System Information Functions¶
Session Information Functions¶
current_catalog
Description: Name of the current database (called "catalog" in the SQL standard)
Return type: name
For example:
SELECT current_catalog; current_database ------------------ gaussdb (1 row)
current_database()
Description: Name of the current database
Return type: name
For example:
SELECT current_database(); current_database ------------------ gaussdb (1 row)
current_query()
Description: Text of the currently executing query, as submitted by the client (might contain more than one statement)
Return type: text
For example:
SELECT current_query(); current_query ------------------------- SELECT current_query(); (1 row)
current_schema[()]
Description: Name of current schema
Return type: name
For example:
SELECT current_schema(); current_schema ---------------- public (1 row)
Remarks: current_schema returns the first valid schema name in the search path. (If the search path is empty or contains no valid schema name, NULL is returned.) This is the schema that will be used for any tables or other named objects that are created without specifying a target schema.
current_schemas(boolean)
Description: Names of schemas in search path
Return type: name[]
For example:
SELECT current_schemas(true); current_schemas --------------------- {pg_catalog,public} (1 row)
Note:
current_schemas(boolean) returns an array of the names of all schemas presently in the search path. The Boolean option determines whether implicitly included system schemas such as pg_catalog are included in the returned search path.
Note
The search path can be altered at run time. The command is:
SET search_path TO schema [, schema, ...]
current_user
Description: User name of current execution context
Return type: name
For example:
SELECT current_user; current_user -------------- dbadmin (1 row)
Note: current_user is the user identifier that is applicable for permission checking. Normally it is equal to the session user, but it can be changed with SET ROLE. It also changes during the execution of functions with the attribute SECURITY DEFINER.
inet_client_addr()
Description: Displays the IP address of the currently connected client.
Note
It is available only in remote connection mode.
If the database is connected to the local PC, the value is empty.
Return type: inet
For example:
SELECT inet_client_addr(); inet_client_addr ------------------ 10.10.0.50 (1 row)
inet_client_port()
Description: Displays the port number of the currently connected client.
Note
It is available only in remote connection mode.
Return type: int
For example:
SELECT inet_client_port(); inet_client_port ------------------ 33143 (1 row)
inet_server_addr()
Description: Displays the IP address of the current server.
Note
It is available only in remote connection mode.
If the database is connected to the local PC, the value is empty.
Return type: inet
For example:
SELECT inet_server_addr(); inet_server_addr ------------------ 10.10.0.13 (1 row)
inet_server_port()
Description: Displays the port of the current server. All these functions return NULL if the current connection is via a Unix-domain socket.
Note
It is available only in remote connection mode.
Return type: int
For example:
SELECT inet_server_port(); inet_server_port ------------------ 8000 (1 row)
pg_backend_pid()
Description: Process ID of the server process attached to the current session
Return type: int
For example:
SELECT pg_backend_pid(); pg_backend_pid ----------------- 140229352617744 (1 row)
pg_conf_load_time()
Description: Configures load time. pg_conf_load_time returns the timestamp with time zone when the server configuration files were last loaded.
Return type: timestamp with time zone
For example:
SELECT pg_conf_load_time(); pg_conf_load_time ------------------------------ 2017-09-01 16:05:23.89868+08 (1 row)
pg_my_temp_schema()
Description: OID of the temporary schema of a session. The value is 0 if the OID does not exist.
Return type: OID
For example:
SELECT pg_my_temp_schema(); pg_my_temp_schema ------------------- 0 (1 row)
Note: pg_my_temp_schema returns the OID of the current session's temporary schema, or zero if it has none (because it has not created any temporary tables). pg_is_other_temp_schema returns true if the given OID is the OID of another session's temporary schema.
pg_is_other_temp_schema(oid)
Description: Whether the schema is the temporary schema of another session.
Return type: boolean
For example:
SELECT pg_is_other_temp_schema(25356); pg_is_other_temp_schema ------------------------- f (1 row)
pg_listening_channels()
Description: Channel names that the session is currently listening on
Return type: setof text
For example:
SELECT pg_listening_channels(); pg_listening_channels ----------------------- (0 rows)
Note: pg_listening_channels returns a set of names of channels that the current session is listening to.
pg_postmaster_start_time()
Description: Server start time pg_postmaster_start_time returns the timestamp with time zone when the server started.
