ibm_db_dbi.py

# +--------------------------------------------------------------------------+
# |  Licensed Materials - Property of IBM                                    |
# |                                                                          |
# | (C) Copyright IBM Corporation 2007-2015                                  |
# +--------------------------------------------------------------------------+
# | This module complies with SQLAlchemy and is                              |
# | Licensed under the Apache License, Version 2.0 (the "License");          |
# | you may not use this file except in compliance with the License.         |
# | You may obtain a copy of the License at                                  |
# | http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable |
# | law or agreed to in writing, software distributed under the License is   |
# | distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
# | KIND, either express or implied. See the License for the specific        |
# | language governing permissions and limitations under the License.        |
# +--------------------------------------------------------------------------+
# | Authors: Swetha Patel, Abhigyan Agrawal, Tarun Pasrija, Rahul Priyadarshi,
# |          Akshay Anand, Saba Kauser
# +--------------------------------------------------------------------------+

"""
This module implements the Python DB API Specification v2.0 for DB2 database.
"""

import types, string, time, datetime, decimal, sys
import weakref
import logging as log_ibmdb_dbi

logger = log_ibmdb_dbi.getLogger(__name__)
log_enabled = False

# Define macros for log levels
DEBUG = "DEBUG"
INFO = "INFO"
WARNING = "WARNING"
ERROR = "ERROR"
EXCEPTION = "EXCEPTION"


def debug(option):
    global log_enabled
    if isinstance(option, bool):
        log_enabled = option
        if log_enabled:
            log_ibmdb_dbi.basicConfig(level=log_ibmdb_dbi.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')
    elif isinstance(option, str):
        log_enabled = True  # Set log_enabled to True
        if '.' not in option:
            option += '.txt'
        log_ibmdb_dbi.basicConfig(filename=option, level=log_ibmdb_dbi.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s',
                            filemode='w')
    else:
        print("Invalid argument in debug. Please give a boolean or a file name in string.")


def LogMsg(log_level, message):
    if log_enabled:
        if log_level == DEBUG:
            logger.debug(message)
        elif log_level == INFO:
            logger.info(message)
        elif log_level == WARNING:
            logger.warning(message)
        elif log_level == ERROR:
            logger.error(message)
        elif log_level == EXCEPTION:
            logger.exception(message)

PY2 = sys.version_info < (3, )

if not PY2:
    buffer = memoryview
if PY2:
    import exceptions

    exception = exceptions.Exception
else:
    exception = Exception

if PY2:
    import __builtin__

    string_types = __builtin__.unicode, bytes
    int_types = __builtin__.long, int
else:
    string_types = str
    int_types = int

import ibm_db

__version__ = ibm_db.__version__

# Constants for specifying database connection options.
SQL_ATTR_AUTOCOMMIT = ibm_db.SQL_ATTR_AUTOCOMMIT
SQL_ATTR_CURRENT_SCHEMA = ibm_db.SQL_ATTR_CURRENT_SCHEMA
SQL_AUTOCOMMIT_OFF = ibm_db.SQL_AUTOCOMMIT_OFF
SQL_AUTOCOMMIT_ON = ibm_db.SQL_AUTOCOMMIT_ON
ATTR_CASE = ibm_db.ATTR_CASE
CASE_NATURAL = ibm_db.CASE_NATURAL
CASE_LOWER = ibm_db.CASE_LOWER
CASE_UPPER = ibm_db.CASE_UPPER
SQL_FALSE = ibm_db.SQL_FALSE
SQL_TRUE = ibm_db.SQL_TRUE
SQL_TABLE_STAT = ibm_db.SQL_TABLE_STAT
SQL_INDEX_CLUSTERED = ibm_db.SQL_INDEX_CLUSTERED
SQL_INDEX_OTHER = ibm_db.SQL_INDEX_OTHER
SQL_DBMS_VER = ibm_db.SQL_DBMS_VER
SQL_DBMS_NAME = ibm_db.SQL_DBMS_NAME
USE_WCHAR = ibm_db.USE_WCHAR
WCHAR_YES = ibm_db.WCHAR_YES
WCHAR_NO = ibm_db.WCHAR_NO
FIX_RETURN_TYPE = 1

SQL_ATTR_TXN_ISOLATION = ibm_db.SQL_ATTR_TXN_ISOLATION
SQL_TXN_READ_UNCOMMITTED = ibm_db.SQL_TXN_READ_UNCOMMITTED
SQL_TXN_READ_COMMITTED = ibm_db.SQL_TXN_READ_COMMITTED
SQL_TXN_REPEATABLE_READ = ibm_db.SQL_TXN_REPEATABLE_READ
SQL_TXN_SERIALIZABLE = ibm_db.SQL_TXN_SERIALIZABLE
SQL_TXN_NO_COMMIT = ibm_db.SQL_TXN_NO_COMMIT

# Module globals
apilevel = '2.0'
threadsafety = 0
paramstyle = 'qmark'
class Error(exception):
    """This is the base class of all other exception thrown by this
    module.  It can be use to catch all exceptions with a single except
    statement.

