MDAC 2.5 SDK - ODBC Programmer's Reference
Chapter 21: ODBC API Reference

SQLSpecialColumns

Conformance

Version Introduced: ODBC 1.0
Standards Compliance: X/Open

Summary

SQLSpecialColumns retrieves the following information about columns within a specified table:

• The optimal set of columns that uniquely identifies a row in the table.
  
  
• Columns that are automatically updated when any value in the row is updated by a transaction.

Syntax

SQLRETURN SQLSpecialColumns(
     SQLHSTMT     StatementHandle,
     SQLSMALLINT     IdentifierType,
     SQLCHAR *     CatalogName,
     SQLSMALLINT     NameLength1,
     SQLCHAR *     SchemaName,
     SQLSMALLINT     NameLength2,
     SQLCHAR *     TableName,
     SQLSMALLINT     NameLength3,
     SQLSMALLINT     Scope,
     SQLSMALLINT     Nullable);

Arguments

StatementHandle

[Input]
Statement handle.

IdentifierType

[Input]
Type of column to return. Must be one of the following values:

SQL_BEST_ROWID: Returns the optimal column or set of columns that, by retrieving values from the column or columns, allows any row in the specified table to be uniquely identified. A column can be either a pseudo-column specifically designed for this purpose (as in Oracle ROWID or Ingres TID) or the column or columns of any unique index for the table.

SQL_ROWVER: Returns the column or columns in the specified table, if any, that are automatically updated by the data source when any value in the row is updated by any transaction (as in SQLBase ROWID or Sybase TIMESTAMP).

CatalogName

[Input]
Catalog name for the table. If a driver supports catalogs for some tables but not for others, such as when the driver retrieves data from different DBMSs, an empty string ("") denotes those tables that do not have catalogs. CatalogName cannot contain a string search pattern.

If the SQL_ATTR_METADATA_ID statement attribute is set to SQL_TRUE, CatalogName is treated as an identifier and its case is not significant. If it is SQL_FALSE, CatalogName is an ordinary argument; it is treated literally, and its case is significant. For more information, see "Arguments in Catalog Functions" in Chapter 7, "Catalog Functions."

NameLength1

[Input]
Length of *CatalogName.

SchemaName

[Input]
Schema name for the table. If a driver supports schemas for some tables but not for others, such as when the driver retrieves data from different DBMSs, an empty string ("") denotes those tables that do not have schemas. SchemaName cannot contain a string search pattern.

If the SQL_ATTR_METADATA_ID statement attribute is set to SQL_TRUE, SchemaName is treated as an identifier and its case is not significant. If it is SQL_FALSE, SchemaName is an ordinary argument; it is treated literally, and its case is significant.

NameLength2

[Input]
Length of *SchemaName.

TableName

[Input]
Table name. This argument cannot be a null pointer. TableName cannot contain a string search pattern.

If the SQL_ATTR_METADATA_ID statement attribute is set to SQL_TRUE, TableName is treated as an identifier and its case is not significant. If it is SQL_FALSE, TableName is an ordinary argument; it is treated literally, and its case is significant.

NameLength3

[Input]
Length of *TableName.

Scope

[Input]
Minimum required scope of the rowid. The returned rowid may be of greater scope. Must be one of the following:

SQL_SCOPE_CURROW: The rowid is guaranteed to be valid only while positioned on that row. A later reselect using rowid may not return a row if the row was updated or deleted by another transaction.

SQL_SCOPE_TRANSACTION: The rowid is guaranteed to be valid for the duration of the current transaction.

SQL_SCOPE_SESSION: The rowid is guaranteed to be valid for the duration of the session (across transaction boundaries).

Nullable

[Input]
Determines whether to return special columns that can have a NULL value. Must be one of the following:

SQL_NO_NULLS: Exclude special columns that can have NULL values. Some drivers cannot support SQL_NO_NULLS, and these drivers will return an empty result set if SQL_NO_NULLS was specified. Applications should be prepared for this case and request SQL_NO_NULLS only if it is absolutely required.