Return type: timestamp with time zone
For example:
SELECT pg_postmaster_start_time(); pg_postmaster_start_time ------------------------------ 2017-08-30 16:02:54.99854+08 (1 row)
pg_trigger_depth()
Description: Current nesting level of triggers
Return type: int
For example:
SELECT pg_trigger_depth(); pg_trigger_depth ------------------ 0 (1 row)
pgxc_version()
Description: Postgres-XC version information
Return type: text
For example:
SELECT pgxc_version(); pgxc_version ------------------------------------------------------------------------------------------------------------- Postgres-XC 1.1 on x86_64-unknown-linux-gnu, based on PostgreSQL 9.2.4, compiled by g++ (GCC) 5.4.0, 64-bit (1 row)
session_user
Description: Session user name
Return type: name
For example:
SELECT session_user; session_user -------------- dbadmin (1 row)
Note: session_user is usually the user who initiated the current database connection, but administrators can change this setting with SET SESSION AUTHORIZATION.
user
Description: Is equivalent to current_user.
Return type: name
For example:
SELECT user; current_user -------------- dbadmin (1 row)
version()
Description: version information. version returns a string describing a server's version.
Return type: text
For example:
SELECT version(); version --------------------------------------------------------------------------------------------------------------------------------------- PostgreSQL 9.2.4 gsql ((GaussDB 8.1.1 build af002019) compiled at 2020-01-10 05:43:20 commit 6995 last mr 11566 ) on x86_64-unknown-linux-gnu, compiled by g++ (GCC) 5.4.0, 64-bit (1 row)
Access Privilege Inquiry Functions¶
has_any_column_privilege(user, table, privilege)
Description: Queries whether a specified user has permission for any column of table.
Return type: boolean
has_any_column_privilege(table, privilege)
Description: Queries whether the current user has permission for any column of table.
Return type: boolean
has_any_column_privilege checks whether a user can access any column of a table in a particular way. Its parameter possibilities are analogous to has_table_privilege, except that the desired access permission type must be some combination of SELECT, INSERT, UPDATE, or REFERENCES.
Note
Note that having any of these permissions at the table level implicitly grants it for each column of the table, so has_any_column_privilege will always return true if has_table_privilege does for the same parameters. But has_any_column_privilege also succeeds if there is a column-level grant of the permission for at least one column.
has_column_privilege(user, table, column, privilege)
Description: Queries whether a specified user has permission for column.
Return type: boolean
has_column_privilege(table, column, privilege)
Description: Queries whether the current user has permission for column.
Return type: boolean
has_column_privilege checks whether a user can access a column in a particular way. Its argument possibilities are analogous to has_table_privilege, with the addition that the column can be specified either by name or attribute number. The desired access permission type must evaluate to some combination of SELECT, INSERT, UPDATE, or REFERENCES.
Note
Note that having any of these permissions at the table level implicitly grants it for each column of the table.
has_database_privilege(user, database, privilege)
Description: Queries whether a specified user has permission for database.
Return type: boolean
has_database_privilege(database, privilege)
Description: Queries whether the current user has permission for database.
Return type: boolean
Note: has_database_privilege checks whether a user can access a database in a particular way. Its argument possibilities are analogous to has_table_privilege. The desired access permission type must evaluate to some combination of CREATE, CONNECT, TEMPORARY, or TEMP (which is equivalent to TEMPORARY).
has_foreign_data_wrapper_privilege(user, fdw, privilege)
Description: Queries whether a specified user has permission for foreign-data wrapper.
The fdw parameter indicates the name or ID of the foreign data wrapper.
Return type: boolean
has_foreign_data_wrapper_privilege(fdw, privilege)
Description: Queries whether the current user has permission for foreign-data wrapper.
Return type: boolean
Note: has_foreign_data_wrapper_privilege checks whether a user can access a foreign-data wrapper in a particular way. Its argument possibilities are analogous to has_table_privilege. The desired access permission type must evaluate to USAGE.
has_function_privilege(user, function, privilege)
Description: Queries whether a specified user has permission for function.
Return type: boolean
has_function_privilege(function, privilege)
Description: Queries whether the current user has permission for function.
Return type: boolean
Note: has_function_privilege checks whether a user can access a function in a particular way. Its argument possibilities are analogous to has_table_privilege. When a function is specified by a text string rather than by OID, the allowed input is the same as that for the regprocedure data type (see Object Identifier Types). The desired access permission type must evaluate to EXECUTE.
has_language_privilege(user, language, privilege)
Description: Queries whether a specified user has permission for language.
