Search moodle.org's
Developer Documentation
Developer Documentation
- Bug fixes for general core bugs in 4.3.x will end 7 October 2024 (12 months).
- Bug fixes for security issues in 4.3.x will end 21 April 2025 (18 months).
- PHP version: minimum PHP 8.0.0 Note: minimum PHP version has increased since Moodle 4.1. PHP 8.2.x is supported too.
/lib/dml/ -> mysqli_native_moodle_database.php (source)
Differences Between: [Versions 310 and 403] [Versions 311 and 403] [Versions 39 and 403] [Versions 400 and 403] [Versions 401 and 403] [Versions 402 and 403]
Native mysqli class representing moodle database interface.
| Copyright: | 2008 Petr Skoda (http://skodak.org) |
| License: | http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later |
| File Size: | 2276 lines (87 kb) |
| Included or required: | 3 times |
| Referenced: | 0 times |
| Includes or requires: | 4 files lib/dml/mysqli_native_moodle_temptables.php lib/dml/moodle_read_slave_trait.php lib/dml/mysqli_native_moodle_recordset.php lib/dml/moodle_database.php |
Defines 1 class
mysqli_native_moodle_database:: (81 methods):
create_database()
driver_installed()
get_dbfamily()
get_dbtype()
get_dblibrary()
get_dbengine()
get_dbcollation()
detect_collation()
is_antelope_file_format_no_more_supported()
get_row_format()
is_compressed_row_format_supported()
is_file_per_table_enabled()
is_large_prefix_enabled()
get_row_format_sql()
get_name()
get_configuration_help()
diagnose()
raw_connect()
dispose()
get_db_handle()
set_db_handle()
can_use_readonly()
get_mysqli_server_info()
get_version_from_db()
should_db_version_be_read_from_db()
get_server_info()
allowed_param_types()
get_last_error()
get_tables()
get_indexes()
fetch_columns()
has_breaking_change_quoted_defaults()
has_breaking_change_sqlmode()
get_column_info()
mysqltype2moodletype()
normalise_value()
setup_is_unicodedb()
change_database_structure()
emulate_bound_params()
execute()
get_recordset_sql()
export_table_recordset()
create_recordset()
get_records_sql()
get_fieldset_sql()
insert_record_raw()
insert_record()
insert_chunk_size()
insert_records()
insert_chunk()
import_record()
update_record_raw()
update_record()
set_field_select()
delete_records_select()
delete_records_subquery()
sql_cast_char2int()
sql_cast_char2real()
sql_equal()
sql_like()
sql_concat()
sql_concat_join()
sql_group_concat()
sql_length()
sql_regex_supported()
sql_regex()
sql_regex_get_word_beginning_boundary_marker()
sql_regex_get_word_end_boundary_marker()
sql_cast_2signed()
sql_intersect()
replace_all_text_supported()
session_lock_supported()
get_session_lock()
release_session_lock()
transactions_supported()
begin_transaction()
commit_transaction()
rollback_transaction()
convert_table_row_format()
is_fulltext_search_supported()
fix_table_name()
Class: mysqli_native_moodle_database - X-Ref
Native mysqli class representing moodle database interface.| create_database($dbhost, $dbuser, $dbpass, $dbname, array $dboptions=null) X-Ref |
| Attempt to create the database param: string $dbhost param: string $dbuser param: string $dbpass param: string $dbname return: bool success |
| driver_installed() X-Ref |
| Detects if all needed PHP stuff installed. Note: can be used before connect() return: mixed true if ok, string if something |
| get_dbfamily() X-Ref |
| Returns database family type - describes SQL dialect Note: can be used before connect() return: string db family name (mysql, postgres, mssql, oracle, etc.) |
| get_dbtype() X-Ref |
| Returns more specific database driver type Note: can be used before connect() return: string db type mysqli, pgsql, oci, mssql, sqlsrv |
| get_dblibrary() X-Ref |
| Returns general database library name Note: can be used before connect() return: string db type pdo, native |
| get_dbengine() X-Ref |
| Returns the current MySQL db engine. This is an ugly workaround for MySQL default engine problems, Moodle is designed to work best on ACID compliant databases with full transaction support. Do not use MyISAM. return: string or null MySQL engine name |
| get_dbcollation() X-Ref |
| Returns the current MySQL db collation. This is an ugly workaround for MySQL default collation problems. return: string or null MySQL collation name |
| detect_collation() X-Ref |
| Set 'dbcollation' option return: string|null $dbcollation |
| is_antelope_file_format_no_more_supported() X-Ref |
| Tests if the Antelope file format is still supported or it has been removed. When removed, only Barracuda file format is supported, given the XtraDB/InnoDB engine. return: bool True if the Antelope file format has been removed; otherwise, false. |