SQL_NULLABLE: Return special columns even if they can have NULL values.

Returns

SQL_SUCCESS, SQL_SUCCESS_WITH_INFO, SQL_STILL_EXECUTING, SQL_ERROR, or SQL_INVALID_HANDLE.

Diagnostics

When SQLSpecialColumns returns SQL_ERROR or SQL_SUCCESS_WITH_INFO, an associated SQLSTATE value may be obtained by calling SQLGetDiagRec with a HandleType of SQL_HANDLE_STMT and a Handle of StatementHandle. The following table lists the SQLSTATE values commonly returned by SQLSpecialColumns and explains each one in the context of this function; the notation "(DM)" precedes the descriptions of SQLSTATEs returned by the Driver Manager. The return code associated with each SQLSTATE value is SQL_ERROR, unless noted otherwise.

Comments

When the IdentifierType argument is SQL_BEST_ROWID, SQLSpecialColumns returns the column or columns that uniquely identify each row in the table. These columns can always be used in a select-list or WHERE clause. SQLColumns, which is used to return a variety of information on the columns of a table, does not necessarily return the columns that uniquely identify each row, or columns that are automatically updated when any value in the row is updated by a transaction. For example, SQLColumns might not return the Oracle pseudo-column ROWID. This is why SQLSpecialColumns is used to return these columns. For more information, see "Uses of Catalog Data" in Chapter 7, "Catalog Functions."

Note   For more information about the general use, arguments, and returned data of ODBC catalog functions, see Chapter 7, "Catalog Functions."

If there are no columns that uniquely identify each row in the table, SQLSpecialColumns returns a rowset with no rows; a subsequent call to SQLFetch or SQLFetchScroll on the statement returns SQL_NO_DATA.

If the IdentifierType, Scope, or Nullable arguments specify characteristics that are not supported by the data source, SQLSpecialColumns returns an empty result set.

If the SQL_ATTR_METADATA_ID statement attribute is set to SQL_TRUE, the CatalogName, SchemaName, and TableName arguments are treated as identifiers, so they cannot be set to a null pointer in certain situations. (For more information, see "Arguments in Catalog Functions" in Chapter 7, "Catalog Functions.")

SQLSpecialColumns returns the results as a standard result set, ordered by SCOPE.

The following columns have been renamed for ODBC 3.x. The column name changes do not affect backward compatibility because applications bind by column number.

To determine the actual length of the COLUMN_NAME column, an application can call SQLGetInfo with the SQL_MAX_COLUMN_NAME_LEN option.

The following table lists the columns in the result set. Additional columns beyond column 8 (PSEUDO_COLUMN) can be defined by the driver. An application should gain access to driver-specific columns by counting down from the end of the result set rather than specifying an explicit ordinal position. For more information, see "Data Returned by Catalog Functions" in Chapter 7, "Catalog Functions."

After the application retrieves values for SQL_BEST_ROWID, the application can use these values to reselect that row within the defined scope. The SELECT statement is guaranteed to return either no rows or one row.

If an application reselects a row based on the rowid column or columns and the row is not found, the application can assume that the row was deleted or the rowid columns were modified. The opposite is not true: even if the rowid has not changed, the other columns in the row may have changed.

Columns returned for column type SQL_BEST_ROWID are useful for applications that need to scroll forward and back within a result set to retrieve the most recent data from a set of rows. The column or columns of the rowid are guaranteed not to change while positioned on that row.

The column or columns of the rowid may remain valid even when the cursor is not positioned on the row; the application can determine this by checking the SCOPE column in the result set.

Columns returned for column type SQL_ROWVER are useful for applications that need the ability to check whether any columns in a given row have been updated while the row was reselected using the rowid. For example, after reselecting a row using rowid, the application can compare the previous values in the SQL_ROWVER columns to the ones just fetched. If the value in a SQL_ROWVER column differs from the previous value, the application can alert the user that data on the display has changed.

Code Example

For a code example of a similar function, see SQLColumns.

Related Functions