];
default:
int l;
if ((l = my_ismbchar(cs, lip->get_ptr() - 1,
lip->get_end_of_query())) == 0)
break;
lip->skip_binary(l - 1);
}
}
} else {
for (result_state = c; ident_map[c = lip->yyGet()]; result_state |= c)
;
/* If there were non-ASCII characters, mark that we must convert */
result_state = result_state & 0x80 ? IDENT_QUOTED : IDENT;
}
length = lip->yyLength();
start = lip->get_ptr();
if (lip->ignore_space) {
/*
If we find a space then this can't be an identifier. We notice this
below by checking start != lex->ptr.
*/
for (; state_map[c] == MY_LEX_SKIP; c = lip->yyGet()) {
if (c == '\n') lip->yylineno++;
}
}
if (start == lip->get_ptr() && c == '.' && ident_map[lip->yyPeek()])
lip->next_state = MY_LEX_IDENT_SEP;
else { // '(' must follow directly if function
lip->yyUnget();
if ((tokval = find_keyword(lip, length, c == '('))) {
lip->next_state = MY_LEX_START; // Allow signed numbers
return (tokval); // Was keyword
}
lip->yySkip(); // next state does a unget
}
yylval->lex_str = get_token(lip, 0, length);
/*
Note: "SELECT _bla AS 'alias'"
_bla should be considered as a IDENT if charset haven't been found.
So we don't want to produce any warning in find_primary.
*/
if (yylval->lex_str.str[0] == '_') {
auto charset_name = yylval->lex_str.str + 1;
const CHARSET_INFO *underscore_cs =
mysql::collation::find_primary(charset_name);
if (underscore_cs) {
lip->warn_on_deprecated_charset(underscore_cs, charset_name);
if (underscore_cs == &my_charset_utf8mb4_0900_ai_ci) {
/*
If underscore_cs is utf8mb4, and the collation of underscore_cs
is the default collation of utf8mb4, then update underscore_cs
with a value of the default_collation_for_utf8mb4 system
variable:
*/
underscore_cs = thd->variables.default_collation_for_utf8mb4;
}
yylval->charset = underscore_cs;
lip->m_underscore_cs = underscore_cs;
lip->body_utf8_append(lip->m_cpp_text_start,
lip->get_cpp_tok_start() + length);
return (UNDERSCORE_CHARSET);
}
}
lip->body_utf8_append(lip->m_cpp_text_start);
lip->body_utf8_append_literal(thd, &yylval->lex_str, cs,
lip->m_cpp_text_end);
return (result_state); // IDENT or IDENT_QUOTED
case MY_LEX_IDENT_SEP: // Found ident and now '.'
yylval->lex_str.str = const_cast<char *>(lip->get_ptr());
yylval->lex_str.length = 1;
c = lip->yyGet(); // should be '.'
if (const uchar next_c = lip->yyPeek(); ident_map[next_c]) {
lip->next_state =
MY_LEX_IDENT_START; // Next is an ident (not a keyword)
if (next_c == '$') // We got .$ident
push_deprecated_warn_no_replacement(
lip->m_thd,
"$ as the first character of an unquoted identifier");
} else // Probably ` or "
lip->next_state = MY_LEX_START;
return ((int)c);
case MY_LEX_NUMBER_IDENT: // number or ident which num-start
if (lip->yyGetLast() == '0') {
c = lip->yyGet();
if (c == 'x') {
while (my_isxdigit(cs, (c = lip->yyGet())))
;
if ((lip->yyLength() >= 3) && !ident_map[c]) {
/* skip '0x' */
yylval->lex_str = get_token(lip, 2, lip->yyLength() - 2);
return (HEX_NUM);
}
lip->yyUnget();
state = MY_LEX_IDENT_START;
break;
} else if (c == 'b') {
while ((c = lip->yyGet()) == '0' || c == '1')
;
if ((lip->yyLength() >= 3) && !ident_map[c]) {
/* Skip '0b' */
yylval->lex_str = get_token(lip, 2, lip->yyLength() - 2);
return (BIN_NUM);
}
lip->yyUnget();
state = MY_LEX_IDENT_START;
break;
}
lip->yyUnget();
}
while (my_isdigit(cs, (c = lip->yyGet())))
;
if (!ident_map[c]) { // Can't be identifier
state = MY_LEX_INT_OR_REAL;
break;
}
if (c == 'e' || c == 'E') {
// The following test is written this way to allow numbers of type 1e1
if (my_isdigit(cs, lip->yyPeek()) || (c = (lip->yyGet())) == '+' ||
c == '-') { // Allow 1E+10
if (my_isdigit(cs,
lip->yyPeek())) // Number must have digit after sign
{
lip->yySkip();
while (my_isdigit(cs, lip->yyGet()))
;
yylval->lex_str = get_token(lip, 0, lip->yyLength());
return (FLOAT_NUM);
}
}
lip->yyUnget();
}
[[fallthrough]];
case MY_LEX_IDENT_START: // We come here after '.'
result_state = IDENT;
if (use_mb(cs)) {
result_state = IDENT_QUOTED;
while (ident_map[c = lip->yyGet()]) {
switch (my_mbcharlen(cs, c)) {
case 1:
break;
case 0:
if (my_mbmaxlenlen(cs) < 2) break;
[[fallthrough]];
default:
int l;
if ((l = my_ismbchar(cs, lip->get_ptr() - 1,
lip->get_end_of_query())) == 0)
break;
lip->skip_binary(l - 1);
}
}
} else {
for (result_state = 0; ident_map[c = lip->yyGet()]; result_state |= c)
;
/* If there were non-ASCII characters, mark that we must convert */
result_state = result_state & 0x80 ? IDENT_QUOTED : IDENT;
}
if (c == '.' && ident_map[lip->yyPeek()])
lip->next_state = MY_LEX_IDENT_SEP; // Next is '.'
yylval->lex_str = get_token(lip, 0, lip->yyLength());
lip->body_utf8_append(lip->m_cpp_text_start);
lip->body_utf8_append_literal(thd, &yylval->lex_str, cs,
lip->m_cpp_text_end);
return (result_state);
case MY_LEX_USER_VARIABLE_DELIMITER: // Found quote char
{
uint double_quotes = 0;
const char quote_char = c; // Used char
for (;;) {
c = lip->yyGet();
if (c == 0) {
lip->yyUnget();
return ABORT_SYM; // Unmatched quotes
}
int var_length;
if ((var_length = my_mbcharlen(cs, c)) == 1) {
if (c == quote_char) {
if (lip->yyPeek() != quote_char) break;
c = lip->yyGet();
double_quotes++;
continue;
}
} else if (use_mb(cs)) {
if ((var_length = my_ismbchar(cs, lip->get_ptr() - 1,
lip->get_end_of_query())))
lip->skip_binary(var_length - 1);
}
}
if (double_quotes)
yylval->lex_str = get_quoted_token(
lip, 1, lip->yyLength() - double_quotes - 1, quote_char);
else
yylval->lex_str = get_token(lip, 1, lip->yyLength() - 1);
if (c == quote_char) lip->yySkip(); // Skip end `
lip->next_state = MY_LEX_START;
lip->body_utf8_append(lip->m_cpp_text_start);
lip->body_utf8_append_literal(thd, &yylval->lex_str, cs,
lip->m_cpp_text_end);
return (IDENT_QUOTED);
}
case MY_LEX_INT_OR_REAL: // Complete int or incomplete real
if (c != '.') { // Found complete integer number.
yylval->lex_str = get_token(lip, 0, lip->yyLength());
return int_token(yylval->lex_str.str, (uint)yylval->lex_str.length);
}
[[fallthrough]];
case MY_LEX_REAL: // Incomplete real number
while (my_isdigit(cs, c = lip->yyGet()))
;
if (c == 'e' || c == 'E') {
c = lip->yyGet();
if (c == '-' || c == '+') c = lip->yyGet(); // Skip sign
if (!my_isdigit(cs, c)) { // No digit after sign
state = MY_LEX_CHAR;
break;
}
while (my_isdigit(cs, lip->yyGet()))
;
yylval->lex_str = get_token(lip, 0, lip->yyLength());
return (FLOAT_NUM);
}
yylval->lex_str = get_token(lip, 0, lip->yyLength());
return (DECIMAL_NUM);
case MY_LEX_HEX_NUMBER: // Found x'hexstring'
lip->yySkip(); // Accept opening '
while (my_isxdigit(cs, (c = lip->yyGet())))
;
if (c != '\'') return (ABORT_SYM); // Illegal hex constant
lip->yySkip(); // Accept closing '
length = lip->yyLength(); // Length of hexnum+3
if ((length % 2) == 0) return (ABORT_SYM); // odd number of hex digits
yylval->lex_str = get_token(lip,
2, // skip x'
length - 3); // don't count x' and last '
return (HEX_NUM);
case MY_LEX_BIN_NUMBER: // Found b'bin-string'
lip->yySkip(); // Accept opening '
while ((c = lip->yyGet()) == '0' || c == '1')
;
if (c != '\'') return (ABORT_SYM); // Illegal hex constant
lip->yySkip(); // Accept closing '
length = lip->yyLength(); // Length of bin-num + 3
yylval->lex_str = get_token(lip,
2, // skip b'
length - 3); // don't count b' and last '
return (BIN_NUM);
case MY_LEX_CMP_OP: // Incomplete comparison operator
if (state_map[lip->yyPeek()] == MY_LEX_CMP_OP ||
state_map[lip->yyPeek()] == MY_LEX_LONG_CMP_OP)
lip->yySkip();
if ((tokval = find_keyword(lip, lip->yyLength() + 1, false))) {
lip->next_state = MY_LEX_START; // Allow signed numbers
return (tokval);
}
state = MY_LEX_CHAR; // Something fishy found
break;
case MY_LEX_LONG_CMP_OP: // Incomplete comparison operator
if (state_map[lip->yyPeek()] == MY_LEX_CMP_OP ||
state_map[lip->yyPeek()] == MY_LEX_LONG_CMP_OP) {
lip->yySkip();
if (state_map[lip->yyPeek()] == MY_LEX_CMP_OP) lip->yySkip();
}
if ((tokval = find_keyword(lip, lip->yyLength() + 1, false))) {
lip->next_state = MY_LEX_START; // Found long op
return (tokval);
}
state = MY_LEX_CHAR; // Something fishy found
break;
case MY_LEX_BOOL:
if (c != lip->yyPeek()) {
state = MY_LEX_CHAR;
break;
}
lip->yySkip();
tokval = find_keyword(lip, 2, false); // Is a bool operator
lip->next_state = MY_LEX_START; // Allow signed numbers
return (tokval);
case MY_LEX_STRING_OR_DELIMITER:
if (thd->variables.sql_mode & MODE_ANSI_QUOTES) {
state = MY_LEX_USER_VARIABLE_DELIMITER;
break;
}
/* " used for strings */
[[fallthrough]];
case MY_LEX_STRING: // Incomplete text string
if (!(yylval->lex_str.str = get_text(lip, 1, 1))) {
state = MY_LEX_CHAR; // Read char by char
break;
}
yylval->lex_str.length = lip->yytoklen;
lip->body_utf8_append(lip->m_cpp_text_start);
lip->body_utf8_append_literal(
thd, &yylval->lex_str,
lip->m_underscore_cs ? lip->m_underscore_cs : cs,
lip->m_cpp_text_end);
lip->m_underscore_cs = nullptr;
return (TEXT_STRING);
case MY_LEX_COMMENT: // Comment
thd->m_parser_state->add_comment();
while ((c = lip->yyGet()) != '\n' && c)
;
lip->yyUnget(); // Safety against eof
state = MY_LEX_START; // Try again
break;
case MY_LEX_LONG_COMMENT: /* Long C comment? */
if (lip->yyPeek() != '*') {
state = MY_LEX_CHAR; // Probable division
break;
}
thd->m_parser_state->add_comment();
/* Reject '/' '*', since we might need to turn off the echo */
lip->yyUnget();
lip->save_in_comment_state();
if (lip->yyPeekn(2) == '!') {
lip->in_comment = DISCARD_COMMENT;
/* Accept '/' '*' '!', but do not keep this marker. */
lip->set_echo(false);
lip->yySkip();
lip->yySkip();
lip->yySkip();
/*
The special comment format is very strict:
'/' '*' '!', followed by either
6 digits: 2 digits (major), 2 digits (mionr), 2 digits (dot), then a
white-space character.
032302 -> 3.23.2
050032 -> 5.0.32
050114 -> 5.1.14
100000 -> 10.0.0
or, if that format is not matched:
5 digits: 1 digit (major), 2 digits (minor), then 2 digits (dot).
32302 -> 3.23.2
50032 -> 5.0.32
50114 -> 5.1.14
*/
char version_str[7] = {0};
if (my_isdigit(cs, (version_str[0] = lip->yyPeekn(0))) &&
my_isdigit(cs, (version_str[1] = lip->yyPeekn(1))) &&
my_isdigit(cs, (version_str[2] = lip->yyPeekn(2))) &&
my_isdigit(cs, (version_str[3] = lip->yyPeekn(3))) &&
my_isdigit(cs, (version_str[4] = lip->yyPeekn(4)))) {
if (my_isdigit(cs, lip->yyPeekn(5)) &&
my_isspace(cs, lip->yyPeekn(6))) {
version_str[5] = lip->yyPeekn(5);
} else if (!my_isspace(cs, lip->yyPeekn(5))) {
push_warning(thd, Sql_condition::SL_WARNING,
ER_WARN_NO_SPACE_VERSION_COMMENT,
ER_THD(thd, ER_WARN_NO_SPACE_VERSION_COMMENT));
}
ulong version = strtol(version_str, nullptr, 10);
if (version <= MYSQL_VERSION_ID) {
/* Accept ('M') 'M' 'm' 'm' 'd' 'd' */
lip->yySkipn(strlen(version_str));
/* Expand the content of the special comment as real code */
lip->set_echo(true);
state = MY_LEX_START;
break; /* Do not treat contents as a comment. */
} else {
/*
Patch and skip the conditional comment to avoid it
being propagated infinitely (eg. to a slave).
*/
char *pcom = lip->yyUnput(' ');
comment_closed = !consume_comment(lip, 1);
if (!comment_closed) {
*pcom = '!';
}
/* version allowed to have one level of comment inside. */
}
} else {
/* Not a version comment. */
state = MY_LEX_START;
lip->set_echo(true);
break;
}
} else {
if (lip->in_comment != NO_COMMENT) {
push_warning(
lip->m_thd, Sql_condition::SL_WARNING,
ER_WARN_DEPRECATED_SYNTAX_NO_REPLACEMENT,
ER_THD(lip->m_thd, ER_WARN_DEPRECATED_NESTED_COMMENT_SYNTAX));
}
lip->in_comment = PRESERVE_COMMENT;
lip->yySkip(); // Accept /
lip->yySkip(); // Accept *
comment_closed = !consume_comment(lip, 0);
/* regular comments can have zero comments inside. */
}
/*
Discard:
- regular '/' '*' comments,
- special comments '/' '*' '!' for a future version,
by scanning until we find a closing '*' '/' marker.