| get_row_format($table = null) X-Ref |
| Get the row format from the database schema. param: string $table return: string row_format name or null if not known or table does not exist. |
| is_compressed_row_format_supported($cached = true) X-Ref |
| Is this database compatible with compressed row format? This feature is necessary for support of large number of text columns in InnoDB/XtraDB database. param: bool $cached use cached result return: bool true if table can be created or changed to compressed row format. |
| is_file_per_table_enabled() X-Ref |
| Check the database to see if innodb_file_per_table is on. return: bool True if on otherwise false. |
| is_large_prefix_enabled() X-Ref |
| Check the database to see if innodb_large_prefix is on. return: bool True if on otherwise false. |
| get_row_format_sql($engine = null, $collation = null) X-Ref |
| Determine if the row format should be set to compressed, dynamic, or default. Terrible kludge. If we're using utf8mb4 AND we're using InnoDB, we need to specify row format to be either dynamic or compressed (default is compact) in order to allow for bigger indexes (MySQL errors #1709 and #1071). param: string $engine The database engine being used. Will be looked up if not supplied. param: string $collation The database collation to use. Will look up the current collation if not supplied. return: string An sql fragment to add to sql statements. |
| get_name() X-Ref |
| Returns localised database type name Note: can be used before connect() return: string |
| get_configuration_help() X-Ref |
| Returns localised database configuration help. Note: can be used before connect() return: string |
| diagnose() X-Ref |
| Diagnose database and tables, this function is used to verify database and driver settings, db engine types, etc. return: string null means everything ok, string means problem found. |
| raw_connect(string $dbhost, string $dbuser, string $dbpass, string $dbname, $prefix, array $dboptions=null) X-Ref |
| Connect to db param: string $dbhost The database host. param: string $dbuser The database username. param: string $dbpass The database username's password. param: string $dbname The name of the database being connected to.e param: mixed $prefix string means moodle db prefix, false used for external databases where prefix not used param: array $dboptions driver specific options return: bool success |
| dispose() X-Ref |
| Close database connection and release all resources and memory (especially circular memory references). Do NOT use connect() again, create a new instance if needed. |
| get_db_handle() X-Ref |
| Gets db handle currently used with queries return: resource |
| set_db_handle($dbh) X-Ref |
| Sets db handle to be used with subsequent queries param: resource $dbh return: void |
| can_use_readonly(int $type, string $sql) X-Ref |
| Check if The query qualifies for readonly connection execution Logging queries are exempt, those are write operations that circumvent standard query_start/query_end paths. param: int $type type of query param: string $sql return: bool |
| get_mysqli_server_info() X-Ref |
| Returns the version of the MySQL server, as reported by the PHP client connection. Wrap $this->mysqli->server_info to improve testing strategy. return: string A string representing the version of the MySQL server that the MySQLi extension is connected to. |
| get_version_from_db() X-Ref |
| Returns the version of the MySQL server, as reported by 'SELECT VERSION()' query. return: string A string that indicates the MySQL server version. |
| should_db_version_be_read_from_db() X-Ref |
| Returns whether $CFG->dboptions['versionfromdb'] has been set to boolean `true`. return: bool True if $CFG->dboptions['versionfromdb'] has been set to boolean `true`. Otherwise, `false`. |
| get_server_info() X-Ref |
| Returns database server info array. return: array Array containing 'description' and 'version' info. |
| allowed_param_types() X-Ref |
| Returns supported query parameter types return: int bitmask of accepted SQL_PARAMS_* |
| get_last_error() X-Ref |
| Returns last error reported by database engine. return: string error message |
| get_tables($usecache=true) X-Ref |
| Return tables in database WITHOUT current prefix param: bool $usecache if true, returns list of cached tables. return: array of table names in lowercase and without prefix |
| get_indexes($table) X-Ref |
