=begin
= DBI Interface Specification Version 0.2.2 (Draft)
by Michael Neumann (neumann@s-direknet.de)

 $Id: DBI_SPEC,v 1.1 2002/10/02 18:10:37 mneumann Exp $
 
== Module DBD

=== Constants

--- API_VERSION
    Use this in your DBD driver to ensure it is used with the correct DBD API-Version

== Module DBI

=== Constants

--- VERSION
    Version of the DBI Interface

--- SQL_FETCH_NEXT
--- SQL_FETCH_PRIOR
--- SQL_FETCH_FIRST
--- SQL_FETCH_LAST
--- SQL_FETCH_ABSOLUTE
    Constants for (({StatementHandle#fetch_scroll})).

--- SQL_BIT
--- SQL_TINYINT
--- SQL_SMALLINT
--- SQL_INTEGER
--- SQL_BIGINT
--- SQL_FLOAT
--- SQL_REAL
--- SQL_DOUBLE
--- SQL_NUMERIC
--- SQL_DECIMAL
--- SQL_CHAR
--- SQL_VARCHAR
--- SQL_LONGVARCHAR
--- SQL_DATE
--- SQL_TIME
--- SQL_TIMESTAMP
--- SQL_BINARY
--- SQL_VARBINARY
--- SQL_LONGVARBINARY
--- SQL_OTHER
    Constant representing SQL types.

=== Exceptions

Exception classes were "borrowed" from Python API 2.0.

--- Warning < RuntimeError
    For important warnings like data truncation etc.

--- Error < RuntimeError
    Base class of all other error exceptions.
    Use this to catch all errors.

--- InterfaceError < Error
    Exception for errors related to the DBI interface rather 
    than the database itself.

--- NotImplementedError < InterfaceError
    Exception raised if the DBD driver has not specified
    a mandantory method (not in Python API 2.0).

--- DatabaseError < Error
    Exception for errors related to the database.
 
    Has three attributes ((|err|)), ((|errstr|)) and ((|state|)).

--- DataError < DatabaseError
    Exception for errors due to problems with the processed 
    data like division by zero, numeric value out of range etc.

--- OperationalError < DatabaseError
    Exception for errors related to the database's operation which
    are not necessarily under the control of the programmer like
    unexpected disconnect, datasource name not found, transaction
    could not be processed, a memory allocation error occured during
    processing etc.

--- IntegrityError < DatabaseError
    Exception raised when the relational integrity of the database
    is affected, e.g. a foreign key check fails

--- InternalError < DatabaseError
    Exception raised when the database encounters an internal error, 
    e.g. the cursor is not valid anymore, the transaction is out of sync.

--- ProgrammingError < DatabaseError
    Exception raised for programming errors, e.g. table not found
    or already exists, syntax error in SQL statement, wrong number
    of parameters specified, etc.

--- NotSupportedError < DatabaseError
    Raised if e.g. commit() is called for a database which do not
    support transactions.


=== Module functions

--- DBI.connect( driver_url, user=nil, auth=nil, params=nil )
    Connect to the database specified by ((*driver_url*)), which may
    look like "dbi:Oracle:oracle.neumann".
 
    Returns an (({DBI::DatabaseHandle})) object, or if called with code-block,
    calls this block with the new (({DBI::DatabaseHandle})) as parameter and
    calls (({disconnect})) after calling the block if it was not yet disconnected by
    the user.

--- DBI.available_drivers
    Returns an (({Array})) of all available DBD driver.
    The strings which represent a DBD driver are partial DSN's
    (e.g. "dbi:Oracle:").

--- DBI.data_sources( driver )
    Returns all available DSN's for the ((*driver*)), which
    is a partial DSN (e.g. "dbi:Oracle:").

--- DBI.disconnect_all( driver=nil )
    Disconnects all active connections of ((*driver*)) or
    all drivers if ((*driver*)) is (({nil})).

--- DBI.trace(mode=nil, output=nil)
    Set the trace mode for all following created Handles to these values.

    If a parameter is (({nil})) the value is not changed.
    ((|mode|)) defaults to 2 if it is (({nil})), and ((|output|)) to (({STDERR})) if a value was not
    set before.
    For ((|mode|)) the values 0, 1, 2 or 3 are allowed.

    Note: Tracing is only activated, if you load the module "dbi/trace", because tracing currently
    depends on AspectR > 0.3.3.

== Class DBI::Handle

Abstract base class for all "Handles" (DriverHandle, DatabaseHandle, StatementHandle).

=== Instance Methods

--- func( function, *values )
    Calls the driver specific extension function named by
    ((|function|)) with ((|values|)) as parameters.

--- trace(mode=nil, output=nil)
    Set the trace mode for this handle as well as for all sub-handles (in the case of DriverHandle and 
    DatabaseHandle).

    If a parameter is (({nil})) the value is not changed.
    ((|mode|)) defaults to 2 if it is (({nil})), and ((|output|)) to (({STDERR})) if a value was not
    set before.
    For ((|mode|)) the values 0, 1, 2 or 3 are allowed.

    Note: Tracing is only activated, if you load the module "dbi/trace", because tracing currently
    depends on AspectR > 0.3.3.


== Class DBI::DatabaseHandle

=== Superclass

--- DBI::Handle


=== Instance Methods

--- connected?
    Returns (({true})) if the connection was not yet disconnected
    by calling ((<disconnect>)), otherwise (({false})).