Nesting regular comments isn't allowed. The first
'*' '/' returns the parser to the previous state.
/#!VERSI oned containing /# regular #/ is allowed #/
Inside one versioned comment, another versioned comment
is treated as a regular discardable comment. It gets
no special parsing.
*/
/* Unbalanced comments with a missing '*' '/' are a syntax error */
if (!comment_closed) return (ABORT_SYM);
state = MY_LEX_START; // Try again
lip->restore_in_comment_state();
break;
case MY_LEX_END_LONG_COMMENT:
if ((lip->in_comment != NO_COMMENT) && lip->yyPeek() == '/') {
/* Reject '*' '/' */
lip->yyUnget();
/* Accept '*' '/', with the proper echo */
lip->set_echo(lip->in_comment == PRESERVE_COMMENT);
lip->yySkipn(2);
/* And start recording the tokens again */
lip->set_echo(true);
/*
C-style comments are replaced with a single space (as it
is in C and C++). If there is already a whitespace
character at this point in the stream, the space is
not inserted.
See also ISO/IEC 9899:1999 §5.1.1.2
("Programming languages — C")
*/
if (!my_isspace(cs, lip->yyPeek()) &&
lip->get_cpp_ptr() != lip->get_cpp_buf() &&
!my_isspace(cs, *(lip->get_cpp_ptr() - 1)))
lip->cpp_inject(' ');
lip->in_comment = NO_COMMENT;
state = MY_LEX_START;
} else
state = MY_LEX_CHAR; // Return '*'
break;
case MY_LEX_SET_VAR: // Check if ':='
if (lip->yyPeek() != '=') {
state = MY_LEX_CHAR; // Return ':'
break;
}
lip->yySkip();
return (SET_VAR);
case MY_LEX_SEMICOLON: // optional line terminator
state = MY_LEX_CHAR; // Return ';'
break;
case MY_LEX_EOL:
if (lip->eof()) {
lip->yyUnget(); // Reject the last '\0'
lip->set_echo(false);
lip->yySkip();
lip->set_echo(true);
/* Unbalanced comments with a missing '*' '/' are a syntax error */
if (lip->in_comment != NO_COMMENT) return (ABORT_SYM);
lip->next_state = MY_LEX_END; // Mark for next loop
return (END_OF_INPUT);
}
state = MY_LEX_CHAR;
break;
case MY_LEX_END:
lip->next_state = MY_LEX_END;
return (0); // We found end of input last time
/* Actually real shouldn't start with . but allow them anyhow */
case MY_LEX_REAL_OR_POINT:
if (my_isdigit(cs, lip->yyPeek()))
state = MY_LEX_REAL; // Real
else {
state = MY_LEX_IDENT_SEP; // return '.'
lip->yyUnget(); // Put back '.'
}
break;
case MY_LEX_USER_END: // end '@' of user@hostname
switch (state_map[lip->yyPeek()]) {
case MY_LEX_STRING:
case MY_LEX_USER_VARIABLE_DELIMITER:
case MY_LEX_STRING_OR_DELIMITER:
break;
case MY_LEX_USER_END:
lip->next_state = MY_LEX_SYSTEM_VAR;
break;
default:
lip->next_state = MY_LEX_HOSTNAME;
break;
}
yylval->lex_str.str = const_cast<char *>(lip->get_ptr());
yylval->lex_str.length = 1;
return ((int)'@');
case MY_LEX_HOSTNAME: // end '@' of user@hostname
for (c = lip->yyGet();
my_isalnum(cs, c) || c == '.' || c == '_' || c == '$';
c = lip->yyGet())
;
yylval->lex_str = get_token(lip, 0, lip->yyLength());
return (LEX_HOSTNAME);
case MY_LEX_SYSTEM_VAR:
yylval->lex_str.str = const_cast<char *>(lip->get_ptr());
yylval->lex_str.length = 1;
lip->yySkip(); // Skip '@'
lip->next_state =
(state_map[lip->yyPeek()] == MY_LEX_USER_VARIABLE_DELIMITER
? MY_LEX_START
: MY_LEX_IDENT_OR_KEYWORD);
return ((int)'@');
case MY_LEX_IDENT_OR_KEYWORD:
/*
We come here when we have found two '@' in a row.
We should now be able to handle:
[(global | local | session) .]variable_name
*/
for (result_state = 0; ident_map[c = lip->yyGet()]; result_state |= c)
;
/* If there were non-ASCII characters, mark that we must convert */
result_state = result_state & 0x80 ? IDENT_QUOTED : IDENT;
if (c == '.') lip->next_state = MY_LEX_IDENT_SEP;
length = lip->yyLength();
if (length == 0) return (ABORT_SYM); // Names must be nonempty.
if ((tokval = find_keyword(lip, length, false))) {
lip->yyUnget(); // Put back 'c'
return (tokval); // Was keyword
}
yylval->lex_str = get_token(lip, 0, length);
lip->body_utf8_append(lip->m_cpp_text_start);
lip->body_utf8_append_literal(thd, &yylval->lex_str, cs,
lip->m_cpp_text_end);
return (result_state);
case MY_LEX_IDENT_OR_DOLLAR_QUOTED_TEXT: {
int len = 0; /* Length of the tag of the dollar quote */
uchar p = lip->yyPeek(); /* Character succeeding first $ */
// Find $ character after the tag
while (p != '$' && ident_map[p] &&
lip->get_ptr() + len <= lip->get_end_of_query()) {
if (use_mb(cs)) {
int l =
my_ismbchar(cs, lip->get_ptr() + len, lip->get_end_of_query());
if (l > 1) len += l - 1;
}
p = lip->yyPeekn(++len);
}
if (p != '$') { /* Not a dollar quote, could be an identifier */
push_deprecated_warn_no_replacement(
lip->m_thd, "$ as the first character of an unquoted identifier");
state = MY_LEX_IDENT;
break;
} else {
LEX_CSTRING text = get_dollar_quoted_text(lip, len);
if (text == NULL_CSTR)
return ABORT_SYM; // error: unterminated text
else {
yylval->lex_str.str = const_cast<char *>(text.str);
yylval->lex_str.length = text.length;
lip->body_utf8_append(text.str);
lip->body_utf8_append_literal(thd, &yylval->lex_str, cs,
text.str + text.length);
return DOLLAR_QUOTED_STRING_SYM; // $$ ... $$
}
}
}
}
}
}
void trim_whitespace(const CHARSET_INFO *cs, LEX_STRING *str) {
/*
TODO:
This code assumes that there are no multi-bytes characters
that can be considered white-space.
*/
while ((str->length > 0) && (my_isspace(cs, str->str[0]))) {
str->length--;
str->str++;
}
/*
FIXME:
Also, parsing backward is not safe with multi bytes characters
*/
while ((str->length > 0) && (my_isspace(cs, str->str[str->length - 1]))) {
str->length--;
/* set trailing spaces to 0 as there're places that don't respect length */
str->str[str->length] = 0;
}
}
/**
Prints into 'str' a comma-separated list of column names, enclosed in
parenthesis.
@param thd Thread handler
@param str Where to print
@param column_names List to print, or NULL
*/
void print_derived_column_names(const THD *thd, String *str,
const Create_col_name_list *column_names) {
if (!column_names) return;
str->append(" (");
for (auto s : *column_names) {
append_identifier(thd, str, s.str, s.length);
str->append(',');
}
str->length(str->length() - 1);
str->append(')');
}
/**
Construct and initialize Query_expression object.
*/
Query_expression::Query_expression(enum_parsing_context parsing_context)
: next(nullptr),
prev(nullptr),
master(nullptr),
slave(nullptr),
explain_marker(CTX_NONE),
prepared(false),
optimized(false),
executed(false),
m_query_result(nullptr),
uncacheable(0),
cleaned(UC_DIRTY),
types(current_thd->mem_root),
select_limit_cnt(HA_POS_ERROR),
offset_limit_cnt(0),
item(nullptr),
m_with_clause(nullptr),
derived_table(nullptr),
first_recursive(nullptr),
m_lateral_deps(0) {
switch (parsing_context) {
case CTX_ORDER_BY:
explain_marker = CTX_ORDER_BY_SQ; // A subquery in ORDER BY
break;
case CTX_GROUP_BY:
explain_marker = CTX_GROUP_BY_SQ; // A subquery in GROUP BY
break;
case CTX_ON:
explain_marker = CTX_WHERE;
break;
case CTX_HAVING: // A subquery elsewhere
case CTX_SELECT_LIST:
case CTX_UPDATE_VALUE:
case CTX_INSERT_VALUES:
case CTX_INSERT_UPDATE:
case CTX_WHERE:
case CTX_DERIVED:
case CTX_QUALIFY:
case CTX_NONE: // A subquery in a non-select
explain_marker = parsing_context;
break;
default:
/* Subquery can't happen outside of those ^. */
assert(false); /* purecov: inspected */
break;
}
}
/**
Construct and initialize Query_block object.
*/
Query_block::Query_block(MEM_ROOT *mem_root, Item *where, Item *having)
: fields(mem_root),
ftfunc_list(&ftfunc_list_alloc),
sj_nests(mem_root),
first_context(&context),
m_table_nest(mem_root),
m_current_table_nest(&m_table_nest),
m_where_cond(where),
m_having_cond(having) {}
/**
Set the name resolution context for the specified query block.
@param outer_context Outer name resolution context.
NULL if none or it will be set later.
*/
bool Query_block::set_context(Name_resolution_context *outer_context) {
context.init();
context.query_block = this;
context.outer_context = outer_context;
/*
Add the name resolution context of this query block to the
stack of contexts for the whole query.
*/
return parent_lex->push_context(&context);
}
/**
Add tables from an array to a list of used tables.
@param thd Current session.
@param tables Tables to add.
@param table_options A set of the following bits:
- TL_OPTION_UPDATING : Table will be updated,
- TL_OPTION_FORCE_INDEX : Force usage of index,
- TL_OPTION_ALIAS : an alias in multi table DELETE.
@param lock_type How table should be locked.
@param mdl_type Type of metadata lock to acquire on the table.
@returns true if error (reported), otherwise false.
*/
bool Query_block::add_tables(THD *thd,
const Mem_root_array<Table_ident *> *tables,
ulong table_options, thr_lock_type lock_type,
enum_mdl_type mdl_type) {
if (tables == nullptr) return false;
for (auto *table : *tables) {
if (!add_table_to_list(thd, table, nullptr, table_options, lock_type,
mdl_type))
return true;
}
return false;
}
/**
Exclude this query expression and its immediately contained query terms
and query blocks from AST.
@note
Query expressions that belong to the query_block objects of the current
query expression will be brought up one level and will replace
the current query expression in the list inside the outer query block.
*/
void Query_expression::exclude_level() {
/*
This change to the AST is done only during statement resolution
so doesn't need LOCK_query_plan
*/
Query_expression *qe_chain = nullptr;
Query_expression **last_qe_ref = &qe_chain;
for (Query_block *sl = first_query_block(); sl != nullptr;
sl = sl->next_query_block()) {
assert(sl->join == nullptr);
// Unlink this query block from global list
if (sl->link_prev && (*sl->link_prev = sl->link_next))
sl->link_next->link_prev = sl->link_prev;
// Bring up underlying query expressions
Query_expression **last = nullptr;
for (Query_expression *u = sl->first_inner_query_expression(); u;
u = u->next_query_expression()) {
/*
We are excluding a query block from the AST. Since this level is
removed, we must also exclude the Name_resolution_context belonging to
this level. Do this by looping through inner subqueries and changing
their contexts' outer context pointers to point to the outer query
block's context.
*/
for (auto qt : u->query_terms<>()) {
if (qt->query_block()->context.outer_context == &sl->context) {
qt->query_block()->context.outer_context =
&sl->outer_query_block()->context;
}
}
u->master = master;
last = &(u->next);
}
if (last != nullptr) {
(*last_qe_ref) = sl->first_inner_query_expression();
last_qe_ref = last;
// Unlink the query expressions that have been moved to the outer level:
sl->slave = nullptr;
}
}
if (qe_chain != nullptr) {
// Include underlying query expressions in place of the current one.
(*prev) = qe_chain;
(*last_qe_ref) = next;
if (next != nullptr) next->prev = last_qe_ref;
qe_chain->prev = prev;
} else {
// Exclude current query expression from list inside query block.
(*prev) = next;
if (next != nullptr) next->prev = prev;
}
// Cleanup and destroy this query expression, including any temporary tables:
cleanup(true);
destroy();
}
/**
Exclude current query expression with all underlying query terms,
query blocks and query expressions from AST.
*/
void Query_expression::exclude_tree() {
for (Query_block *sl = first_query_block(); sl != nullptr;
sl = sl->next_query_block()) {
/*
Exclusion is only done during preparation, however some table-less
subqueries may have been evaluated during preparation.
*/
assert(sl->join == nullptr || is_executed());
if (sl->join != nullptr) {
sl->join->destroy();
sl->join = nullptr;
}
// Unlink from global query block list
if (sl->link_prev && (*sl->link_prev = sl->link_next))
sl->link_next->link_prev = sl->link_prev;
// Exclude subtrees of all the inner query expressions of this query block
for (Query_expression *u = sl->first_inner_query_expression();
u != nullptr;) {
Query_expression *next = u->next_query_expression();
u->exclude_tree();
u = next;
}
// All underlying query expressions are now deleted:
assert(sl->slave == nullptr);
}
// Exclude current query expression from list inside query block.
(*prev) = next;
if (next != nullptr) next->prev = prev;
// Cleanup and destroy the internal objects for this query expression.
cleanup(true);
destroy();
}
/**
Invalidate by nulling out pointers to other Query expressions and
Query blocks.
*/
void Query_expression::invalidate() {
next = nullptr;
prev = nullptr;
master = nullptr;
slave = nullptr;
}
/**
Make active options from base options, supplied options, any statement
options and the environment.
@param added_options Options that are added to the active options
@param removed_options Options that are removed from the active options
*/
void Query_block::make_active_options(ulonglong added_options,
ulonglong removed_options) {
m_active_options =
(m_base_options | added_options | parent_lex->statement_options() |
parent_lex->thd->variables.option_bits) &
~removed_options;
}
/**
Mark all query blocks from this to 'last' as dependent
@param last Pointer to last Query_block struct, before which all
Query_block are marked as as dependent.
@param aggregate true if the dependency is due to a set function, such as
COUNT(*), which is aggregated within the query block 'last'.