| Return table indexes - everything lowercased. param: string $table The table we want to get indexes from. return: array An associative array of indexes containing 'unique' flag and 'columns' being indexed |
| fetch_columns(string $table) X-Ref |
| Fetches detailed information about columns in table. param: string $table name return: database_column_info[] array of database_column_info objects indexed with column names |
| has_breaking_change_quoted_defaults() X-Ref |
| Indicates whether column information retrieved from `information_schema.columns` has default values quoted or not. return: boolean True when default values are quoted (breaking change); otherwise, false. |
| has_breaking_change_sqlmode() X-Ref |
| Indicates whether SQL_MODE default value has changed in a not backward compatible way. return: boolean True when SQL_MODE breaks BC; otherwise, false. |
| get_column_info(stdClass $rawcolumn) X-Ref |
| Returns moodle column info for raw column from information schema. param: stdClass $rawcolumn return: stdClass standardised colum info |
| mysqltype2moodletype($mysql_type) X-Ref |
| Normalise column type. param: string $mysql_type return: string one character |
| normalise_value($column, $value) X-Ref |
| Normalise values based in RDBMS dependencies (booleans, LOBs...) param: database_column_info $column column metadata corresponding with the value we are going to normalise param: mixed $value value we are going to normalise return: mixed the normalised value |
| setup_is_unicodedb() X-Ref |
| Is this database compatible with utf8? return: bool |
| change_database_structure($sql, $tablenames = null) X-Ref |
| Do NOT use in code, to be used by database_manager only! param: string|array $sql query param: array|null $tablenames an array of xmldb table names affected by this request. return: bool true |
| emulate_bound_params($sql, array $params=null) X-Ref |
| Very ugly hack which emulates bound parameters in queries because prepared statements do not use query cache. |
| execute($sql, array $params=null) X-Ref |
| Execute general sql query. Should be used only when no other method suitable. Do NOT use this to make changes in db structure, use database_manager methods instead! param: string $sql query param: array $params query parameters return: bool true |
| get_recordset_sql($sql, array $params=null, $limitfrom=0, $limitnum=0) X-Ref |
| Get a number of records as a moodle_recordset using a SQL statement. Since this method is a little less readable, use of it should be restricted to code where it's possible there might be large datasets being returned. For known small datasets use get_records_sql - it leads to simpler code. The return type is like: param: string $sql the SQL select query to execute. param: array $params array of sql parameters param: int $limitfrom return a subset of records, starting at this point (optional, required if $limitnum is set). param: int $limitnum return a subset comprising this many records (optional, required if $limitfrom is set). return: moodle_recordset instance |
| export_table_recordset($table) X-Ref |
| Get all records from a table. This method works around potential memory problems and may improve performance, this method may block access to table until the recordset is closed. param: string $table Name of database table. return: moodle_recordset A moodle_recordset instance {@link function get_recordset}. |
| create_recordset($result) X-Ref |
| No description |
| get_records_sql($sql, array $params=null, $limitfrom=0, $limitnum=0) X-Ref |
| Get a number of records as an array of objects using a SQL statement. Return value is like: param: string $sql the SQL select query to execute. The first column of this SELECT statement param: array $params array of sql parameters param: int $limitfrom return a subset of records, starting at this point (optional, required if $limitnum is set). param: int $limitnum return a subset comprising this many records (optional, required if $limitfrom is set). return: array of objects, or empty array if no records were found |
| get_fieldset_sql($sql, array $params=null) X-Ref |
| Selects records and return values (first field) as an array using a SQL statement. param: string $sql The SQL query param: array $params array of sql parameters return: array of values |
| insert_record_raw($table, $params, $returnid=true, $bulk=false, $customsequence=false) X-Ref |