--- disconnect
    Disconnects the connection.

--- prepare( stmt )
--- prepare( stmt ) {|statement_handle| aBlock}
    Prepare the SQL statement ((|stmt|)) and return a
    (({DBI::StatementHandle})) or if called with a code-block
    calls the block with the handle as parameter and after that
    calls (({#finish})) onto the handle to free all resources

--- execute( stmt, *bindvars )
--- execute( stmt, *bindvars ) {|statement_handle| aBlock}
    Executes immediately the SQL statement ((*stmt*)) with binding
    the placeholders with values given in ((*bindvars*)) before.

    Returns a (({DBI::StatementHandle})) or if called with code-block
    calls the block with the handle as parameter and after that
    calls (({#finish})) onto the handle to free all resources.

--- do( stmt, *bindvars )
    Same as ((<execute>)) only that no (({DBI::StatementHandle})) is
    returned but the RPC (Row Processed Count).
   
--- select_one( stmt, *bindvars)
    Executes the statement with binding the values to the parameters and returns the
    first row as a reference to a Row object. 

--- select_all( stmt, *bindvars)
    Executes the statement with binding the values to the parameters and returns all
    resulting rows. 

    If called as iterator the passed (({DBI::Row})) objects are only references.
 
--- tables
    Returns a list of all tables and views.

--- columns( table )
    Get more information about the columns of table ((|table|)). 
    Returns an array containing for each column one (({DBI::ColumnInfo})) object.

--- ping
    Returns (({true})) if the connection is active, otherwise (({false})).
    
    In contranst to ((<connected?>)), ((<ping>)) tests if the connection is
    still active by executing some SQL or doing something else.
    
--- quote( value )
    Quotes the given value ((*value*)) database specific and returns the result.

--- commit
    Commits current transaction.

--- rollback
    Rolls the current transaction back.

--- transaction {|database_handle| aBlock}
    First commits the current transaction, then
    executes the given block where the parameter is
    the object itself (the database handle). If the
    block raises an exception, then it rolls the transaction
    back otherwise commits it.

--- [](attr)
--- []=(attr)
    Sets or gets the attribute ((*attr*)).
    An attribute can for example be "AutoCommit", which can be set to 
    (({true})) or (({false})). Attributes are database dependant.

== Class DBI::StatementHandle

=== Superclass

--- DBI::Handle

=== Mixins

--- Enumerable

=== Instance Methods

--- bind_param( param, value, attribs=nil )
    Bind ((*param*)) which is either a (({String})) which is then the name of the 
    placeholder used in the SQL statement (e.g. Oracle: "SELECT * FROM EMP WHERE ENAME = :ename") 
    or it is an integer which is then the number of the placeholder where counting starts at 1.

    ((*value*)) is the value which is bound to the placeholder.

    ((*attribs*)) is not yet used in this version, but could later be a hash containing more information
    like parameter type etc..


--- execute( *bindvars )
    Execute the statement but before binds the placeholders with ((*bindvars*)).

--- finish
    Frees the resources for the statement.
    After calling ((<finish>)) no other operation on this
    statement is valid.

--- cancel
    Frees any result set resources which were made after a call 
    to (({execute})).
    After calling this method, a call to one of the ((*fetch*)) methods 
    is no more valid.

--- column_names
    Returns an (({Array})) of all column names.


--- column_info
    Returns an (({Array})) containing fore each column one (({DBI::ColumnInfo})) object.

--- rows
    Returns the RPC (Row Processed Count) of the last executed statement, or
    (({nil})) if no such exist.


--- fetchable?
    Returns true if you can fetch rows using fetch etc..
 

--- fetch 
    Returns a (({DBI::Row})) object, or (({nil})) if there are
    no more rows to fetch.

    When called as iterator, the block is called for each row
    until no more row is available. Each row is passed to the
    block as (({DBI::Row})) object.

    Note that the returned or passed (({DBI::Row})) object is only a reference and
    should be copied (dup) if it is store elsewhere.


--- each  {|row| aBlock }
    Same as ((<fetch>)) called as iterator.


--- fetch_array
    Returns the current row as (({Array})) or nil if no more
    row is available.

    Can be also called as iterator.

--- fetch_hash 
    Returns the current row as (({Hash})) or nil if no more
    row is available.
    
    Can be also called as iterator.

--- fetch_many( cnt )
    Returns an (({Array})) of the next ((*cnt*)) rows, which are
    stored as (({DBI::Row})) objects. 

    Returns the empty array (({[]})) if there are no more rows.

--- fetch_all
    Same as ((<fetch_many>)) only that all rows are returned.
 
--- fetch_scroll( direction, offset=1 )
    ((*direction*)) is one of the following constants:
    * SQL_FETCH_NEXT
    * SQL_FETCH_PRIOR
    * SQL_FETCH_FIRST
    * SQL_FETCH_LAST
    * SQL_FETCH_ABSOLUTE
    * SQL_FETCH_RELATIVE
    
    ((*offset*)) is a positive or negativ number (only when SQL_FETCH_RELATIVE is used).

    ((<fetch_scroll>)) do not automatically free the result set if no more rows are available
    if e.g. you get the last row. 
    
    Returns a (({DBI::Row})) object, if not possible, returns (({nil})).

    Note that the returned (({DBI::Row})) object is only a reference and
    should be copied (dup) if it is stored elsewhere.

--- [](attr)
--- []=(attr)
    Sets or gets the attribute ((*attr*)).

=end