Such functions must have a dependency on all tables of
the aggregating query block.
@note
last should be reachable from this Query_block
@todo Update OUTER_REF_TABLE_BIT for intermediate subquery items, by
replacing the below "if (aggregate)" block with:
if (last == s->outer_query_block())
{
if (aggregate)
munit->item->accumulate_used_tables(last->all_tables_map());
}
else
{
munit->item->accumulate_used_tables(OUTER_REF_TABLE_BIT);
}
and remove settings from Item_field::fix_outer_field(),
Item_ref::fix_fields().
*/
void Query_block::mark_as_dependent(Query_block *last, bool aggregate) {
// The top level query block cannot be dependent, so do not go above this:
assert(last != nullptr);
/*
Mark all selects from resolved to 1 before select where was
found table as depended (of select where was found table)
*/
for (Query_block *s = this; s && s != last; s = s->outer_query_block()) {
Query_expression *munit = s->master_query_expression();
if (!(s->uncacheable & UNCACHEABLE_DEPENDENT)) {
// Select is dependent of outer select
s->uncacheable =
(s->uncacheable & ~UNCACHEABLE_UNITED) | UNCACHEABLE_DEPENDENT;
munit->uncacheable =
(munit->uncacheable & ~UNCACHEABLE_UNITED) | UNCACHEABLE_DEPENDENT;
for (Query_block *sl = munit->first_query_block(); sl;
sl = sl->next_query_block()) {
if (sl != s &&
!(sl->uncacheable & (UNCACHEABLE_DEPENDENT | UNCACHEABLE_UNITED))) {
// Prevent early freeing in JOIN::join_free()
sl->uncacheable |= UNCACHEABLE_UNITED;
}
}
}
if (aggregate) {
munit->accumulate_used_tables(last == s->outer_query_block()
? last->all_tables_map()
: OUTER_REF_TABLE_BIT);
}
}
}
/*
prohibit using LIMIT clause
*/
bool Query_block::test_limit() {
if (select_limit != nullptr) {
my_error(ER_NOT_SUPPORTED_YET, MYF(0), "LIMIT & IN/ALL/ANY/SOME subquery");
return (true);
}
return (false);
}
enum_parsing_context Query_expression::get_explain_marker(
const THD *thd) const {
thd->query_plan.assert_plan_is_locked_if_other();
return explain_marker;
}
void Query_expression::set_explain_marker(THD *thd, enum_parsing_context m) {
thd->lock_query_plan();
explain_marker = m;
thd->unlock_query_plan();
}
void Query_expression::set_explain_marker_from(THD *thd,
const Query_expression *u) {
thd->lock_query_plan();
explain_marker = u->explain_marker;
thd->unlock_query_plan();
}
ha_rows Query_block::get_offset(const THD *) const {
if (offset_limit != nullptr)
return ha_rows{offset_limit->val_uint()};
else
return ha_rows{0};
}
ha_rows Query_block::get_limit(const THD *thd) const {
/*
If m_use_select_limit is set in the query block, return the value
of the variable select_limit, unless an explicit limit is set.
This is used to implement SQL_SELECT_LIMIT for SELECT statements.
*/
if (select_limit != nullptr)
return ha_rows{select_limit->val_uint()};
else if (m_use_select_limit)
return ha_rows{thd->variables.select_limit};
else
return ha_rows{HA_POS_ERROR};
}
bool Query_block::add_item_to_list(Item *item) {
DBUG_TRACE;
DBUG_PRINT("info", ("Item: %p", item));
assert_consistent_hidden_flags(fields, item, /*hidden=*/false);
fields.push_back(item);
item->hidden = false;
return false;
}
bool Query_block::add_ftfunc_to_list(Item_func_match *func) {
return !func || ftfunc_list->push_back(func); // end of memory?
}
/**
Invalidate by nulling out pointers to other Query_expressions and
Query_blockes.
*/
void Query_block::invalidate() {
next = nullptr;
master = nullptr;
slave = nullptr;
link_next = nullptr;
link_prev = nullptr;
}
bool Query_block::setup_base_ref_items(THD *thd) {
uint order_group_num = order_list.elements + group_list.elements;
// find_order_in_list() may need some extra space, so multiply by two.
order_group_num *= 2;
// create_distinct_group() may need some extra space
if (is_distinct()) {
uint bitcount = 0;
for (Item *item : visible_fields()) {
/*
Same test as in create_distinct_group, when it pushes new items to the
end of base_ref_items. An extra test for 'fixed' which, at this
stage, will be true only for columns inserted for a '*' wildcard.
*/
if (item->fixed && item->type() == Item::FIELD_ITEM &&
item->data_type() == MYSQL_TYPE_BIT)
++bitcount;
}
order_group_num += bitcount;
}
/*
We have to create array in prepared statement memory if it is
prepared statement
*/
Query_arena *arena = thd->stmt_arena;
uint n_elems = n_sum_items + n_child_sum_items + fields.size() +
select_n_having_items + select_n_where_fields +
order_group_num + n_scalar_subqueries;
/*
If it is possible that we transform IN(subquery) to a join to a derived
table, we will be adding DISTINCT (this possibly has the problem of BIT
columns as in the logic above), and we will also be adding one expression to
the SELECT list per decorrelated equality in WHERE. So we have to allocate
more space.
The number of decorrelatable equalities is bounded by
select_n_where_fields. Indeed an equality isn't counted in
select_n_where_fields if it's:
expr-without_Item_field = expr-without_Item_field;
but we decorrelate an equality if one member has OUTER_REF_TABLE_BIT, so
it has an Item_field inside.
Note that cond_count cannot be used, as setup_cond() hasn't run yet. So we
use select_n_where_fields instead.
*/
if (master_query_expression()->item &&
(thd->optimizer_switch_flag(OPTIMIZER_SWITCH_SUBQUERY_TO_DERIVED) ||
(thd->lex->m_sql_cmd != nullptr &&
thd->secondary_engine_optimization() ==
Secondary_engine_optimization::SECONDARY))) {
Item_subselect *subq_predicate = master_query_expression()->item;
if (subq_predicate->subquery_type() == Item_subselect::EXISTS_SUBQUERY ||
subq_predicate->subquery_type() == Item_subselect::IN_SUBQUERY) {
// might be transformed to derived table, so:
n_elems +=
// possible additions to SELECT list from decorrelation of WHERE
select_n_where_fields +
// add size of new SELECT list, for DISTINCT and BIT type
(select_n_where_fields + fields.size());
}
}
DBUG_PRINT(
"info",
("setup_ref_array this %p %4u : %4u %4u %4zu %4u %4u %4u %4u", this,
n_elems, // :
n_sum_items, n_child_sum_items, fields.size(), select_n_having_items,
select_n_where_fields, order_group_num, n_scalar_subqueries));
if (!base_ref_items.is_null()) {
/*
This should not happen, as it's the sign of preparing an already-prepared
Query_block. It does happen (in test main.sp-error, section for bug13037):
a table-less substatement fails due to wrong identifier, and
LEX::mark_broken() doesn't mark it as broken as it uses no tables; so it
will be reused by the next CALL. WL#6570.
*/
if (base_ref_items.size() >= n_elems) return false;
}
Item **array = static_cast<Item **>(arena->alloc(sizeof(Item *) * n_elems));
if (array == nullptr) return true;
base_ref_items = Ref_item_array(array, n_elems);
return false;
}
void print_set_operation(const THD *thd, Query_term *op, String *str, int level,
enum_query_type query_type) {
if (op->term_type() == QT_QUERY_BLOCK) {
Query_block *const block = op->query_block();
const bool needs_parens =
block->has_limit() || block->order_list.elements > 0;
if (needs_parens) str->append('(');
op->query_block()->print(thd, str, query_type);
if (needs_parens) str->append(')');
} else {
Query_term_set_op *qts = down_cast<Query_term_set_op *>(op);
const bool needs_parens = level > 0;
if (needs_parens) str->append('(');
for (uint i = 0; i < qts->m_children.size(); ++i) {
print_set_operation(thd, qts->m_children[i], str, level + 1, query_type);
if (i < qts->m_children.size() - 1) {
switch (op->term_type()) {
case QT_UNION:
str->append(STRING_WITH_LEN(" union "));
break;
case QT_INTERSECT:
str->append(STRING_WITH_LEN(" intersect "));
break;
case QT_EXCEPT:
str->append(STRING_WITH_LEN(" except "));
break;
default:
assert(false);
}
if (static_cast<signed int>(i) + 1 > qts->m_last_distinct) {
str->append(STRING_WITH_LEN("all "));
}
}
}
if (op->query_block()->order_list.elements > 0) {
str->append(STRING_WITH_LEN(" order by "));
op->query_block()->print_order(
thd, str, op->query_block()->order_list.first, query_type);
}
op->query_block()->print_limit(thd, str, query_type);
if (needs_parens) str->append(')');
}
}
void Query_expression::print(const THD *thd, String *str,
enum_query_type query_type) {
if (m_with_clause) m_with_clause->print(thd, str, query_type);
if (is_simple()) {
Query_block *sl = query_term()->query_block();
assert(sl->next_query_block() == nullptr);
sl->print(thd, str, query_type);
} else {
print_set_operation(thd, query_term(), str, 0, query_type);
}
}
void Query_block::print_limit(const THD *thd, String *str,
enum_query_type query_type) const {
Query_expression *unit = master_query_expression();
Item_subselect *item = unit->item;
if (item != nullptr && unit->global_parameters() == this) {
Item_subselect::Subquery_type type = item->subquery_type();
if (type == Item_subselect::EXISTS_SUBQUERY ||
type == Item_subselect::IN_SUBQUERY ||
type == Item_subselect::ALL_SUBQUERY)
return;
}
if (has_limit() && !m_internal_limit) {
str->append(STRING_WITH_LEN(" limit "));
if (offset_limit) {
offset_limit->print(thd, str, query_type);
str->append(',');
}
select_limit->print(thd, str, query_type);
}
}
/**
@brief Print an index hint
@details Prints out the USE|FORCE|IGNORE index hint.
@param thd the current thread
@param[out] str appends the index hint here
*/
void Index_hint::print(const THD *thd, String *str) {
switch (type) {
case INDEX_HINT_IGNORE:
str->append(STRING_WITH_LEN("IGNORE INDEX"));
break;
case INDEX_HINT_USE:
str->append(STRING_WITH_LEN("USE INDEX"));
break;
case INDEX_HINT_FORCE:
str->append(STRING_WITH_LEN("FORCE INDEX"));
break;
}
switch (clause) {
case INDEX_HINT_MASK_ALL:
break;
case INDEX_HINT_MASK_JOIN:
str->append(STRING_WITH_LEN(" FOR JOIN"));
break;
case INDEX_HINT_MASK_ORDER:
str->append(STRING_WITH_LEN(" FOR ORDER BY"));
break;
case INDEX_HINT_MASK_GROUP:
str->append(STRING_WITH_LEN(" FOR GROUP BY"));
break;
}
str->append(STRING_WITH_LEN(" ("));
if (key_name.length) {
if (thd && !my_strnncoll(system_charset_info, (const uchar *)key_name.str,
key_name.length, (const uchar *)primary_key_name,
strlen(primary_key_name)))
str->append(primary_key_name);
else
append_identifier(thd, str, key_name.str, key_name.length);
}
str->append(')');
}
typedef Prealloced_array<Table_ref *, 8> Table_array;
static void print_table_array(const THD *thd, String *str,
const Table_array &tables,
enum_query_type query_type) {
assert(!tables.empty());
Table_array::const_iterator it = tables.begin();
bool first = true;
for (; it != tables.end(); ++it) {
Table_ref *curr = *it;
const bool is_optimized =
curr->query_block->join && curr->query_block->join->is_optimized();
// the JOIN ON condition
Item *const cond =
is_optimized ? curr->join_cond_optim() : curr->join_cond();
// Print the join operator which relates this table to the previous one
const char *op = nullptr;
if (curr->is_aj_nest())
op = " anti join ";
else if (curr->is_sj_nest())
op = " semi join ";
else if (curr->outer_join) {
/* MySQL converts right to left joins */
op = " left join ";
} else if (!first || cond) {
/*
If it's the first table, and it has an ON condition (can happen due to
query transformations, e.g. merging a single-table view moves view's
WHERE to table's ON): ON also needs JOIN.
*/
op = curr->straight ? " straight_join " : " join ";
}
if (op) {
if (first) {
// Add a dummy table before the operator, to have sensible SQL:
str->append(STRING_WITH_LEN("<constant table>"));
}
str->append(op);
}
curr->print(thd, str, query_type); // Print table
/*
Print table hint info after the table name. Used only
for explaining views. There is no functionality, just
additional info for user.
*/
if (thd->lex->is_explain() && curr->opt_hints_table &&
curr->belong_to_view) {
str->append(STRING_WITH_LEN(" /*+ "));
curr->opt_hints_table->print(thd, str, query_type);
str->append(STRING_WITH_LEN("*/ "));
}
// Print join condition
if (cond) {
str->append(STRING_WITH_LEN(" on("));
cond->print(thd, str, query_type);
str->append(')');
}
first = false;
}
}
/**
Print joins from the FROM clause.
@param thd thread handler
@param str string where table should be printed
@param tables list of tables in join
@param query_type type of the query is being generated
*/
static void print_join(const THD *thd, String *str,
mem_root_deque<Table_ref *> *tables,
enum_query_type query_type) {
/* List is reversed => we should reverse it before using */
/*
If the QT_NO_DATA_EXPANSION flag is specified, we print the
original table list, including constant tables that have been
optimized away, as the constant tables may be referenced in the
expression printed by Item_field::print() when this flag is given.
Otherwise, only non-const tables are printed.
Example:
Original SQL:
select * from (select 1) t
Printed without QT_NO_DATA_EXPANSION:
select '1' AS `1` from dual
Printed with QT_NO_DATA_EXPANSION:
select `t`.`1` from (select 1 AS `1`) `t`
*/
const bool print_const_tables = (query_type & QT_NO_DATA_EXPANSION);
Table_array tables_to_print(PSI_NOT_INSTRUMENTED);
for (Table_ref *t : *tables) {
// The single table added to fake_query_block has no name;
// “from dual” looks slightly better than “from ``”, so drop it.
// (The fake_query_block query is invalid either way.)
if (t->alias[0] == '\0') continue;
if (print_const_tables || !t->optimized_away)
if (tables_to_print.push_back(t)) return; /* purecov: inspected */
}
if (tables_to_print.empty()) {
str->append(STRING_WITH_LEN("dual"));
return; // all tables were optimized away
}
std::reverse(tables_to_print.begin(), tables_to_print.end());
print_table_array(thd, str, tables_to_print, query_type);
}
/**
@returns whether a database is equal to the connection's default database
*/
bool db_is_default_db(const char *db, size_t db_len, const THD *thd) {
return thd != nullptr && thd->db().str != nullptr &&
thd->db().length == db_len && !memcmp(db, thd->db().str, db_len);
}
/*.*
Print table as it should be in join list.