    """

    def __init__(self, message):
        """This is the constructor which take one string argument."""
        self._message = message
        super(Error, self).__init__(message)

    def __str__(self):
        """Converts the message to a string."""
        return 'ibm_db_dbi::' + str(self.__class__.__name__) + ': ' + str(self._message)


class Warning(exception):
    """This exception is used to inform the user about important
    warnings such as data truncations.

    """

    def __init__(self, message):
        """This is the constructor which take one string argument."""
        self._message = message
        super(Warning, self).__init__(message)

    def __str__(self):
        """Converts the message to a string."""
        return 'ibm_db_dbi::' + str(self.__class__.__name__) + ': ' + str(self._message)


class InterfaceError(Error):
    """This exception is raised when the module interface is being
    used incorrectly.

    """
    pass


class DatabaseError(Error):
    """This exception is raised for errors related to database."""
    pass


class InternalError(DatabaseError):
    """This exception is raised when internal database error occurs,
    such as cursor is not valid anymore.

    """
    pass


class OperationalError(DatabaseError):
    """This exception is raised when database operation errors that are
    not under the programmer control occur, such as unexpected
    disconnect.

    """
    pass


class ProgrammingError(DatabaseError):
    """This exception is raised for programming errors, such as table
    not found.

    """
    pass


class IntegrityError(DatabaseError):
    """This exception is thrown when errors occur when the relational
    integrity of database fails, such as foreign key check fails.

    """
    pass


class DataError(DatabaseError):
    """This exception is raised when errors due to data processing,
    occur, such as divide by zero.

    """
    pass


class NotSupportedError(DatabaseError):
    """This exception is thrown when a method in this module or an
    database API is not supported.

    """
    pass


def Date(year, month, day):
    """This method can be used to get date object from integers, for
    inserting it into a DATE column in the database.

    """
    return datetime.date(year, month, day)


def Time(hour, minute, second):
    """This method can be used to get time object from integers, for
    inserting it into a TIME column in the database.

    """
    return datetime.time(hour, minute, second)


def Timestamp(year, month, day, hour, minute, second):
    """This method can be used to get timestamp object from integers,
    for inserting it into a TIMESTAMP column in the database.

    """
    return datetime.datetime(year, month, day, hour, minute, second)


def DateFromTicks(ticks):
    """This method can be used to get date object from ticks seconds,
    for inserting it into a DATE column in the database.

    """
    time_tuple = time.localtime(ticks)
    return datetime.date(time_tuple[0], time_tuple[1], time_tuple[2])


def TimeFromTicks(ticks):
    """This method can be used to get time object from ticks seconds,
    for inserting it into a TIME column in the database.

    """
    time_tuple = time.localtime(ticks)
    return datetime.time(time_tuple[3], time_tuple[4], time_tuple[5])


def TimestampFromTicks(ticks):
    """This method can be used to get timestamp object from ticks
    seconds, for inserting it into a TIMESTAMP column in the database.

    """
    time_tuple = time.localtime(ticks)
    return datetime.datetime(time_tuple[0], time_tuple[1], time_tuple[2],
                             time_tuple[3], time_tuple[4], time_tuple[5])


def Binary(string):
    """This method can be used to store binary information, for
    inserting it into a binary type column in the database.

    """
    if not isinstance(string, (bytes, memoryview)):
        raise InterfaceError("Binary function expects type string argument.")
    return buffer(string)


class DBAPITypeObject(frozenset):
    """Class used for creating objects that can be used to compare
    in order to determine the python type to provide in parameter
    sequence argument of the execute method.