Return type: boolean
has_language_privilege(language, privilege)
Description: Queries whether the current user has permission for language.
Return type: boolean
Note: has_language_privilege checks whether a user can access a procedural language in a particular way. Its argument possibilities are analogous to has_table_privilege. The desired access permission type must evaluate to USAGE.
has_schema_privilege(user, schema, privilege)
Description: Queries whether a specified user has permission for schema.
Return type: boolean
has_schema_privilege(schema, privilege)
Description: Queries whether the current user has permission for schema.
Return type: boolean
Note: has_schema_privilege checks whether a user can access a schema in a particular way. Its argument possibilities are analogous to has_table_privilege. The desired access permission type must evaluate to some combination of CREATE or USAGE.
has_server_privilege(user, server, privilege)
Description: Queries whether a specified user has permission for foreign server.
Return type: boolean
has_server_privilege(server, privilege)
Description: Queries whether the current user has permission for foreign server.
Return type: boolean
Note: has_server_privilege checks whether a user can access a foreign server in a particular way. Its argument possibilities are analogous to has_table_privilege. The desired access permission type must evaluate to USAGE.
has_table_privilege(user, table, privilege)
Description: Queries whether a specified user has permission for table.
Return type: boolean
has_table_privilege(table, privilege)
Description: Queries whether the current user has permission for table.
Return type: boolean
has_table_privilege checks whether a user can access a table in a particular way. The user can be specified by name, by OID (pg_authid.oid), public to indicate the PUBLIC pseudo-role, or if the argument is omitted current_user is assumed. The table can be specified by name or by OID. When specifying by name, the name can be schema-qualified if necessary. The desired access permission type is specified by a text string, which must be one of the values SELECT, INSERT, UPDATE, DELETE, TRUNCATE, REFERENCES, or TRIGGER. Optionally, WITH GRANT OPTION can be added to a permission type to test whether the permission is held with grant option. Also, multiple permission types can be listed separated by commas, in which case the result will be true if any of the listed permissions is held.
For example:
SELECT has_table_privilege('tpcds.web_site', 'select'); has_table_privilege --------------------- t (1 row) SELECT has_table_privilege('dbadmin', 'tpcds.web_site', 'select,INSERT WITH GRANT OPTION '); has_table_privilege --------------------- t (1 row)
pg_has_role(user, role, privilege)
Description: Queries whether a specified user has permission for role.
Return type: boolean
pg_has_role(role, privilege)
Description: Specifies whether the current user has permission for role.
Return type: boolean
Note: pg_has_role checks whether a user can access a role in a particular way. Its argument possibilities are analogous to has_table_privilege, except that public is not allowed as a user name. The desired access permission type must evaluate to some combination of MEMBER or USAGE. MEMBER denotes direct or indirect membership in the role (that is, the right to do SET ROLE), while USAGE denotes the permissions of the role are available without doing SET ROLE.
Schema Visibility Inquiry Functions¶
Each function performs the visibility check for one type of database object. For functions and operators, an object in the search path is visible if there is no object of the same name and argument data type(s) earlier in the path. For operator classes, both name and associated index access method are considered.
All these functions require OIDs to identify the objects to be checked. If you want to test an object by name, it is convenient to use the OID alias types (regclass, regtype, regprocedure, regoperator, regconfig, or regdictionary).
For example, a table is said to be visible if its containing schema is in the search path and no table of the same name appears earlier in the search path. This is equivalent to the statement that the table can be referenced by name without explicit schema qualification. For example, to list the names of all visible tables:
SELECT relname FROM pg_class WHERE pg_table_is_visible(oid);
pg_collation_is_visible(collation_oid)
Description: Queries whether the collation is visible in search path.
Return type: boolean
pg_conversion_is_visible(conversion_oid)
Description: Queries whether the conversion is visible in search path.
Return type: boolean
pg_function_is_visible(function_oid)
Description: Queries whether the function is visible in search path.
Return type: boolean
pg_opclass_is_visible(opclass_oid)
Description: Queries whether the operator class is visible in search path.
Return type: boolean
pg_operator_is_visible(operator_oid)
Description: Queries whether the operator is visible in search path.
Return type: boolean
pg_opfamily_is_visible(opclass_oid)
Description: Queries whether the operator family is visible in search path.
Return type: boolean
pg_table_is_visible(table_oid)
Description: Queries whether the table is visible in search path.
Return type: boolean
pg_ts_config_is_visible(config_oid)
Description: Queries whether the text search configuration is visible in search path.