| Insert new record into database, as fast as possible, no safety checks, lobs not supported. param: string $table name param: mixed $params data record as object or array param: bool $returnit return it of inserted record param: bool $bulk true means repeated inserts expected param: bool $customsequence true if 'id' included in $params, disables $returnid return: bool|int true or new id |
| insert_record($table, $dataobject, $returnid=true, $bulk=false) X-Ref |
| Insert a record into a table and return the "id" field if required. Some conversions and safety checks are carried out. Lobs are supported. If the return ID isn't required, then this just reports success as true/false. $data is an object containing needed data param: string $table The database table to be inserted into param: object|array $dataobject A data object with values for one or more fields in the record param: bool $returnid Should the id of the newly created record entry be returned? If this option is not requested then true/false is returned. return: bool|int true or new id |
| insert_chunk_size() X-Ref |
| Get chunk size for multiple records insert return: int |
| insert_records($table, $dataobjects) X-Ref |
| Insert multiple records into database as fast as possible. Order of inserts is maintained, but the operation is not atomic, use transactions if necessary. This method is intended for inserting of large number of small objects, do not use for huge objects with text or binary fields. param: string $table The database table to be inserted into param: array|Traversable $dataobjects list of objects to be inserted, must be compatible with foreach return: void does not return new record ids |
| insert_chunk($table, array $chunk, array $columns) X-Ref |
| Insert records in chunks. Note: can be used only from insert_records(). param: string $table param: array $chunk param: database_column_info[] $columns |
| import_record($table, $dataobject) X-Ref |
| Import a record into a table, id field is required. Safety checks are NOT carried out. Lobs are supported. param: string $table name of database table to be inserted into param: object $dataobject A data object with values for one or more fields in the record return: bool true |
| update_record_raw($table, $params, $bulk=false) X-Ref |
| Update record in database, as fast as possible, no safety checks, lobs not supported. param: string $table name param: stdClass|array $params data record as object or array param: bool true means repeated updates expected return: bool true |
| update_record($table, $dataobject, $bulk=false) X-Ref |
| Update a record in a table $dataobject is an object containing needed data Relies on $dataobject having a variable "id" to specify the record to update param: string $table The database table to be checked against. param: stdClass|array $dataobject An object with contents equal to fieldname=>fieldvalue. param: bool true means repeated updates expected return: bool true |
| set_field_select($table, $newfield, $newvalue, $select, array $params=null) X-Ref |
| Set a single field in every table record which match a particular WHERE clause. param: string $table The database table to be checked against. param: string $newfield the field to set. param: string $newvalue the value to set the field to. param: string $select A fragment of SQL to be used in a where clause in the SQL call. param: array $params array of sql parameters return: bool true |
| delete_records_select($table, $select, array $params=null) X-Ref |
| Delete one or more records from a table which match a particular WHERE clause. param: string $table The database table to be checked against. param: string $select A fragment of SQL to be used in a where clause in the SQL call (used to define the selection criteria). param: array $params array of sql parameters return: bool true |
| delete_records_subquery(string $table, string $field, string $alias, string $subquery, array $params = []) X-Ref |
| Deletes records using a subquery, which is done with a strange DELETE...JOIN syntax in MySQL because it performs very badly with normal subqueries. param: string $table Table to delete from param: string $field Field in table to match param: string $alias Name of single column in subquery e.g. 'id' param: string $subquery Query that will return values of the field to delete param: array $params Parameters for query |
| sql_cast_char2int($fieldname, $text=false) X-Ref |
| No description |
| sql_cast_char2real($fieldname, $text=false) X-Ref |
| No description |
| sql_equal($fieldname, $param, $casesensitive = true, $accentsensitive = true, $notequal = false) X-Ref |
| No description |