    """

    def __new__(cls, col_types):
        return frozenset.__new__(cls, col_types)

    def __init__(self, col_types):
        """Constructor for DBAPITypeObject.  It takes a tuple of
        database column type as an argument.
        """
        self.col_types = col_types

    def __cmp__(self, cmp):
        """This method checks if the string compared with is in the
        tuple provided to the constructor of this object.  It takes
        string as an argument.
        """
        if cmp in self.col_types:
            return 0
        if sys.version_info < (3,):
            if cmp < self.col_types:
                return 1
            else:
                return -1
        else:
            return 1

    def __eq__(self, cmp):
        """This method checks if the string compared with is in the
        tuple provided to the constructor of this object.  It takes
        string as an argument.
        """
        return cmp in self.col_types

    def __ne__(self, cmp):
        """This method checks if the string compared with is not in the
        tuple provided to the constructor of this object.  It takes
        string as an argument.
        """
        return cmp not in self.col_types

    def __hash__(self):
        return id(self)


# The user can use these objects to compare the database column types
# with in order to determine the python type to provide in the
# parameter sequence argument of the execute method.
STRING = DBAPITypeObject(("CHARACTER", "CHAR", "VARCHAR",
                          "CHARACTER VARYING", "CHAR VARYING", "STRING",))

TEXT = DBAPITypeObject(("CLOB", "CHARACTER LARGE OBJECT", "CHAR LARGE OBJECT", "DBCLOB"))

XML = DBAPITypeObject(("XML",))

BINARY = DBAPITypeObject(("BLOB", "BINARY LARGE OBJECT",))

NUMBER = DBAPITypeObject(("INTEGER", "INT", "SMALLINT",))

BIGINT = DBAPITypeObject(("BIGINT",))

FLOAT = DBAPITypeObject(("FLOAT", "REAL", "DOUBLE", "DECFLOAT"))

DECIMAL = DBAPITypeObject(("DECIMAL", "DEC", "NUMERIC", "NUM",))

DATE = DBAPITypeObject(("DATE",))

TIME = DBAPITypeObject(("TIME",))

DATETIME = DBAPITypeObject(("TIMESTAMP",))

ROWID = DBAPITypeObject(())

BOOLEAN = DBAPITypeObject(("BOOLEAN",))


# This method is used to determine the type of error that was
# generated.  It takes an exception instance as an argument, and
# returns exception object of the appropriate type.
def _get_exception(inst):
    # These tuple are used to determine the type of exceptions that are
    # thrown by the database.  They store the SQLSTATE code and the
    # SQLSTATE class code(the 2 digit prefix of the SQLSTATE code)
    warning_error_tuple = ('01',)
    data_error_tuple = ('02', '22', '10601', '10603', '10605', '10901', '10902',
                        '38552', '54')

    operational_error_tuple = ('08', '09', '10502', '10000', '10611', '38501',
                               '38503', '38553', '38H01', '38H02', '38H03', '38H04',
                               '38H05', '38H06', '38H07', '38H09', '38H0A')

    integrity_error_tuple = ('23',)

    internal_error_tuple = ('24', '25', '26', '2D', '51', '57')

    programming_error_tuple = ('08002', '07', 'OD', 'OF', 'OK', 'ON', '10', '27',
                               '28', '2E', '34', '36', '38', '39', '56', '42',
                               '3B', '40', '44', '53', '55', '58', '5U', '21')

    not_supported_error_tuple = ('0A', '10509')

    # These tuple are used to determine the type of exceptions that are
    # thrown from the driver module.
    interface_exceptions = ("Supplied parameter is invalid",
                            "ATTR_CASE attribute must be one of "
                            "CASE_LOWER, CASE_UPPER, or CASE_NATURAL",
                            "Connection or statement handle must be passed in.",
                            "Param is not a tuple")

    programming_exceptions = ("Connection is not active",
                              "qualifier must be a string",
                              "unique must be a boolean",
                              "Parameters not bound",
                              "owner must be a string",
                              "table_name must be a string",
                              "table type must be a string",
                              "column_name must be a string",
                              "Column ordinal out of range",
                              "procedure name must be a string",
                              "Requested row number must be a positive value",
                              "Options Array must have string indexes")

    database_exceptions = ("Binding Error",
                           "Column information cannot be retrieved: ",
                           "Column binding cannot be done: ",
                           "Failed to Determine XML Size: ")

    statement_exceptions = ("Statement Execute Failed: ",
                            "Describe Param Failed: ",
                            "Sending data failed: ",
                            "Fetch Failure: ",
                            "SQLNumResultCols failed: ",
                            "SQLRowCount failed: ",
                            "SQLGetDiagField failed: ",
                            "Statement prepare Failed: ")

    operational_exceptions = ("Connection Resource cannot be found",
                              "Failed to Allocate Memory",
                              "Describe Param Failed: ",
                              "Statement Execute Failed: ",
                              "Sending data failed: ",
                              "Failed to Allocate Memory for XML Data",
                              "Failed to Allocate Memory for LOB Data")