Return type: boolean
pg_ts_dict_is_visible(dict_oid)
Description: Queries whether the text search dictionary is visible in search path.
Return type: boolean
pg_ts_parser_is_visible(parser_oid)
Description: Queries whether the text search parser is visible in search path.
Return type: boolean
pg_ts_template_is_visible(template_oid)
Description: Queries whether the text search template is visible in search path.
Return type: boolean
pg_type_is_visible(type_oid)
Description: Queries whether the type (or domain) is visible in search path.
Return type: boolean
System Catalog Information Functions¶
format_type(type_oid, typemod)
Description: Gets SQL name of a data type.
Return type: text
Note:
format_type returns the SQL name of a data type that is identified by its type OID and possibly a type modifier. Pass NULL for the type modifier if no specific modifier is known. Certain type modifiers are passed for data types with length limitations. The SQL name returned from format_type contains the length of the data type, which can be calculated by taking sizeof(int32) from actual storage length [actual storage len - sizeof(int32)] in the unit of bytes. 32-bit space is required to store the customized length set by users. So the actual storage length contains 4 bytes more than the customized length. In the following example, the SQL name returned from format_type is character varying(6), indicating the length of varchar type is 6 bytes. So the actual storage length of varchar type is 10 bytes.
SELECT format_type((SELECT oid FROM pg_type WHERE typname='varchar'), 10); format_type ---------------------- character varying(6) (1 row)
pg_check_authid(role_oid)
Description: Checks whether a role name with given OID exists.
Return type: bool
pg_describe_object(catalog_id, object_id, object_sub_id)
Description: Gets description of a database object.
Return type: text
Note: pg_describe_object returns a description of a database object specified by catalog OID, object OID and a (possibly zero) sub-object ID. This is useful to determine the identity of an object as stored in the pg_depend catalog.
pg_get_constraintdef(constraint_oid)
Description: Gets definition of a constraint.
Return type: text
pg_get_constraintdef(constraint_oid, pretty_bool)
Description: Gets definition of a constraint.
Return type: text
Note: pg_get_constraintdef and pg_get_indexdef respectively reconstruct the creating command for a constraint and an index.
pg_get_expr(pg_node_tree, relation_oid)
Description: Decompiles internal form of an expression, assuming that any Vars in it refer to the relationship indicated by the second parameter.
Return type: text
pg_get_expr(pg_node_tree, relation_oid, pretty_bool)
Description: Decompiles internal form of an expression, assuming that any Vars in it refer to the relationship indicated by the second parameter.
Return type: text
Note: pg_get_expr decompiles the internal form of an individual expression, such as the default value for a column. It can be useful when examining the contents of system catalogs. If the expression might contain Vars, specify the OID of the relationship they refer to as the second parameter; if no Vars are expected, zero is sufficient.
pg_get_functiondef(func_oid)
Description: Gets definition of a function.
Return type: text
func_oid is the OID of the function, which can be queried in the PG_PROC system catalog.
Example: Query the OID and definition of the justify_days function.
SELECT oid FROM pg_proc WHERE proname ='justify_days'; oid ------ 1295 (1 row) SELECT * FROM pg_get_functiondef(1295); headerlines | definition -------------+-------------------------------------------------------------- 4 | CREATE OR REPLACE FUNCTION pg_catalog.justify_days(interval)+ | RETURNS interval + | LANGUAGE internal + | IMMUTABLE STRICT NOT FENCED NOT SHIPPABLE + | AS $function$interval_justify_days$function$ + | (1 row)pg_get_function_arguments(func_oid)
Description: Gets argument list of function's definition (with default values).
Return type: text
Note: pg_get_function_arguments returns the argument list of a function, in the form it would need to appear in within CREATE FUNCTION.
pg_get_function_identity_arguments(func_oid)
Description: Gets argument list to identify a function (without default values).
Return type: text
Note: pg_get_function_identity_arguments returns the argument list necessary to identify a function, in the form it would need to appear in within ALTER FUNCTION. This form omits default values.
pg_get_function_result(func_oid)
Description: Gets RETURNS clause for function.
Return type: text
Note: pg_get_function_result returns the appropriate RETURNS clause for the function.
pg_get_indexdef(index_oid)
Description: Gets CREATE INDEX command for index.
Return type: text
index_oid indicates the index OID, which can be queried in the PG_STATIO_ALL_INDEXES system view.