@param str string where table should be printed
*/
void Table_ref::print(const THD *thd, String *str,
enum_query_type query_type) const {
if (nested_join) {
str->append('(');
print_join(thd, str, &nested_join->m_tables, query_type);
str->append(')');
} else {
const char *cmp_name; // Name to compare with alias
if (is_table_function()) {
table_function->print(thd, str, query_type);
cmp_name = table_name;
} else if (is_derived() && !is_merged() && !common_table_expr()) {
// A derived table that is materialized or without specified algorithm
if (!(query_type & QT_DERIVED_TABLE_ONLY_ALIAS)) {
if (derived_query_expression()->m_lateral_deps)
str->append(STRING_WITH_LEN("lateral "));
str->append('(');
derived->print(thd, str, query_type);
str->append(')');
}
cmp_name = ""; // Force printing of alias
} else {
// A normal table, or a view, or a CTE
if (db_length && !(query_type & QT_NO_DB) &&
!((query_type & QT_NO_DEFAULT_DB) &&
db_is_default_db(db, db_length, thd))) {
append_identifier(thd, str, db, db_length);
str->append('.');
}
append_identifier(thd, str, table_name, table_name_length);
cmp_name = table_name;
if (partition_names && partition_names->elements) {
int i, num_parts = partition_names->elements;
List_iterator<String> name_it(*(partition_names));
str->append(STRING_WITH_LEN(" PARTITION ("));
for (i = 1; i <= num_parts; i++) {
String *name = name_it++;
append_identifier(thd, str, name->c_ptr(), name->length());
if (i != num_parts) str->append(',');
}
str->append(')');
}
if (has_tablesample()) {
str->append(" TABLESAMPLE ");
str->append(SamplingTypeToString(get_sampling_type()));
str->append('(');
auto *item_decimal =
new (thd->mem_root) Item_decimal(get_sampling_percentage());
item_decimal->print(thd, str, QT_ORDINARY);
str->append(") ");
}
}
if (my_strcasecmp(table_alias_charset, cmp_name, alias)) {
char t_alias_buff[MAX_ALIAS_NAME];
const char *t_alias = alias;
str->append(' ');
if (lower_case_table_names == 1) {
if (alias && alias[0]) // Print alias in lowercase
{
my_stpcpy(t_alias_buff, alias);
my_casedn_str(files_charset_info, t_alias_buff);
t_alias = t_alias_buff;
}
}
append_identifier(thd, str, t_alias, strlen(t_alias));
}
/*
The optional column list is to be specified in the definition. For a
CTE, the definition is in WITH, and here we only have a
reference. For a Derived Table, the definition is here.
*/
if (is_derived() && !common_table_expr())
print_derived_column_names(thd, str, m_derived_column_names);
if (index_hints) {
List_iterator<Index_hint> it(*index_hints);
Index_hint *hint;
while ((hint = it++)) {
str->append(STRING_WITH_LEN(" "));
hint->print(thd, str);
}
}
}
}
void Query_block::print(const THD *thd, String *str,
enum_query_type query_type) {
/* QQ: thd may not be set for sub queries, but this should be fixed */
if (!thd) thd = current_thd;
if (select_number == 1) {
if (print_error(thd, str)) return;
switch (parent_lex->sql_command) {
case SQLCOM_UPDATE:
[[fallthrough]];
case SQLCOM_UPDATE_MULTI:
print_update(thd, str, query_type);
return;
case SQLCOM_DELETE:
[[fallthrough]];
case SQLCOM_DELETE_MULTI:
print_delete(thd, str, query_type);
return;
case SQLCOM_INSERT:
[[fallthrough]];
case SQLCOM_INSERT_SELECT:
case SQLCOM_REPLACE:
case SQLCOM_REPLACE_SELECT:
print_insert(thd, str, query_type);
return;
case SQLCOM_SELECT:
[[fallthrough]];
default:
break;
}
}
if (is_table_value_constructor) {
print_values(thd, str, query_type, *row_value_list, "row");
} else {
print_query_block(thd, str, query_type);
}
}
void Query_block::print_query_block(const THD *thd, String *str,
enum_query_type query_type) {
if (query_type & QT_SHOW_SELECT_NUMBER) {
/* it makes EXPLAIN's "id" column understandable */
str->append("/* select#");
str->append_ulonglong(select_number);
str->append(" */ select ");
} else
str->append(STRING_WITH_LEN("select "));
print_hints(thd, str, query_type);
print_select_options(str);
print_item_list(thd, str, query_type);
print_from_clause(thd, str, query_type);
print_where_cond(thd, str, query_type);
print_group_by(thd, str, query_type);
print_having(thd, str, query_type);
print_windows(thd, str, query_type);
print_qualify(thd, str, query_type);
print_order_by(thd, str, query_type);
print_limit(thd, str, query_type);
// PROCEDURE unsupported here
}
void Query_block::print_update(const THD *thd, String *str,
enum_query_type query_type) {
Sql_cmd_update *sql_cmd_update =
(static_cast<Sql_cmd_update *>(parent_lex->m_sql_cmd));
str->append(STRING_WITH_LEN("update "));
print_hints(thd, str, query_type);
print_update_options(str);
if (parent_lex->sql_command == SQLCOM_UPDATE) {
// Single table update
Table_ref *t = get_table_list();
t->print(thd, str, query_type); // table identifier
str->append(STRING_WITH_LEN(" set "));
print_update_list(thd, str, query_type, fields,
*sql_cmd_update->update_value_list);
/*
Print join condition (may happen with a merged view's WHERE condition
and disappears in simplify_joins(); visible in opt trace only).
*/
Item *const cond = t->join_cond();
if (cond) {
str->append(STRING_WITH_LEN(" on("));
cond->print(thd, str, query_type);
str->append(')');
}
print_where_cond(thd, str, query_type);
print_order_by(thd, str, query_type);
print_limit(thd, str, query_type);
} else {
// Multi table update
print_join(thd, str, &m_table_nest, query_type);
str->append(STRING_WITH_LEN(" set "));
print_update_list(thd, str, query_type, fields,
*sql_cmd_update->update_value_list);
print_where_cond(thd, str, query_type);
}
}
void Query_block::print_delete(const THD *thd, String *str,
enum_query_type query_type) {
str->append(STRING_WITH_LEN("delete "));
print_hints(thd, str, query_type);
print_delete_options(str);
if (parent_lex->sql_command == SQLCOM_DELETE) {
Table_ref *t = get_table_list();
// Single table delete
str->append(STRING_WITH_LEN("from "));
t->print(thd, str, query_type); // table identifier
/*
Print join condition (may happen with a merged view's WHERE condition
and disappears in simplify_joins(); visible in opt trace only).
*/
Item *const cond = t->join_cond();
if (cond) {
str->append(STRING_WITH_LEN(" on("));
cond->print(thd, str, query_type);
str->append(')');
}
print_where_cond(thd, str, query_type);
print_order_by(thd, str, query_type);
print_limit(thd, str, query_type);
} else {
// Multi table delete
print_table_references(thd, str, parent_lex->query_tables, query_type);
str->append(STRING_WITH_LEN(" from "));
print_join(thd, str, &m_table_nest, query_type);
print_where_cond(thd, str, query_type);
}
}
void Query_block::print_insert(const THD *thd, String *str,
enum_query_type query_type) {
/**
USES: 'INSERT INTO table (fields) VALUES values' syntax over
'INSERT INTO table SET field = value, ...'
*/
Sql_cmd_insert_base *sql_cmd_insert =
down_cast<Sql_cmd_insert_base *>(parent_lex->m_sql_cmd);
if (parent_lex->sql_command == SQLCOM_REPLACE ||
parent_lex->sql_command == SQLCOM_REPLACE_SELECT)
str->append(STRING_WITH_LEN("replace "));
else
str->append(STRING_WITH_LEN("insert "));
// Don't print QB name hints since it will be printed through
// print_query_block.
print_hints(thd, str, enum_query_type(query_type | QT_IGNORE_QB_NAME));
print_insert_options(str);
str->append(STRING_WITH_LEN("into "));
Table_ref *tbl = (parent_lex->insert_table_leaf)
? parent_lex->insert_table_leaf
: get_table_list();
tbl->print(thd, str, query_type); // table identifier
print_insert_fields(thd, str, query_type);
str->append(STRING_WITH_LEN(" "));
if (parent_lex->sql_command == SQLCOM_INSERT ||
parent_lex->sql_command == SQLCOM_REPLACE) {
print_values(thd, str, query_type, sql_cmd_insert->insert_many_values,
nullptr);
} else {
/*
Print only QB name hint here since other hints were printed in the
earlier call to print_hints.
*/
print_query_block(thd, str, enum_query_type(query_type | QT_ONLY_QB_NAME));
}
if (!sql_cmd_insert->update_field_list.empty()) {
str->append(STRING_WITH_LEN(" on duplicate key update "));
print_update_list(thd, str, query_type, sql_cmd_insert->update_field_list,
sql_cmd_insert->update_value_list);
}
}
void Query_block::print_hints(const THD *thd, String *str,
enum_query_type query_type) {
if (thd->lex->opt_hints_global) {
char buff[NAME_LEN];
String hint_str(buff, sizeof(buff), system_charset_info);
hint_str.length(0);
if (select_number == 1 ||
// First select number is 2 for SHOW CREATE VIEW
(select_number == 2 && parent_lex->sql_command == SQLCOM_SHOW_CREATE)) {
if (opt_hints_qb && !(query_type & QT_IGNORE_QB_NAME))
opt_hints_qb->append_qb_hint(thd, &hint_str);
if (!(query_type & QT_ONLY_QB_NAME))
thd->lex->opt_hints_global->print(thd, &hint_str, query_type);
} else if (opt_hints_qb)
opt_hints_qb->append_qb_hint(thd, &hint_str);
if (hint_str.length() > 0) {
str->append(STRING_WITH_LEN("/*+ "));
str->append(hint_str.ptr(), hint_str.length());
str->append(STRING_WITH_LEN("*/ "));
}
}
}
bool Query_block::print_error(const THD *thd, String *str) {
if (thd->is_error()) {
/*
It is possible that this query block had an optimization error, but the
caller didn't notice (caller evaluated this as a subquery and Item::val*()
don't have an error status). In this case the query block may be broken
and printing it may crash.
*/
str->append(STRING_WITH_LEN("had some error"));
return true;
}
/*
In order to provide info for EXPLAIN FOR CONNECTION units shouldn't be
completely cleaned till the end of the query. This is valid only for
explainable commands.
*/
assert(!(master_query_expression()->cleaned == Query_expression::UC_CLEAN &&
is_explainable_query(thd->lex->sql_command)));
return false;
}
void Query_block::print_select_options(String *str) {
std::string select_options_str;
get_select_options_str(active_options(), &select_options_str);
str->append(select_options_str);
if (!select_options_str.empty()) str->append(" ");
}
void Query_block::print_update_options(String *str) {
if (get_table_list() &&
get_table_list()->mdl_request.type == MDL_SHARED_WRITE_LOW_PRIO)
str->append(STRING_WITH_LEN("low_priority "));
if (parent_lex->is_ignore()) str->append(STRING_WITH_LEN("ignore "));
}
void Query_block::print_delete_options(String *str) {
if (get_table_list() &&
get_table_list()->mdl_request.type == MDL_SHARED_WRITE_LOW_PRIO)
str->append(STRING_WITH_LEN("low_priority "));
if (active_options() & OPTION_QUICK) str->append(STRING_WITH_LEN("quick "));
if (parent_lex->is_ignore()) str->append(STRING_WITH_LEN("ignore "));
}
void Query_block::print_insert_options(String *str) {
if (get_table_list()) {
const int type = static_cast<int>(get_table_list()->lock_descriptor().type);
// Lock option
if (type == static_cast<int>(TL_WRITE_LOW_PRIORITY))
str->append(STRING_WITH_LEN("low_priority "));
else if (type == static_cast<int>(TL_WRITE))
str->append(STRING_WITH_LEN("high_priority "));
}
if (parent_lex->is_ignore()) str->append(STRING_WITH_LEN("ignore "));
}
void Query_block::print_table_references(const THD *thd, String *str,
Table_ref *table_list,
enum_query_type query_type) {
bool first = true;
for (Table_ref *tbl = table_list; tbl; tbl = tbl->next_local) {
if (tbl->updating) {
if (first)
first = false;
else
str->append(STRING_WITH_LEN(", "));
Table_ref *t = tbl;
/*
Query Rewrite Plugin will not have is_view() set even for a view. This
is because operations like open_table haven't happened yet. So the
underlying target tables will not be added, only the original
table/view list will be reproduced. Ideally, it would be better if
Table_ref::updatable_base_table() were used here, but that isn't
possible due to QRP.
*/
while (t->is_view()) t = t->merge_underlying_list;
if (!(query_type & QT_NO_DB) &&
!((query_type & QT_NO_DEFAULT_DB) &&
db_is_default_db(t->db, t->db_length, thd))) {
append_identifier(thd, str, t->db, t->db_length);
str->append('.');
}
append_identifier(thd, str, t->table_name, t->table_name_length);
}
}
}
void Query_block::print_item_list(const THD *thd, String *str,
enum_query_type query_type) {
// Item List
bool first = true;
for (Item *item : visible_fields()) {
if (first)
first = false;
else
str->append(',');
if ((master_query_expression()->item &&
item->item_name.is_autogenerated()) ||
(query_type & QT_NORMALIZED_FORMAT)) {
/*
Do not print auto-generated aliases in subqueries. It has no purpose
in a view definition or other contexts where the query is printed.