| sql_like($fieldname, $param, $casesensitive = true, $accentsensitive = true, $notlike = false, $escapechar = '\\') X-Ref |
| Returns 'LIKE' part of a query. Note that mysql does not support $casesensitive = true and $accentsensitive = false. More information in http://bugs.mysql.com/bug.php?id=19567. param: string $fieldname usually name of the table column param: string $param usually bound query parameter (?, :named) param: bool $casesensitive use case sensitive search param: bool $accensensitive use accent sensitive search (ignored if $casesensitive is true) param: bool $notlike true means "NOT LIKE" param: string $escapechar escape char for '%' and '_' return: string SQL code fragment |
| sql_concat() X-Ref |
| Returns the proper SQL to do CONCAT between the elements passed Can take many parameters param: string $str,... 1 or more fields/strings to concat return: string The concat sql |
| sql_concat_join($separator="' '", $elements=array() X-Ref |
| Returns the proper SQL to do CONCAT between the elements passed with a given separator param: string $separator The string to use as the separator param: array $elements An array of items to concatenate return: string The concat SQL |
| sql_group_concat(string $field, string $separator = ', ', string $sort = '') X-Ref |
| Return SQL for performing group concatenation on given field/expression param: string $field param: string $separator param: string $sort return: string |
| sql_length($fieldname) X-Ref |
| Returns the SQL text to be used to calculate the length in characters of one expression. param: string fieldname or expression to calculate its length in characters. return: string the piece of SQL code to be used in the statement. |
| sql_regex_supported() X-Ref |
| Does this driver support regex syntax when searching |
| sql_regex($positivematch = true, $casesensitive = false) X-Ref |
| Return regex positive or negative match sql param: bool $positivematch param: bool $casesensitive return: string or empty if not supported |
| sql_regex_get_word_beginning_boundary_marker() X-Ref |
| Returns the word-beginning boundary marker based on MySQL version. return: string The word-beginning boundary marker. |
| sql_regex_get_word_end_boundary_marker() X-Ref |
| Returns the word-end boundary marker based on MySQL version. return: string The word-end boundary marker. |
| sql_cast_2signed($fieldname) X-Ref |
| Returns the SQL to be used in order to an UNSIGNED INTEGER column to SIGNED. param: string $fieldname The name of the field to be cast return: string The piece of SQL code to be used in your statement. |
| sql_intersect($selects, $fields) X-Ref |
| Returns the SQL that allows to find intersection of two or more queries param: array $selects array of SQL select queries, each of them only returns fields with the names from $fields param: string $fields comma-separated list of fields return: string SQL query that will return only values that are present in each of selects |
| replace_all_text_supported() X-Ref |
| No description |
| session_lock_supported() X-Ref |
| No description |
| get_session_lock($rowid, $timeout) X-Ref |
| Obtain session lock param: int $rowid id of the row with session record param: int $timeout max allowed time to wait for the lock in seconds return: void |
| release_session_lock($rowid) X-Ref |
| No description |
| transactions_supported() X-Ref |
| Are transactions supported? It is not responsible to run productions servers on databases without transaction support ;-) MyISAM does not support support transactions. You can override this via the dbtransactions option. return: bool |
| begin_transaction() X-Ref |
| Driver specific start of real database transaction, this can not be used directly in code. return: void |
| commit_transaction() X-Ref |
| Driver specific commit of real database transaction, this can not be used directly in code. return: void |
| rollback_transaction() X-Ref |
| Driver specific abort of real database transaction, this can not be used directly in code. return: void |
| convert_table_row_format($tablename) X-Ref |
| Converts a table to either 'Compressed' or 'Dynamic' row format. param: string $tablename Name of the table to convert to the new row format. |
| is_fulltext_search_supported() X-Ref |
| Does this mysql instance support fulltext indexes? return: bool |
| fix_table_name($tablename) X-Ref |
| Fixes any table names that clash with reserved words. param: string $tablename The table name return: string The fixed table name |