Example: Query the OID and CREATE INDEX command of index ds_ship_mode_t1_index1.
SELECT indexrelid FROM PG_STATIO_ALL_INDEXES WHERE indexrelname = 'ds_ship_mode_t1_index1'; indexrelid ------------ 136035 (1 row) SELECT * FROM pg_get_indexdef(136035); pg_get_indexdef --------------------------------------------------------------------------------------------------------------- CREATE INDEX ds_ship_mode_t1_index1 ON tpcds.ship_mode_t1 USING psort (sm_ship_mode_sk) TABLESPACE pg_default (1 row)
pg_get_indexdef(index_oid, column_no, pretty_bool)
Description: Gets CREATE INDEX command for index, or definition of just one index column when column_no is not zero.
Return type: text
SELECT * FROM pg_get_indexdef(136035,0,false); pg_get_indexdef --------------------------------------------------------------------------------------------------------------- CREATE INDEX ds_ship_mode_t1_index1 ON tpcds.ship_mode_t1 USING psort (sm_ship_mode_sk) TABLESPACE pg_default (1 row) SELECT * FROM pg_get_indexdef(136035,1,false); pg_get_indexdef ----------------- sm_ship_mode_sk (1 row)
pg_get_keywords()
Description: Gets list of SQL keywords and their categories.
Return type: setof record
Note: pg_get_keywords returns a set of records describing the SQL keywords recognized by the server. The word column contains the keyword. The catcode column contains a category code: U for unreserved, C for column name, T for type or function name, or R for reserved. The catdesc column contains a possibly-localized string describing the category.
pg_get_ruledef(rule_oid)
Description: Gets CREATE RULE command for a rule.
Return type: text
pg_get_ruledef(rule_oid, pretty_bool)
Description: Gets CREATE RULE command for a rule.
Return type: text
pg_get_userbyid(role_oid)
Description: Gets role name with given OID.
Return type: name
Note: pg_get_userbyid extracts a role's name given its OID.
pg_get_viewdef(viewname text [, pretty bool [, fullflag bool]])
Description: gets underlying SELECT command for views.
Return type: text
Note:
pg_get_viewdef reconstructs the SELECT query that defines a view. If the value of pretty bool is set to true, the display format is suitable for printing and more readable. The default value of pretty bool is false, and the display format is not readable. Use the default format for dump purposes whenever possible. The pretty bool parameter can be applied only to valid views.
When fullflag bool is set to true, the complete definition of the view is displayed. The default value is false.
pg_get_viewdef(viewoid oid [, pretty bool [, fullflag bool]])
Description: gets underlying SELECT command for views.
Return type: text
pg_get_viewdef(view_oid, wrap_column_int)
Description: Gets underlying SELECT command for view, wrapping lines with columns as specified, printing is implied.
Return type: text
pg_get_tabledef(table_oid)
Description: Obtains a table definition based on table_oid.
Return type: text
Example: Obtain the OID of the table customer_t2 from the system catalog pg_class, and then use this function to query the definition of customer_t2 to obtain the table columns, storage mode (row-store or column-store), and table distribution mode configured for customer_t2 when it is created.
select oid from pg_class where relname ='customer_t2'; oid ------- 17353 (1 row) select * from pg_get_tabledef(17353); pg_get_tabledef -------------------------------------------- SET search_path = dbadmin; + CREATE TABLE customer_t2 ( + state_id character(2), + state_name character varying(40), + area_id numeric + ) + WITH (orientation=column, compression=low)+ DISTRIBUTE BY HASH(state_id) + TO GROUP group_version1; (1 row)
pg_get_tabledef(table_name)
Description: Obtains a table definition based on table_name.
Return type: text
Remarks: pg_get_tabledef reconstructs the CREATE statement of the table definition, including the table definition, index information, and comments. Users need to create the dependent objects of the table, such as groups, schemas, tablespaces, and servers. The table definition does not include the statements for creating these dependent objects.
pg_options_to_table(reloptions)
Description: Gets the set of storage option name/value pairs.
Return type: setof record
Note: pg_options_to_table returns the set of storage option name/value pairs (option_name/option_value) when passing pg_class.reloptions or pg_attribute.attoptions.
pg_typeof(any)
Description: Gets the data type of any value.
Return type: regtype
Note:
pg_typeof returns the OID of the data type of the value that is passed to it. This can be helpful for troubleshooting or dynamically constructing SQL queries. The function is declared as returning regtype, which is an OID alias type (see Object Identifier Types). This means that it is the same as an OID for comparison purposes but displays as a type name.