*/
item->print(thd, str, query_type);
} else
item->print_item_w_name(thd, str, query_type);
/** @note that 'INTO variable' clauses are not printed */
}
}
void Query_block::print_update_list(const THD *thd, String *str,
enum_query_type query_type,
const mem_root_deque<Item *> &fields,
const mem_root_deque<Item *> &values) {
auto it_column = VisibleFields(fields).begin();
auto it_value = values.begin();
bool first = true;
while (it_column != VisibleFields(fields).end() && it_value != values.end()) {
Item *column = *it_column++;
Item *value = *it_value++;
if (first)
first = false;
else
str->append(',');
column->print(thd, str, query_type);
str->append(STRING_WITH_LEN(" = "));
value->print(thd, str, enum_query_type(query_type & ~QT_NO_DATA_EXPANSION));
}
}
void Query_block::print_insert_fields(const THD *thd, String *str,
enum_query_type query_type) {
Sql_cmd_insert_base *const cmd =
down_cast<Sql_cmd_insert_base *>(parent_lex->m_sql_cmd);
const mem_root_deque<Item *> &fields = cmd->insert_field_list;
if (cmd->column_count > 0) {
str->append(STRING_WITH_LEN(" ("));
bool first = true;
for (Item *field : fields) {
if (first)
first = false;
else
str->append(',');
field->print(thd, str, query_type);
}
str->append(')');
}
}
void Query_block::print_values(
const THD *thd, String *str, enum_query_type query_type,
const mem_root_deque<mem_root_deque<Item *> *> &values,
const char *prefix) {
str->append(STRING_WITH_LEN("values "));
bool row_first = true;
for (const mem_root_deque<Item *> *row : values) {
if (row_first)
row_first = false;
else
str->append(',');
if (prefix != nullptr) str->append(prefix);
str->append('(');
bool col_first = true;
for (Item *item : *row) {
if (col_first)
col_first = false;
else
str->append(',');
item->print(thd, str, query_type);
}
str->append(')');
}
}
void Query_block::print_from_clause(const THD *thd, String *str,
enum_query_type query_type) {
/*
from clause
*/
if (has_tables()) {
str->append(STRING_WITH_LEN(" from "));
/* go through join tree */
print_join(thd, str, &m_table_nest, query_type);
} else if (m_where_cond) {
/*
"SELECT 1 FROM DUAL WHERE 2" should not be printed as
"SELECT 1 WHERE 2": the 1st syntax is valid, but the 2nd is not.
*/
str->append(STRING_WITH_LEN(" from DUAL "));
}
}
void Query_block::print_where_cond(const THD *thd, String *str,
enum_query_type query_type) {
// Where
Item *const cur_where =
(join && join->is_optimized()) ? join->where_cond : m_where_cond;
if (cur_where || cond_value != Item::COND_UNDEF) {
str->append(STRING_WITH_LEN(" where "));
if (cur_where)
cur_where->print(thd, str, query_type);
else
str->append(cond_value != Item::COND_FALSE ? "true" : "false");
}
}
void Query_block::print_group_by(const THD *thd, String *str,
enum_query_type query_type) {
// group by & olap
if (group_list.elements) {
str->append(STRING_WITH_LEN(" group by "));
if (olap == CUBE_TYPE) {
str->append(STRING_WITH_LEN("cube ("));
}
print_order(thd, str, group_list.first, query_type);
switch (olap) {
case ROLLUP_TYPE:
str->append(STRING_WITH_LEN(" with rollup"));
break;
case CUBE_TYPE:
str->append(STRING_WITH_LEN(")"));
break;
default:; // satisfy compiler
}
}
}
void Query_block::print_having(const THD *thd, String *str,
enum_query_type query_type) {
// having
Item *const cur_having = (join && join->having_for_explain != (Item *)1)
? join->having_for_explain
: m_having_cond;
if (cur_having || having_value != Item::COND_UNDEF) {
str->append(STRING_WITH_LEN(" having "));
if (cur_having)
cur_having->print(thd, str, query_type);
else
str->append(having_value != Item::COND_FALSE ? "true" : "false");
}
}
void Query_block::print_qualify(const THD *thd, String *str,
enum_query_type query_type) const {
if (qualify_cond() != nullptr) {
str->append(STRING_WITH_LEN(" qualify "));
qualify_cond()->print(thd, str, query_type);
}
}
void Query_block::print_windows(const THD *thd, String *str,
enum_query_type query_type) {
List_iterator<Window> li(m_windows);
Window *w;
bool first = true;
while ((w = li++)) {
if (w->name() == nullptr) continue; // will be printed with function
if (first) {
first = false;
str->append(" window ");
} else {
str->append(", ");
}
append_identifier(thd, str, w->name()->item_name.ptr(),
strlen(w->name()->item_name.ptr()));
str->append(" AS ");
w->print(thd, str, query_type, true);
}
}
void Query_block::print_order_by(const THD *thd, String *str,
enum_query_type query_type) const {
if (order_list.elements) {
str->append(STRING_WITH_LEN(" order by "));
print_order(thd, str, order_list.first, query_type);
}
}
static enum_walk get_walk_flags(const Select_lex_visitor *visitor) {
if (visitor->visits_in_prefix_order())
return enum_walk::SUBQUERY_PREFIX;
else
return enum_walk::SUBQUERY_POSTFIX;
}
bool walk_item(Item *item, Select_lex_visitor *visitor) {
if (item == nullptr) return false;
return item->walk(&Item::visitor_processor, get_walk_flags(visitor),
pointer_cast<uchar *>(visitor));
}
bool accept_for_order(SQL_I_List<ORDER> orders, Select_lex_visitor *visitor) {
if (orders.elements == 0) return false;
for (ORDER *order = orders.first; order != nullptr; order = order->next)
if (walk_item(*order->item, visitor)) return true;
return false;
}
bool Query_expression::accept(Select_lex_visitor *visitor) {
for (auto qt : query_terms<>()) {
if (qt->term_type() == QT_QUERY_BLOCK)
qt->query_block()->accept(visitor);
else
// FIXME: why doesn't this also visit limit? done for Query_block's limit
// FIXME: Worse, Query_block::accept doesn't visit windows' ordering
// expressions
accept_for_order(qt->query_block()->order_list, visitor);
}
return visitor->visit(this);
}
bool accept_for_join(mem_root_deque<Table_ref *> *tables,
Select_lex_visitor *visitor) {
for (Table_ref *t : *tables) {
if (accept_table(t, visitor)) return true;
}
return false;
}
bool accept_table(Table_ref *t, Select_lex_visitor *visitor) {
if (t->nested_join && accept_for_join(&t->nested_join->m_tables, visitor))
return true;
if (t->is_derived()) t->derived_query_expression()->accept(visitor);
if (walk_item(t->join_cond(), visitor)) return true;
return false;
}
bool Query_block::accept(Select_lex_visitor *visitor) {
// Select clause
for (Item *item : visible_fields()) {
if (walk_item(item, visitor)) return true;
}
// From clause
if (has_tables() && accept_for_join(m_current_table_nest, visitor))
return true;
// Where clause
Item *where_condition = join != nullptr ? join->where_cond : m_where_cond;
if (where_condition != nullptr && walk_item(where_condition, visitor))
return true;
// Group by and olap clauses
if (accept_for_order(group_list, visitor)) return true;
// Having clause
Item *having_condition =
join != nullptr ? join->having_for_explain : m_having_cond;
if (walk_item(having_condition, visitor)) return true;
// Order clause
if (accept_for_order(order_list, visitor)) return true;
// Limit clause
if (has_limit()) {
if (walk_item(offset_limit, visitor) || walk_item(select_limit, visitor))
return true;
}
return visitor->visit(this);
}
void LEX::clear_privileges() {
users_list.clear();
columns.clear();
grant = grant_tot_col = grant_privilege = false;
all_privileges = false;
ssl_type = SSL_TYPE_NOT_SPECIFIED;
ssl_cipher = x509_subject = x509_issuer = nullptr;
alter_password.cleanup();
memset(&mqh, 0, sizeof(mqh));
dynamic_privileges.clear();
default_roles = nullptr;
}
/*
Initialize (or reset) Query_tables_list object.
SYNOPSIS
reset_query_tables_list()
init true - we should perform full initialization of object with
allocating needed memory
false - object is already initialized so we should only reset
its state so it can be used for parsing/processing
of new statement
DESCRIPTION
This method initializes Query_tables_list so it can be used as part
of LEX object for parsing/processing of statement. One can also use
this method to reset state of already initialized Query_tables_list
so it can be used for processing of new statement.
*/
void Query_tables_list::reset_query_tables_list(bool init) {
sql_command = SQLCOM_END;
if (!init && query_tables) {
Table_ref *table = query_tables;
for (;;) {
delete table->view_query();
if (query_tables_last == &table->next_global ||
!(table = table->next_global))
break;
}
}
query_tables = nullptr;
query_tables_last = &query_tables;
query_tables_own_last = nullptr;
if (init) {
/*
We delay real initialization of hash (and therefore related
memory allocation) until first insertion into this hash.
*/
sroutines.reset();
} else if (sroutines != nullptr) {
sroutines->clear();
}
sroutines_list.clear();
sroutines_list_own_last = sroutines_list.next;
sroutines_list_own_elements = 0;
binlog_stmt_flags = 0;
stmt_accessed_table_flag = 0;
lock_tables_state = LTS_NOT_LOCKED;
table_count = 0;
using_match = false;
stmt_unsafe_with_mixed_mode = false;
/* Check the max size of the enum to control new enum values definitions. */
static_assert(BINLOG_STMT_UNSAFE_COUNT <= 32, "");
}
/*
Destroy Query_tables_list object with freeing all resources used by it.
SYNOPSIS
destroy_query_tables_list()
*/
void Query_tables_list::destroy_query_tables_list() { sroutines.reset(); }
/*
Initialize LEX object.
SYNOPSIS
LEX::LEX()
NOTE
LEX object initialized with this constructor can be used as part of
THD object for which one can safely call open_tables(), lock_tables()
and close_thread_tables() functions. But it is not yet ready for
statement parsing. On should use lex_start() function to prepare LEX
for this.
*/
LEX::LEX()
: unit(nullptr),
query_block(nullptr),
all_query_blocks_list(nullptr),
m_current_query_block(nullptr),
result(nullptr),
thd(nullptr),
opt_hints_global(nullptr),
// Quite unlikely to overflow initial allocation, so no instrumentation.
plugins(PSI_NOT_INSTRUMENTED),
insert_update_values_map(nullptr),
option_type(OPT_DEFAULT),
drop_temporary(false),
sphead(nullptr),
// Initialize here to avoid uninitialized variable warnings.
contains_plaintext_password(false),
keep_diagnostics(DA_KEEP_UNSPECIFIED),
is_lex_started(false),
in_update_value_clause(false),
will_contextualize(true) {
reset_query_tables_list(true);
}
/**
check if command can use VIEW with MERGE algorithm (for top VIEWs)
@details
Only listed here commands can use merge algorithm in top level
Query_block (for subqueries will be used merge algorithm if
LEX::can_not_use_merged() is not true).
@todo - Add SET as a command that can use merged views. Due to how
all uses would be embedded in subqueries, this test is worthless
for the SET command anyway.
@returns true if command can use merged VIEWs, false otherwise
*/
bool LEX::can_use_merged() {
switch (sql_command) {
case SQLCOM_SELECT:
case SQLCOM_CREATE_TABLE:
case SQLCOM_UPDATE:
case SQLCOM_UPDATE_MULTI:
case SQLCOM_DELETE:
case SQLCOM_DELETE_MULTI:
case SQLCOM_INSERT:
case SQLCOM_INSERT_SELECT:
case SQLCOM_REPLACE:
case SQLCOM_REPLACE_SELECT:
case SQLCOM_LOAD:
/*
With WL#6599 following SHOW commands are implemented over the
INFORMATION_SCHEMA system views, and we do not create
temporary tables anymore now. So these queries should be
allowed to be mergeable, which makes the INFORMATION_SCHEMA
query execution faster.
According to optimizer team (Roy), making this decision based
on the command type here is a hack. This should probably change when we
introduce Sql_cmd_show class, which should treat the following SHOW
commands same as SQLCOM_SELECT.
*/
case SQLCOM_SHOW_CHARSETS:
case SQLCOM_SHOW_COLLATIONS:
case SQLCOM_SHOW_DATABASES:
case SQLCOM_SHOW_EVENTS:
case SQLCOM_SHOW_FIELDS:
case SQLCOM_SHOW_KEYS:
case SQLCOM_SHOW_STATUS_FUNC:
case SQLCOM_SHOW_STATUS_PROC:
case SQLCOM_SHOW_TABLES:
case SQLCOM_SHOW_TABLE_STATUS:
case SQLCOM_SHOW_TRIGGERS:
return true;
default:
return false;
}
}
/**
Check if command can't use merged views in any part of command
@details
Temporary table algorithm will be used on all SELECT levels for queries
listed here (see also LEX::can_use_merged()).
@returns true if command cannot use merged view, false otherwise
*/
bool LEX::can_not_use_merged() {
switch (sql_command) {
case SQLCOM_CREATE_VIEW:
case SQLCOM_SHOW_CREATE:
return true;
default:
return false;
}
}
/*
case SQLCOM_REVOKE_ROLE:
case SQLCOM_GRANT_ROLE:
Should Items_ident be printed correctly
SYNOPSIS
need_correct_ident()
RETURN
true yes, we need only structure
false no, we need data
*/
bool LEX::need_correct_ident() {
if (is_explain() && explain_format->is_iterator_based()) return true;
switch (sql_command) {
case SQLCOM_SHOW_CREATE:
case SQLCOM_SHOW_TABLES:
case SQLCOM_CREATE_VIEW:
return true;
default:
return false;
}
}
/**
This method should be called only during parsing.
It is aware of compound statements (stored routine bodies)
and will initialize the destination with the default
database of the stored routine, rather than the default
database of the connection it is parsed in.
E.g. if one has no current database selected, or current database
set to 'bar' and then issues:
CREATE PROCEDURE foo.p1() BEGIN SELECT * FROM t1 END//
t1 is meant to refer to foo.t1, not to bar.t1.
This method is needed to support this rule.
@return true in case of error (parsing should be aborted, false in
case of success
*/
bool LEX::copy_db_to(char const **p_db, size_t *p_db_length) const {
/*
When both is_explain_for_schema() and sphead() are true, it means EXPLAIN
FOR SCHEMA is called from inside a procedure. In such case, the EXPLAIN db
overrides the procedure's db for that statement.
*/
if (!(is_explain() && explain_format->is_explain_for_schema()) && sphead) {
assert(sphead->m_db.str && sphead->m_db.length);
/*
It is safe to assign the string by-pointer, both sphead and
its statements reside in the same memory root.
*/
*p_db = sphead->m_db.str;
if (p_db_length) *p_db_length = sphead->m_db.length;
return false;
}
return thd->copy_db_to(p_db, p_db_length);
}
/**
Set limit and offset for query expression object
@param thd thread handler
@param provider Query_block to get offset and limit from.
@returns false if success, true if error
*/
bool Query_expression::set_limit(THD *thd, Query_block *provider) {
offset_limit_cnt = provider->get_offset(thd);
select_limit_cnt = provider->get_limit(thd);
if (select_limit_cnt + offset_limit_cnt >= select_limit_cnt)
select_limit_cnt += offset_limit_cnt;
else
select_limit_cnt = HA_POS_ERROR;
return false;
}
/**
Checks if this query expression has limit defined. For a query expression
with set operation it checks if any of the query blocks has limit defined.
@returns true if the query expression has limit.
false otherwise.
*/
bool Query_expression::has_any_limit() const {
for (auto qt : query_terms<>())
if (qt->query_block()->has_limit()) return true;
return false;
}
/**
Include a query expression below a query block.
