Installation guide

5. mxODBC Overview

Introspection via cursor.execute()

In mxODBC this can be done right after executing a SQL statement using one of

the

cursor.execute*() methods by looking at the cursor.description

attribute.

Introspection via cursor.prepare()

Alternatively, you can use the cursor.prepare() method to just prepare

execution of a SQL statement - without actually executing it. This may be desirable

in case the result set is not immediately needed or the query would require a long

time to execute.

The cursor.description attribute

This cursor attribute provides access to a sequence of tuples, each describing the

result set column at that position:

(name, type_code, display_size,

internal_size, precision, scale, null_ok).

If no information is available the

cursor.description is set to None.

The column tuple entries have the following meanings (index given in square

brackets):

name [0]

Name of the column as returned by the database.

type_code [1]

In mxODBC, this is the SQL type integer describing the database data type of

the result set column. These are described in the section 8 Supported Data

Types and are available through the SQL singleton defined at subpackage

module level for comparisons.

display_size [2]

mxODBC will always return None for this field. For database tables, this

information can be determined by using the

cursor.getcolattribute(position, SQL.DESC_DISPLAY_SIZE) method,

if needed.

internal_size [3]

mxODBC will always return None for this field. For database tables, this

information can be determined by using the

cursor.getcolattribute(position, SQL.DESC_OCTET_LENGTH) method,

if needed.

precision [4]

Precision of numeric columns.

scale [5]

Scale of numeric columns.