For example:
SELECT pg_typeof(33); pg_typeof ----------- integer (1 row) SELECT typlen FROM pg_type WHERE oid = pg_typeof(33); typlen -------- 4 (1 row)
collation for (any)
Description: Gets the collation of the parameter.
Return type: text
Note:
The expression collation for returns the collation of the value that is passed to it. For example:
SELECT collation for (description) FROM pg_description LIMIT 1; pg_collation_for ------------------ "default" (1 row)
The value might be quoted and schema-qualified. If no collation is derived for the argument expression, then a null value is returned. If the parameter is not of a collectable data type, then an error is thrown.
getdistributekey(table_name)
Description: Gets a distribution column for a hash table.
Return type: text
For example:
SELECT getdistributekey('item'); getdistributekey ------------------ i_item_sk (1 row)
Transaction IDs and Snapshots¶
The following functions provide server transaction information in an exportable form. The main use of these functions is to determine which transactions were committed between two snapshots.
pgxc_is_committed(transaction_id)
Description: Determines whether the given XID is committed or ignored. NULL indicates the unknown status (such as running, preparing, and freezing).
Return type: bool
txid_current()
Description: Gets current transaction ID.
Return type: bigint
txid_current_snapshot()
Description: Gets current snapshot.
Return type: txid_snapshot
txid_snapshot_xip(txid_snapshot)
Description: Gets in-progress transaction IDs in snapshot.
Return type: setof bigint
txid_snapshot_xmax(txid_snapshot)
Description: Gets xmax of snapshot.
Return type: bigint
txid_snapshot_xmin(txid_snapshot)
Description: Gets xmin of snapshot.
Return type: bigint
txid_visible_in_snapshot(bigint, txid_snapshot)
Description: Queries whether the transaction ID is visible in snapshot. (do not use with subtransaction ids)
Return type: boolean
The internal transaction ID type (xid) is 32 bits wide and wraps around every 4 billion transactions. txid_snapshot, the data type used by these functions, stores information about transaction ID visibility at a particular moment in time. Table 1 describes its components.
Name | Description |
|---|---|
xmin | Earliest transaction ID (txid) that is still active. All earlier transactions will either be committed and visible, or rolled back. |
xmax | First as-yet-unassigned txid. All txids greater than or equal to this are not yet started as of the time of the snapshot, so they are invisible. |
xip_list | Active txids at the time of the snapshot. The list includes only those active txids between xmin and xmax; there might be active txids higher than xmax. A txid that is xmin <= txid < xmax and not in this list was already completed at the time of the snapshot, and is either visible or dead according to its commit status. The list does not include txids of subtransactions. |
txid_snapshot's textual representation is xmin:xmax:xip_list.
For example: 10:20:10,14,15 means xmin=10, xmax=20, xip_list=10, 14, 15.
Computing Node Group Function¶
pv_compute_pool_workload()
Description: Load status of a computing Node Group.
Return type: void
For example:
SELECT * from pv_compute_pool_workload();
nodename | rpinuse | maxrp | nodestate
-----------+---------+-------+-----------
datanode1 | 0 | 1000 | normal
datanode2 | 0 | 1000 | normal
(2 rows)
Lock Information Function¶
pgxc_get_lock_conflicts()
Description: Obtains information about conflicting locks in the cluster. When a lock is waiting for another lock or another lock is waiting for it, a lock conflict occurs.
Return type: setof record
Comment Information Functions¶
col_description(table_oid, column_number)
Description: Gets comment for a table column.
Return type: text
Note: col_description returns the comment for a table column, which is specified by the OID of its table and its column number.
obj_description(object_oid, catalog_name)
Description: Gets comment for a database object.
Return type: text
Note: The two-parameter form of obj_description returns the comment for a database object specified by its OID and the name of the containing system catalog. For example, obj_description(123456,'pg_class') would retrieve the comment for the table with OID 123456. The one-parameter form of obj_description requires only the object OID.
obj_description cannot be used for table columns since columns do not have OIDs of their own.
obj_description(object_oid)
Description: Gets comment for a database object.
Return type: text
shobj_description(object_oid, catalog_name)
Description: Gets comment for a shared database object.
Return type: text
Note: shobj_description is used just like obj_description except the former is used for retrieving comments on shared objects. Some system catalogs are global to all databases within each cluster, and the comments for objects in them are stored globally as well.