@param lex Containing LEX object
@param outer The query block that this query expression is included below.
*/
void Query_expression::include_down(LEX *lex, Query_block *outer) {
if ((next = outer->slave)) next->prev = &next;
prev = &outer->slave;
outer->slave = this;
master = outer;
renumber_selects(lex);
}
/**
Return true if query expression can be merged into an outer query, based on
technical constraints.
Being mergeable also means that derived table/view is updatable.
A view/derived table is not mergeable if it is one of the following:
- A set operation (implementation restriction).
- An aggregated query, or has HAVING, or has DISTINCT
(A general aggregated query cannot be merged with a non-aggregated one).
- A table-less query (unimportant special case).
- A query with a LIMIT (limit applies to subquery, so the implementation
strategy is to materialize this subquery, including row count constraint).
- It has windows
*/
bool Query_expression::is_mergeable() const {
if (is_set_operation()) return false;
Query_block *const select = first_query_block();
return !select->is_grouped() && select->having_cond() == nullptr &&
!select->is_distinct() && select->has_tables() &&
!select->has_limit() && !select->has_windows();
}
/**
True if heuristics suggest to merge this query expression.
A view/derived table is not suggested for merging if it contains subqueries
in the SELECT list that depend on columns from itself.
Merging such objects is possible, but we assume they are made derived
tables because the user wants them to be materialized, for performance
reasons.
One possible case is a derived table with dependent subqueries in the select
list, used as the inner table of a left outer join. Such tables will always
be read as many times as there are qualifying rows in the outer table,
and the select list subqueries are evaluated for each row combination.
The select list subqueries are evaluated the same number of times also with
join buffering enabled, even though the table then only will be read once.
Another case is, a query that modifies variables: then try to preserve the
original structure of the query. This is less likely to cause changes in
variable assignment order.
*/
bool Query_expression::merge_heuristic(const LEX *lex) const {
if (lex->set_var_list.elements != 0) return false;
Query_block *const select = first_query_block();
for (Item *item : select->visible_fields()) {
if (item->has_subquery() && !item->const_for_execution()) return false;
}
return true;
}
/**
Renumber contained query_block objects.
@param lex Containing LEX object
*/
void Query_expression::renumber_selects(LEX *lex) {
for (auto qt : query_terms<>()) qt->query_block()->renumber(lex);
}
/**
Save prepared statement properties for a query expression and underlying
query blocks. Required for repeated optimizations of the command.
@param thd thread handler
@returns false if success, true if error (out of memory)
*/
bool Query_expression::save_cmd_properties(THD *thd) {
assert(is_prepared());
for (auto qt : query_terms<>()) qt->query_block()->save_cmd_properties(thd);
return false;
}
/**
Loop over all query blocks and restore information needed for optimization,
including binding data for all associated tables.
*/
void Query_expression::restore_cmd_properties() {
for (auto qt : query_terms<>()) {
qt->query_block()->restore_cmd_properties();
}
}
/**
@brief Set the initial purpose of this Table_ref object in the list of
used tables.
We need to track this information on table-by-table basis, since when this
table becomes an element of the pre-locked list, it's impossible to identify
which SQL sub-statement it has been originally used in.
E.g.:
User request: SELECT * FROM t1 WHERE f1();
FUNCTION f1(): DELETE FROM t2; RETURN 1;
BEFORE DELETE trigger on t2: INSERT INTO t3 VALUES (old.a);
For this user request, the pre-locked list will contain t1, t2, t3
table elements, each needed for different DML.
The trigger event map is updated to reflect INSERT, UPDATE, DELETE,
REPLACE, LOAD DATA, CREATE TABLE .. SELECT, CREATE TABLE ..
REPLACE SELECT statements, and additionally ON DUPLICATE KEY UPDATE
clause.
*/
void LEX::set_trg_event_type_for_tables() {
uint8 new_trg_event_map = 0;
/*
Some auxiliary operations
(e.g. GRANT processing) create Table_ref instances outside
the parser. Additionally, some commands (e.g. OPTIMIZE) change
the lock type for a table only after parsing is done. Luckily,
these do not fire triggers and do not need to pre-load them.
For these TABLE_LISTs set_trg_event_type is never called, and
trg_event_map is always empty. That means that the pre-locking
algorithm will ignore triggers defined on these tables, if
any, and the execution will either fail with an assert in
sql_trigger.cc or with an error that a used table was not
pre-locked, in case of a production build.
TODO: this usage pattern creates unnecessary module dependencies
and should be rewritten to go through the parser.
Table list instances created outside the parser in most cases
refer to mysql.* system tables. It is not allowed to have
a trigger on a system table, but keeping track of
initialization provides extra safety in case this limitation
is circumvented.
*/
switch (sql_command) {
case SQLCOM_LOCK_TABLES:
/*
On a LOCK TABLE, all triggers must be pre-loaded for this
Table_ref when opening an associated TABLE.
*/
new_trg_event_map =
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_INSERT)) |
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_UPDATE)) |
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_DELETE));
break;
case SQLCOM_INSERT:
case SQLCOM_INSERT_SELECT:
/*
Basic INSERT. If there is an additional ON DUPLICATE KEY
UPDATE clause, it will be handled later in this method.
*/
case SQLCOM_LOAD:
/*
LOAD DATA ... INFILE is expected to fire BEFORE/AFTER
INSERT triggers. If the statement also has REPLACE clause, it will be
handled later in this method.
*/
case SQLCOM_REPLACE:
case SQLCOM_REPLACE_SELECT:
/*
REPLACE is semantically equivalent to INSERT. In case
of a primary or unique key conflict, it deletes the old
record and inserts a new one. So we also may need to
fire ON DELETE triggers. This functionality is handled
later in this method.
*/
case SQLCOM_CREATE_TABLE:
/*
CREATE TABLE ... SELECT defaults to INSERT if the table
or view already exists. REPLACE option of CREATE TABLE ... REPLACE
SELECT is handled later in this method.
*/
new_trg_event_map |=
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_INSERT));
break;
case SQLCOM_UPDATE:
case SQLCOM_UPDATE_MULTI:
/* Basic update and multi-update */
new_trg_event_map |=
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_UPDATE));
break;
case SQLCOM_DELETE:
case SQLCOM_DELETE_MULTI:
/* Basic delete and multi-delete */
new_trg_event_map |=
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_DELETE));
break;
default:
break;
}
switch (duplicates) {
case DUP_UPDATE:
new_trg_event_map |=
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_UPDATE));
break;
case DUP_REPLACE:
new_trg_event_map |=
static_cast<uint8>(1 << static_cast<int>(TRG_EVENT_DELETE));
break;
case DUP_ERROR:
default:
break;
}
/*
Do not iterate over sub-selects, only the tables in the outermost
Query_block can be modified, if any.
*/
Table_ref *tables = query_block ? query_block->get_table_list() : nullptr;
while (tables) {
/*
This is a fast check to filter out statements that do
not change data, or tables on the right side, in case of
INSERT .. SELECT, CREATE TABLE .. SELECT and so on.
Here we also filter out OPTIMIZE statement and non-updateable
views, for which lock_type is TL_UNLOCK or TL_READ after
parsing.
*/
if (static_cast<int>(tables->lock_descriptor().type) >=
static_cast<int>(TL_WRITE_ALLOW_WRITE))
tables->trg_event_map = new_trg_event_map;
tables = tables->next_local;
}
}
/*
Unlink the first table from the global table list and the first table from
outer select (lex->query_block) local list
SYNOPSIS
unlink_first_table()
link_to_local Set to 1 if caller should link this table to local list
NOTES
We assume that first tables in both lists is the same table or the local
list is empty.
RETURN
0 If 'query_tables' == 0
unlinked table
In this case link_to_local is set.
*/
Table_ref *LEX::unlink_first_table(bool *link_to_local) {
Table_ref *first;
if ((first = query_tables)) {
/*
Exclude from global table list
*/
if ((query_tables = query_tables->next_global))
query_tables->prev_global = &query_tables;
else
query_tables_last = &query_tables;
first->next_global = nullptr;
if (query_tables_own_last == &first->next_global)
query_tables_own_last = &query_tables;
/*
and from local list if it is not empty
*/
if ((*link_to_local = query_block->get_table_list() != nullptr)) {
query_block->context.table_list =
query_block->context.first_name_resolution_table = first->next_local;
query_block->m_table_list.first = first->next_local;
query_block->m_table_list.elements--; // safety
first->next_local = nullptr;
/*
Ensure that the global list has the same first table as the local
list.
*/
first_lists_tables_same();
}
}
return first;
}
/*
Bring first local table of first most outer select to first place in global
table list
SYNOPSIS
LEX::first_lists_tables_same()
NOTES
In many cases (for example, usual INSERT/DELETE/...) the first table of
main Query_block have special meaning => check that it is the first table
in global list and re-link to be first in the global list if it is
necessary. We need such re-linking only for queries with sub-queries in
the select list, as only in this case tables of sub-queries will go to
the global list first.
*/
void LEX::first_lists_tables_same() {
Table_ref *first_table = query_block->get_table_list();
if (query_tables != first_table && first_table != nullptr) {
Table_ref *next;
if (query_tables_last == &first_table->next_global)
query_tables_last = first_table->prev_global;
if (query_tables_own_last == &first_table->next_global)
query_tables_own_last = first_table->prev_global;
if ((next = *first_table->prev_global = first_table->next_global))
next->prev_global = first_table->prev_global;
/* include in new place */
first_table->next_global = query_tables;
/*
We are sure that query_tables is not 0, because first_table was not
first table in the global list => we can use
query_tables->prev_global without check of query_tables
*/
query_tables->prev_global = &first_table->next_global;
first_table->prev_global = &query_tables;
query_tables = first_table;
}
}
/*
Link table back that was unlinked with unlink_first_table()
SYNOPSIS
link_first_table_back()
link_to_local do we need link this table to local
RETURN
global list
*/
void LEX::link_first_table_back(Table_ref *first, bool link_to_local) {
if (first) {
if ((first->next_global = query_tables))
query_tables->prev_global = &first->next_global;
else
query_tables_last = &first->next_global;
if (query_tables_own_last == &query_tables)
query_tables_own_last = &first->next_global;
query_tables = first;
if (link_to_local) {
first->next_local = query_block->m_table_list.first;
query_block->context.table_list = first;
query_block->m_table_list.first = first;
query_block->m_table_list.elements++; // safety
}
}
}
/*
cleanup lex for case when we open table by table for processing
SYNOPSIS
LEX::cleanup_after_one_table_open()
NOTE
This method is mostly responsible for cleaning up of selects lists and
derived tables state. To rollback changes in Query_tables_list one has
to call Query_tables_list::reset_query_tables_list(false).
*/
void LEX::cleanup_after_one_table_open() {
if (all_query_blocks_list != query_block) {
/* cleunup underlying units (units of VIEW) */
for (Query_expression *un = query_block->first_inner_query_expression(); un;
un = un->next_query_expression()) {
un->cleanup(true);
un->destroy();
}
/* reduce all selects list to default state */
all_query_blocks_list = query_block;
/* remove underlying units (units of VIEW) subtree */
query_block->cut_subtree();
}
}
/*
Save current state of Query_tables_list for this LEX, and prepare it
for processing of new statemnt.
SYNOPSIS
reset_n_backup_query_tables_list()
backup Pointer to Query_tables_list instance to be used for backup
*/
void LEX::reset_n_backup_query_tables_list(Query_tables_list *backup) {
backup->set_query_tables_list(this);
/*
We have to perform full initialization here since otherwise we
will damage backed up state.
*/
this->reset_query_tables_list(true);
}
/*
Restore state of Query_tables_list for this LEX from backup.
SYNOPSIS
restore_backup_query_tables_list()
backup Pointer to Query_tables_list instance used for backup
*/
void LEX::restore_backup_query_tables_list(Query_tables_list *backup) {
this->destroy_query_tables_list();
this->set_query_tables_list(backup);
}
/*
Checks for usage of routines and/or tables in a parsed statement
SYNOPSIS
LEX:table_or_sp_used()
RETURN
false No routines and tables used
true Either or both routines and tables are used.
*/
bool LEX::table_or_sp_used() {
DBUG_TRACE;
if ((sroutines != nullptr && !sroutines->empty()) || query_tables)
return true;
return false;
}
/**
Locate an assignment to a user variable with a given name, within statement.
@param name Name of variable to search for
@returns true if variable is assigned to, false otherwise.
*/
bool LEX::locate_var_assignment(const Name_string &name) {
List_iterator<Item_func_set_user_var> li(set_var_list);
Item_func_set_user_var *var;
while ((var = li++)) {
if (var->name.eq(name)) return true;
}
return false;
}
/**
Save properties for ORDER clauses so that they can be reconstructed
for a new optimization of the query block.
@param thd thread handler
@param list list of ORDER elements to be saved
@param[out] list_ptrs Saved list of ORDER elements
@returns false if success, true if error (out of memory)
*/
bool Query_block::save_order_properties(THD *thd, SQL_I_List<ORDER> *list,
Group_list_ptrs **list_ptrs) {
assert(*list_ptrs == nullptr);
void *mem = thd->stmt_arena->alloc(sizeof(Group_list_ptrs));
if (mem == nullptr) return true;
Group_list_ptrs *p = new (mem) Group_list_ptrs(thd->stmt_arena->mem_root);
if (p == nullptr) return true;
*list_ptrs = p;
if (p->reserve(list->elements)) return true;
for (ORDER *order = list->first; order; order = order->next)
if (p->push_back(order)) return true;
return false;
}
/**
Save properties of a prepared statement needed for repeated optimization.
Saves the chain of ORDER::next in group_list and order_list, in
case the list is modified by remove_const().
