This function has been deprecated in favor of dbListFields().
For a given table this function returns detailed information on
all fields / columns. The expectation is that this is a relatively thin
wrapper around the ODBC SQLColumns function call, with some of the field
names renamed / re-ordered according to the return specifications below.
odbcConnectionColumns(conn, name, ..., exact = FALSE)data.frame with columns
name
field.type - equivalent to type_name in SQLColumns output
table_name
schema_name
catalog_name
data_type
column_size
buffer_length
decimal_digits
numeric_precision_radix
remarks
column_default
sql_data_type
sql_datetime_subtype
char_octet_length
ordinal_position
nullable
OdbcConnection
Table identifiers.
By default, are interpreted as a ODBC search pattern where _ and % are
wild cards. Set exact = TRUE to match _ exactly.
additional parameters to methods
Set to TRUE to escape _ in identifier names so that it
matches exactly, rather than matching any single character. % always
matches any number of characters as this is unlikely to appear in a
table name.
In dbWriteTable() we make a call to this method
to get details on the fields of the table we are writing to. In particular
the columns data_type, column_size, and decimal_digits are used. An
implementation is not necessary for dbWriteTable() to work.
odbcConnectionColumns is routed through the SQLColumns ODBC
method. This function, together with remaining catalog functions
(SQLTables, etc), by default ( SQL_ATTR_METADATA_ID == false ) expect
either ordinary arguments (OA) in the case of the catalog, or pattern value
arguments (PV) in the case of schema/table name. For these, quoted
identifiers do not make sense, so we unquote identifiers prior to the call.
The ODBC documentation on SQLColumns for further details.