@param thd thread handler
@returns false if success, true if error (out of memory)
*/
bool Query_block::save_properties(THD *thd) {
assert(first_execution);
first_execution = false;
assert(!thd->stmt_arena->is_regular());
if (thd->stmt_arena->is_regular()) return false;
saved_cond_count = cond_count;
if (!base_ref_items.empty()) {
m_saved_base_items =
base_ref_items.prefix(fields.size()).Clone(thd->mem_root);
}
if (group_list.first &&
save_order_properties(thd, &group_list, &group_list_ptrs))
return true;
if (order_list.first &&
save_order_properties(thd, &order_list, &order_list_ptrs))
return true;
return false;
}
static enum_explain_type setop2result(Query_term *qt) {
switch (qt->term_type()) {
case QT_UNION:
return enum_explain_type::EXPLAIN_UNION_RESULT;
case QT_INTERSECT:
return enum_explain_type::EXPLAIN_INTERSECT_RESULT;
case QT_EXCEPT:
return enum_explain_type::EXPLAIN_EXCEPT_RESULT;
case QT_UNARY:
return enum_explain_type::EXPLAIN_UNARY_RESULT;
default:
assert(false);
}
return enum_explain_type::EXPLAIN_UNION_RESULT;
}
/*
There are Query_block::add_table_to_list &
Query_block::set_lock_for_tables are in sql_parse.cc
Query_block::print is in sql_select.cc
Query_expression::prepare, Query_expression::exec,
Query_expression::cleanup, Query_expression::change_query_result
are in sql_union.cc
*/
enum_explain_type Query_block::type() const {
Query_term *qt = master_query_expression()->find_blocks_query_term(this);
if (qt->term_type() != QT_QUERY_BLOCK) {
return setop2result(qt);
} else if (!master_query_expression()->outer_query_block() &&
master_query_expression()->first_query_block() == this) {
if (first_inner_query_expression() || next_query_block() ||
m_parent != nullptr)
return enum_explain_type::EXPLAIN_PRIMARY;
else
return enum_explain_type::EXPLAIN_SIMPLE;
} else if (this == master_query_expression()->first_query_block()) {
if (linkage == DERIVED_TABLE_TYPE)
return enum_explain_type::EXPLAIN_DERIVED;
else
return enum_explain_type::EXPLAIN_SUBQUERY;
} else {
assert(m_parent != nullptr);
// if left child, call block PRIMARY, else UNION/INTERSECT/EXCEPT
switch (m_parent->term_type()) {
case QT_EXCEPT:
if (m_parent->m_children[0] == this)
return enum_explain_type::EXPLAIN_PRIMARY;
else
return enum_explain_type::EXPLAIN_EXCEPT;
case QT_UNION:
if (m_parent->m_children[0] == this)
return enum_explain_type::EXPLAIN_PRIMARY;
else
return enum_explain_type::EXPLAIN_UNION;
case QT_INTERSECT:
if (m_parent->m_children[0] == this)
return enum_explain_type::EXPLAIN_PRIMARY;
else
return enum_explain_type::EXPLAIN_INTERSECT;
case QT_UNARY:
return enum_explain_type::EXPLAIN_PRIMARY;
default:
assert(false);
}
return enum_explain_type::EXPLAIN_UNION;
}
}
/**
Add this query block below the specified query expression.
@param lex Containing LEX object
@param outer Query expression that query block is added to.
@note that this query block can never have any underlying query expressions,
hence it is not necessary to e.g. renumber those, like e.g.
Query_expression::include_down() does.
*/
void Query_block::include_down(LEX *lex, Query_expression *outer) {
assert(slave == nullptr);
next = outer->slave;
outer->slave = this;
master = outer;
select_number = ++lex->select_number;
nest_level =
outer_query_block() == nullptr ? 0 : outer_query_block()->nest_level + 1;
}
/**
Add this query block after the specified query block.
@param lex Containing LEX object
@param before Query block that this object is added after.
*/
void Query_block::include_neighbour(LEX *lex, Query_block *before) {
next = before->next;
before->next = this;
master = before->master;
select_number = ++lex->select_number;
nest_level = before->nest_level;
}
/**
Include query block within the supplied unit.
Do not link the query block into the global chain of query blocks.
This function is exclusive for Query_expression::add_fake_query_block() -
use it with caution.
@param outer Query expression this node is included below.
*/
void Query_block::include_standalone(Query_expression *outer) {
next = nullptr;
master = outer;
nest_level = master->first_query_block()->nest_level;
}
/**
Renumber query_block object, and apply renumbering recursively to
contained objects.
@param lex Containing LEX object
*/
void Query_block::renumber(LEX *lex) {
select_number = ++lex->select_number;
nest_level =
outer_query_block() == nullptr ? 0 : outer_query_block()->nest_level + 1;
for (Query_expression *u = first_inner_query_expression(); u;
u = u->next_query_expression())
u->renumber_selects(lex);
}
/**
Include query block into global list.
@param plink - Pointer to start of list
*/
void Query_block::include_in_global(Query_block **plink) {
if ((link_next = *plink)) link_next->link_prev = &link_next;
link_prev = plink;
*plink = this;
}
/**
Include chain of query blocks into global list.
@param start - Pointer to start of list
*/
void Query_block::include_chain_in_global(Query_block **start) {
Query_block *last_query_block;
for (last_query_block = this; last_query_block->link_next != nullptr;
last_query_block = last_query_block->link_next) {
}
last_query_block->link_next = *start;
last_query_block->link_next->link_prev = &last_query_block->link_next;
link_prev = start;
*start = this;
}
/**
Helper function which handles the "ON conditions" part of
Query_block::get_optimizable_conditions().
@returns true if OOM
*/
static bool get_optimizable_join_conditions(
THD *thd, mem_root_deque<Table_ref *> &join_list) {
for (Table_ref *table : join_list) {
NESTED_JOIN *const nested_join = table->nested_join;
if (nested_join &&
get_optimizable_join_conditions(thd, nested_join->m_tables))
return true;
Item *const jc = table->join_cond();
if (jc && !thd->stmt_arena->is_regular()) {
table->set_join_cond_optim(jc->copy_andor_structure(thd));
if (!table->join_cond_optim()) return true;
} else
table->set_join_cond_optim(jc);
}
return false;
}
/**
Returns disposable copies of WHERE/HAVING/ON conditions.
This function returns a copy which can be thrashed during
this execution of the statement. Only AND/OR items are trashable!
If in conventional execution, no copy is created, the permanent clauses are
returned instead, as trashing them is no problem.
@param thd thread handle
@param[out] new_where copy of WHERE
@param[out] new_having copy of HAVING (if passed pointer is not NULL)
Copies of join (ON) conditions are placed in
Table_ref::m_join_cond_optim.
@returns true if OOM
*/
bool Query_block::get_optimizable_conditions(THD *thd, Item **new_where,
Item **new_having) {
/*
We want to guarantee that
join->optimized is true => conditions are ready for reading.
So if we are here, this should hold:
*/
assert(!(join && join->is_optimized()));
if (m_where_cond && !thd->stmt_arena->is_regular()) {
*new_where = m_where_cond->copy_andor_structure(thd);
if (!*new_where) return true;
} else
*new_where = m_where_cond;
if (new_having) {
if (m_having_cond && !thd->stmt_arena->is_regular()) {
*new_having = m_having_cond->copy_andor_structure(thd);
if (!*new_having) return true;
} else
*new_having = m_having_cond;
}
return get_optimizable_join_conditions(thd, m_table_nest);
}
Subquery_strategy Query_block::subquery_strategy(const THD *thd) const {
if (m_windows.elements > 0)
/*
A window function is in the SELECT list.
In-to-exists could not work: it would attach an equality like
outer_expr = WF to either WHERE or HAVING; but a WF is not allowed in
those clauses, and even if we allowed it, it would modify the result
rows over which the WF is supposed to be calculated.
So, subquery materialization is imposed. Grep for (and read) WL#10431.
*/
return Subquery_strategy::SUBQ_MATERIALIZATION;
if (opt_hints_qb) {
const Subquery_strategy strategy = opt_hints_qb->subquery_strategy();
if (strategy != Subquery_strategy::UNSPECIFIED) return strategy;
}
// No SUBQUERY hint given, base possible strategies on optimizer_switch
if (thd->optimizer_switch_flag(OPTIMIZER_SWITCH_MATERIALIZATION))
return thd->optimizer_switch_flag(OPTIMIZER_SWITCH_SUBQ_MAT_COST_BASED)
? Subquery_strategy::CANDIDATE_FOR_IN2EXISTS_OR_MAT
: Subquery_strategy::SUBQ_MATERIALIZATION;
return Subquery_strategy::SUBQ_EXISTS;
}
bool Query_block::semijoin_enabled(const THD *thd) const {
return opt_hints_qb ? opt_hints_qb->semijoin_enabled(thd)
: thd->optimizer_switch_flag(OPTIMIZER_SWITCH_SEMIJOIN);
}
void Query_block::update_semijoin_strategies(THD *thd) {
const uint sj_strategy_mask =
OPTIMIZER_SWITCH_FIRSTMATCH | OPTIMIZER_SWITCH_LOOSE_SCAN |
OPTIMIZER_SWITCH_MATERIALIZATION | OPTIMIZER_SWITCH_DUPSWEEDOUT;
const uint opt_switches = thd->variables.optimizer_switch & sj_strategy_mask;
const bool is_secondary_engine_optimization =
parent_lex->m_sql_cmd != nullptr &&
parent_lex->m_sql_cmd->using_secondary_storage_engine();
for (Table_ref *sj_nest : sj_nests) {
/*
After semi-join transformation, original Query_block with hints is lost.
Fetch hints from last table in semijoin nest, as join_list has the
convention to list join operators' arguments in reverse order.
*/
Table_ref *table = sj_nest->nested_join->m_tables.back();
/*
Do not respect opt_hints_qb for secondary engine optimization.
Secondary storage engines may not support all strategies that are
supported by the MySQL executor. Secondary engines should set their
supported semi-join strategies in thd->variables.optimizer_switch and not
respect optimizer hints or optimizer switches specified by the user.
*/
sj_nest->nested_join->sj_enabled_strategies =
(table->opt_hints_qb && !is_secondary_engine_optimization)
? table->opt_hints_qb->sj_enabled_strategies(opt_switches)
: opt_switches;
if (sj_nest->is_aj_nest()) {
// only these are possible with NOT EXISTS/IN:
sj_nest->nested_join->sj_enabled_strategies &=
OPTIMIZER_SWITCH_FIRSTMATCH | OPTIMIZER_SWITCH_MATERIALIZATION |
OPTIMIZER_SWITCH_DUPSWEEDOUT;
}
}
}
/**
Check if an option that can be used only for an outer-most query block is
applicable to this query block.
@param lex LEX of current statement
@param option option name to output within the error message
@returns false if valid, true if invalid, error is sent to client
*/
bool Query_block::validate_outermost_option(LEX *lex,
const char *option) const {
if (this != lex->query_block) {
my_error(ER_CANT_USE_OPTION_HERE, MYF(0), option);
return true;
}
return false;
}
/**
Validate base options for a query block.
@param lex LEX of current statement
@param options_arg base options for a SELECT statement.
@returns false if success, true if validation failed
These options are supported, per DML statement:
SELECT: SELECT_STRAIGHT_JOIN
SELECT_HIGH_PRIORITY
SELECT_DISTINCT
SELECT_ALL
SELECT_SMALL_RESULT
SELECT_BIG_RESULT
OPTION_BUFFER_RESULT
OPTION_FOUND_ROWS
OPTION_SELECT_FOR_SHOW
DELETE: OPTION_QUICK
LOW_PRIORITY
INSERT: LOW_PRIORITY
HIGH_PRIORITY
UPDATE: LOW_PRIORITY
Note that validation is only performed for SELECT statements.
*/
bool Query_block::validate_base_options(LEX *lex, ulonglong options_arg) const {
assert(!(options_arg & ~(SELECT_STRAIGHT_JOIN | SELECT_HIGH_PRIORITY |
SELECT_DISTINCT | SELECT_ALL | SELECT_SMALL_RESULT |
SELECT_BIG_RESULT | OPTION_BUFFER_RESULT |
OPTION_FOUND_ROWS | OPTION_SELECT_FOR_SHOW)));
if (options_arg & SELECT_DISTINCT && options_arg & SELECT_ALL) {
my_error(ER_WRONG_USAGE, MYF(0), "ALL", "DISTINCT");
return true;
}
if (options_arg & SELECT_HIGH_PRIORITY &&
validate_outermost_option(lex, "HIGH_PRIORITY"))
return true;
if (options_arg & OPTION_BUFFER_RESULT &&
validate_outermost_option(lex, "SQL_BUFFER_RESULT"))
return true;
if (options_arg & OPTION_FOUND_ROWS &&
validate_outermost_option(lex, "SQL_CALC_FOUND_ROWS"))
return true;
return false;
}
/**
Apply walk() processor to join conditions.
JOINs may be nested. Walk nested joins recursively to apply the
processor.
*/
static bool walk_join_condition(mem_root_deque<Table_ref *> *tables,
Item_processor processor, enum_walk walk,
uchar *arg) {
for (const Table_ref *table : *tables) {
if (table->join_cond() && table->join_cond()->walk(processor, walk, arg))
return true;
if (table->nested_join != nullptr &&
walk_join_condition(&table->nested_join->m_tables, processor, walk,
arg))
return true;
}
return false;
}
void Query_expression::accumulate_used_tables(table_map map) {
assert(outer_query_block());
if (item)
item->accumulate_used_tables(map);
else if (m_lateral_deps)
m_lateral_deps |= map;
}
enum_parsing_context Query_expression::place() const {
assert(outer_query_block());
if (item != nullptr) return item->place();
return CTX_DERIVED;
}
bool Query_block::walk(Item_processor processor, enum_walk walk, uchar *arg) {
for (Item *item : visible_fields()) {
if (item->walk(processor, walk, arg)) return true;
}
if (m_current_table_nest != nullptr &&
walk_join_condition(m_current_table_nest, processor, walk, arg))
return true;
if ((walk & enum_walk::SUBQUERY)) {
/*
for each leaf: if a materialized table, walk the unit
*/
for (Table_ref *tbl = leaf_tables; tbl; tbl = tbl->next_leaf) {
if (!tbl->uses_materialization()) continue;
if (tbl->is_derived()) {
if (tbl->derived_query_expression()->walk(processor, walk, arg))
return true;
} else if (tbl->is_table_function()) {
if (tbl->table_function->walk(processor, walk, arg)) return true;
}
}
}
// @todo: Roy thinks that we should always use where_cond.
Item *const where_cond =
(join && join->is_optimized()) ? join->where_cond : this->where_cond();
if (where_cond && where_cond->walk(processor, walk, arg)) return true;
for (auto order = group_list.first; order; order = order->next) {
if ((*order->item)->walk(processor, walk, arg)) return true;
}
if (having_cond() && having_cond()->walk(processor, walk, arg)) return true;
for (auto order = order_list.first; order; order = order->next) {
if ((*order->item)->walk(processor, walk, arg)) return true;
}
// walk windows' ORDER BY and PARTITION BY clauses.
List_iterator<Window> liw(m_windows);
for (Window *w = liw++; w != nullptr; w = liw++) {
/*
We use first_order_by() instead of order() because if a window
references another window and they thus share the same ORDER BY,
we want to walk that clause only once here
(Same for partition as well)".
*/
for (auto it : {w->first_partition_by(), w->first_order_by()}) {
if (it != nullptr) {
for (ORDER *o = it; o != nullptr; o = o->next) {
if ((*o->item)->walk(processor, walk, arg)) return true;
}
}
}
}
return false;
}
/**
Finds a (possibly unresolved) table reference in the from clause by name.
There is a hack in the parser which adorns table references with the current
database. This function piggy-backs on that hack to find fully qualified
table references without having to resolve the name.
@param ident The table name, may be qualified or unqualified.
@retval NULL If not found.
*/
Table_ref *Query_block::find_table_by_name(const Table_ident *ident) {
const LEX_CSTRING db_name = ident->db;
const LEX_CSTRING table_name = ident->table;
for (Table_ref *table = m_table_list.first; table;
table = table->next_local) {
if ((db_name.length == 0 || strcmp(db_name.str, table->db) == 0) &&
strcmp(table_name.str, table->alias) == 0)
return table;
}
return nullptr;
}
/**
Save prepared statement properties for a query block and underlying
query expressions. Required for repeated optimizations of the command.
@param thd thread handler
@returns false if success, true if error (out of memory)
*/
bool Query_block::save_cmd_properties(THD *thd) {
for (Query_expression *u = first_inner_query_expression(); u;
u = u->next_query_expression())
if (u->save_cmd_properties(thd)) return true;
if (save_properties(thd)) return true;
for (Table_ref *tbl = leaf_tables; tbl; tbl = tbl->next_leaf) {
if (!tbl->is_base_table()) continue;
if (tbl->save_properties()) return true;
}
return false;
}
/**
Restore prepared statement properties for this query block and all
underlying query expressions so they are ready for optimization.
Restores properties saved in Table_ref objects into corresponding
TABLEs. Restores ORDER BY and GROUP by clauses, and window definitions, so
they are ready for optimization.
*/
void Query_block::restore_cmd_properties() {
// Restore base_ref_items. Do this before we dive into subqueries, so that
// their outer references point to valid items when they update used tables.
std::copy(m_saved_base_items.begin(), m_saved_base_items.end(),
base_ref_items.begin());
for (Query_expression *u = first_inner_query_expression(); u;
u = u->next_query_expression())
u->restore_cmd_properties();
for (Table_ref *tbl = leaf_tables; tbl; tbl = tbl->next_leaf) {
if (!tbl->is_base_table()) continue;
tbl->restore_properties();
tbl->table->m_record_buffer = Record_buffer{0, 0, nullptr};
}
assert(join == nullptr);
cond_count = saved_cond_count;
// Restore GROUP BY list
if (group_list_ptrs && group_list_ptrs->size() > 0) {
for (uint ix = 0; ix < group_list_ptrs->size() - 1; ++ix) {
ORDER *order = group_list_ptrs->at(ix);
order->next = group_list_ptrs->at(ix + 1);
}
}
// Restore ORDER BY list
if (order_list_ptrs && order_list_ptrs->size() > 0) {
for (uint ix = 0; ix < order_list_ptrs->size() - 1; ++ix) {
ORDER *order = order_list_ptrs->at(ix);
order->next = order_list_ptrs->at(ix + 1);
}
}
if (m_windows.elements > 0) {
List_iterator<Window> li(m_windows);
Window *w;
while ((w = li++)) w->reset_round();
}
/*
Unconditionally update used table information for all items referenced from
query block. This is required in case const table substitution, or other
kind of optimization, has been performed in earlier rounds.
*/
update_used_tables();
}
bool Query_options::merge(const Query_options &a, const Query_options &b) {
query_spec_options = a.query_spec_options | b.query_spec_options;
return false;
}
bool Query_options::save_to(Parse_context *pc) {
LEX *lex = pc->thd->lex;
const ulonglong options = query_spec_options;
if (pc->select->validate_base_options(lex, options)) return true;
pc->select->set_base_options(options);
return false;
}
bool LEX::accept(Select_lex_visitor *visitor) {
return m_sql_cmd->accept(thd, visitor);
}
bool LEX::set_wild(LEX_STRING w) {
if (w.str == nullptr) {
wild = nullptr;
return false;
}
wild = new (thd->mem_root) String(w.str, w.length, system_charset_info);
return wild == nullptr;
}
void LEX_SOURCE_INFO::initialize() {
host = user = password = log_file_name = bind_addr = nullptr;
network_namespace = nullptr;
port = connect_retry = 0;
heartbeat_period = 0;
sql_delay = 0;
pos = 0;
server_id = retry_count = 0;
gtid = nullptr;
gtid_until_condition = UNTIL_SQL_BEFORE_GTIDS;
view_id = nullptr;
until_after_gaps = false;
ssl = ssl_verify_server_cert = heartbeat_opt = repl_ignore_server_ids_opt =
retry_count_opt = auto_position = port_opt = get_public_key =
m_source_connection_auto_failover = m_gtid_only = LEX_MI_UNCHANGED;
ssl_key = ssl_cert = ssl_ca = ssl_capath = ssl_cipher = nullptr;
ssl_crl = ssl_crlpath = nullptr;
public_key_path = nullptr;
tls_version = nullptr;
tls_ciphersuites = UNSPECIFIED;
tls_ciphersuites_string = nullptr;
relay_log_name = nullptr;
relay_log_pos = 0;
repl_ignore_server_ids.clear();
channel = nullptr;
for_channel = false;
compression_algorithm = nullptr;
zstd_compression_level = 0;
privilege_checks_none = false;
privilege_checks_username = privilege_checks_hostname = nullptr;
require_row_format = LEX_MI_UNCHANGED;
require_table_primary_key_check = LEX_MI_PK_CHECK_UNCHANGED;
assign_gtids_to_anonymous_transactions_type =
LEX_MI_ANONYMOUS_TO_GTID_UNCHANGED;
assign_gtids_to_anonymous_transactions_manual_uuid = nullptr;
}
void LEX_SOURCE_INFO::set_unspecified() {
initialize();
sql_delay = -1;
}
uint binlog_unsafe_map[256];
#define UNSAFE(a, b, c) \
{ \
DBUG_PRINT("unsafe_mixed_statement", \
("SETTING BASE VALUES: %s, %s, %02X\n", \
LEX::stmt_accessed_table_string(a), \
LEX::stmt_accessed_table_string(b), c)); \
unsafe_mixed_statement(a, b, c); \
}
/*
Sets the combination given by "a" and "b" and automatically combinations
given by other types of access, i.e. 2^(8 - 2), as unsafe.
It may happen a collision when automatically defining a combination as unsafe.
For that reason, a combination has its unsafe condition redefined only when
the new_condition is greater then the old. For instance,
. (BINLOG_DIRECT_ON & TRX_CACHE_NOT_EMPTY) is never overwritten by
. (BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF).
*/
static void unsafe_mixed_statement(LEX::enum_stmt_accessed_table a,
LEX::enum_stmt_accessed_table b,
uint condition) {
int type = 0;
const int index = (1U << a) | (1U << b);
for (type = 0; type < 256; type++) {
if ((type & index) == index) {
binlog_unsafe_map[type] |= condition;
}
}
}
/**
Uses parse_tree to instantiate an Sql_cmd object and assigns it to the Lex.
@param parse_tree The parse tree.
@returns false on success, true on error.
*/
bool LEX::make_sql_cmd(Parse_tree_root *parse_tree) {
if (!will_contextualize) return false;
m_sql_cmd = parse_tree->make_cmd(thd);
if (m_sql_cmd == nullptr) return true;
assert(m_sql_cmd->sql_command_code() == sql_command);
return false;
}
/**
Set replication channel name
@param name If @p name is null string then reset channel name to default.
Otherwise set it to @p name.
@returns false if success, true if error (OOM).
*/
bool LEX::set_channel_name(LEX_CSTRING name) {
if (name.str == nullptr) {
mi.channel = "";
mi.for_channel = false;
} else {
/*
Channel names are case insensitive. This means, even the results
displayed to the user are converted to lower cases.
system_charset_info is utf8mb3_general_ci as required by channel name
restrictions
*/
char *buf = thd->strmake(name.str, name.length);
if (buf == nullptr) return true; // OOM
my_casedn_str(system_charset_info, buf);
mi.channel = buf;
mi.for_channel = true;
}
return false;
}
/*
The BINLOG_* AND TRX_CACHE_* values can be combined by using '&' or '|',
which means that both conditions need to be satisfied or any of them is
enough. For example,
. BINLOG_DIRECT_ON & TRX_CACHE_NOT_EMPTY means that the statement is
unsafe when the option is on and trx-cache is not empty;
. BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF means the statement is unsafe
in all cases.
. TRX_CACHE_EMPTY | TRX_CACHE_NOT_EMPTY means the statement is unsafe
in all cases. Similar as above.
*/
void binlog_unsafe_map_init() {
memset((void *)binlog_unsafe_map, 0, sizeof(uint) * 256);
/*
Classify a statement as unsafe when there is a mixed statement and an
on-going transaction at any point of the execution if:
1. The mixed statement is about to update a transactional table and
a non-transactional table.
2. The mixed statement is about to update a transactional table and
read from a non-transactional table.
3. The mixed statement is about to update a non-transactional table
and temporary transactional table.
4. The mixed statement is about to update a temporary transactional
table and read from a non-transactional table.
5. The mixed statement is about to update a transactional table and
a temporary non-transactional table.
6. The mixed statement is about to update a transactional table and
read from a temporary non-transactional table.
7. The mixed statement is about to update a temporary transactional
table and temporary non-transactional table.
8. The mixed statement is about to update a temporary transactional
table and read from a temporary non-transactional table.
After updating a transactional table if:
9. The mixed statement is about to update a non-transactional table
and read from a transactional table.
10. The mixed statement is about to update a non-transactional table
and read from a temporary transactional table.
11. The mixed statement is about to update a temporary non-transactional
table and read from a transactional table.
12. The mixed statement is about to update a temporary non-transactional
table and read from a temporary transactional table.
13. The mixed statement is about to update a temporary non-transactional
table and read from a non-transactional table.
The reason for this is that locks acquired may not protected a concurrent
transaction of interfering in the current execution and by consequence in
the result.
*/
/* Case 1. */
UNSAFE(LEX::STMT_WRITES_TRANS_TABLE, LEX::STMT_WRITES_NON_TRANS_TABLE,
BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF);
/* Case 2. */
UNSAFE(LEX::STMT_WRITES_TRANS_TABLE, LEX::STMT_READS_NON_TRANS_TABLE,
BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF);
/* Case 3. */
UNSAFE(LEX::STMT_WRITES_NON_TRANS_TABLE, LEX::STMT_WRITES_TEMP_TRANS_TABLE,
BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF);
/* Case 4. */
UNSAFE(LEX::STMT_WRITES_TEMP_TRANS_TABLE, LEX::STMT_READS_NON_TRANS_TABLE,
BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF);
/* Case 5. */
UNSAFE(LEX::STMT_WRITES_TRANS_TABLE, LEX::STMT_WRITES_TEMP_NON_TRANS_TABLE,
BINLOG_DIRECT_ON);
/* Case 6. */
UNSAFE(LEX::STMT_WRITES_TRANS_TABLE, LEX::STMT_READS_TEMP_NON_TRANS_TABLE,
BINLOG_DIRECT_ON);
/* Case 7. */
UNSAFE(LEX::STMT_WRITES_TEMP_TRANS_TABLE,
LEX::STMT_WRITES_TEMP_NON_TRANS_TABLE, BINLOG_DIRECT_ON);
/* Case 8. */
UNSAFE(LEX::STMT_WRITES_TEMP_TRANS_TABLE,
LEX::STMT_READS_TEMP_NON_TRANS_TABLE, BINLOG_DIRECT_ON);
/* Case 9. */
UNSAFE(LEX::STMT_WRITES_NON_TRANS_TABLE, LEX::STMT_READS_TRANS_TABLE,
(BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF) & TRX_CACHE_NOT_EMPTY);
/* Case 10 */
UNSAFE(LEX::STMT_WRITES_NON_TRANS_TABLE, LEX::STMT_READS_TEMP_TRANS_TABLE,
(BINLOG_DIRECT_ON | BINLOG_DIRECT_OFF) & TRX_CACHE_NOT_EMPTY);
/* Case 11. */
UNSAFE(LEX::STMT_WRITES_TEMP_NON_TRANS_TABLE, LEX::STMT_READS_TRANS_TABLE,
BINLOG_DIRECT_ON & TRX_CACHE_NOT_EMPTY);
/* Case 12. */
UNSAFE(LEX::STMT_WRITES_TEMP_NON_TRANS_TABLE,
LEX::STMT_READS_TEMP_TRANS_TABLE,
BINLOG_DIRECT_ON & TRX_CACHE_NOT_EMPTY);
/* Case 13. */
UNSAFE(LEX::STMT_WRITES_TEMP_NON_TRANS_TABLE, LEX::STMT_READS_NON_TRANS_TABLE,
BINLOG_DIRECT_OFF & TRX_CACHE_NOT_EMPTY);
}
void LEX::set_secondary_engine_execution_context(
Secondary_engine_execution_context *context) {
assert(m_secondary_engine_context == nullptr || context == nullptr);
if (m_secondary_engine_context != nullptr)
::destroy_at(m_secondary_engine_context);
m_secondary_engine_context = context;
}
bool LEX::validate_use_in_old_optimizer() {
if (sql_command == SQLCOM_CREATE_VIEW || sql_command == SQLCOM_SHOW_CREATE)
return false;
if (can_execute_only_in_hypergraph_optimizer() &&
!using_hypergraph_optimizer()) {
// Certain queries cannot be executed in the old optimizer.
my_error(ER_SUPPORTED_ONLY_WITH_HYPERGRAPH, MYF(0),
get_only_supported_in_hypergraph_reason_str());
return true;
}
return false;
}
void LEX_GRANT_AS::cleanup() {
grant_as_used = false;
role_type = role_enum::ROLE_NONE;
user = nullptr;
role_list = nullptr;
}
LEX_GRANT_AS::LEX_GRANT_AS() { cleanup(); }
void get_select_options_str(ulonglong options, std::string *str) {
size_t len = str->length();
if ((options & SELECT_STRAIGHT_JOIN) != 0u)
str->append(STRING_WITH_LEN("straight_join "));
if ((options & SELECT_HIGH_PRIORITY) != 0u)
str->append(STRING_WITH_LEN("high_priority "));
if ((options & SELECT_DISTINCT) != 0u)
str->append(STRING_WITH_LEN("distinct "));
if ((options & SELECT_SMALL_RESULT) != 0u)
str->append(STRING_WITH_LEN("sql_small_result "));
if ((options & SELECT_BIG_RESULT) != 0u)
str->append(STRING_WITH_LEN("sql_big_result "));
if ((options & OPTION_BUFFER_RESULT) != 0u)
str->append(STRING_WITH_LEN("sql_buffer_result "));
if ((options & OPTION_FOUND_ROWS) != 0u)
str->append(STRING_WITH_LEN("sql_calc_found_rows "));
// Delete the last space character.
if (str->length() > len) str->pop_back();
}