    # First check if the exception is from the database.  If it is
    # determine the SQLSTATE code which is used further to determine
    # the exception type.  If not check if the exception is thrown by
    # by the driver and return the appropriate exception type.  If it
    # is not possible to determine the type of exception generated
    # return the generic Error exception.
    if inst is not None:
        message = repr(inst)
        if message.startswith("Exception('"):
            if message.endswith("',)"):  # python 2
                message = message[11:]
                message = message[:len(message) - 3]
            elif message.endswith("')"):  # python 3
                message = message[11:]
                message = message[:len(message) - 2]

        index = message.find('SQLSTATE=')
        if (message != '') & (index != -1):
            error_code = message[(index + 9):(index + 14)]
            prefix_code = error_code[:2]
        else:
            for key in interface_exceptions:
                if message.find(key) != -1:
                    return InterfaceError(message)
            for key in programming_exceptions:
                if message.find(key) != -1:
                    return ProgrammingError(message)
            for key in operational_exceptions:
                if message.find(key) != -1:
                    return OperationalError(message)
            for key in database_exceptions:
                if message.find(key) != -1:
                    return DatabaseError(message)
            for key in statement_exceptions:
                if message.find(key) != -1:
                    return DatabaseError(message)
            return Error(message)
    else:
        return Error('An error has occured')

    # First check if the SQLSTATE is in the tuples, if not check
    # if the SQLSTATE class code is in the tuples to determine the
    # exception type.
    if (error_code in warning_error_tuple or
            prefix_code in warning_error_tuple):
        return Warning(message)
    if (error_code in data_error_tuple or
            prefix_code in data_error_tuple):
        return DataError(message)
    if (error_code in operational_error_tuple or
            prefix_code in operational_error_tuple):
        return OperationalError(message)
    if (error_code in integrity_error_tuple or
            prefix_code in integrity_error_tuple):
        return IntegrityError(message)
    if (error_code in internal_error_tuple or
            prefix_code in internal_error_tuple):
        return InternalError(message)
    if (error_code in programming_error_tuple or
            prefix_code in programming_error_tuple):
        return ProgrammingError(message)
    if (error_code in not_supported_error_tuple or
            prefix_code in not_supported_error_tuple):
        return NotSupportedError(message)
    return DatabaseError(message)


def _retrieve_current_schema(dsn):
    """This method retrieve the value of ODBC keyword CURRENTSCHEMA from DSN
    """
    LogMsg(INFO, "entry _retrieve_current_schema()")
    ODBC_CURRENTSCHEMA_KEYWORD = 'CURRENTSCHEMA='
    current_schema_value = None
    current_schema_start = dsn.find(ODBC_CURRENTSCHEMA_KEYWORD)
    message = f"current_schema_start: {current_schema_start}"
    LogMsg(DEBUG, message)

    if current_schema_start > -1:
        current_schema_end = dsn.find(';', current_schema_start)
        message = f"ODBC_CURRENTSCHEMA_KEYWORD: {ODBC_CURRENTSCHEMA_KEYWORD}"
        LogMsg(DEBUG, message)
        LogMsg(DEBUG, f"current_schema_end: {current_schema_end}")
        current_schema_value = dsn[
                               (current_schema_start + len(ODBC_CURRENTSCHEMA_KEYWORD))
                               :current_schema_end
                               ]
    LogMsg(DEBUG, f"current_schema_value: {current_schema_value}")
    LogMsg(INFO, "exit _retrieve_current_schema()")
    return current_schema_value


def _server_connect(dsn, user='', password='', host=''):
    """This method create connection with server
    """
    LogMsg(INFO, "entry _server_connect()")
    if dsn is None:
        LogMsg(ERROR, "dsn value should not be None")
        raise InterfaceError("dsn value should not be None")

    if (not isinstance(dsn, string_types)) | \
            (not isinstance(user, string_types)) | \
            (not isinstance(password, string_types)) | \
            (not isinstance(host, string_types)):
        LogMsg(ERROR, "Arguments should be of type string or unicode")
        raise InterfaceError("Arguments should be of type string or unicode")

    # If the dsn does not contain port and protocal adding database
    # and hostname is no good.  Add these when required, that is,
    # if there is a '=' in the dsn.  Else the dsn string is taken to be
    # a DSN entry.
    if dsn.find('=') != -1:
        if dsn[len(dsn) - 1] != ';':
            dsn = dsn + ";"
        if host != '' and dsn.find('HOSTNAME=') == -1:
            dsn = dsn + "HOSTNAME=" + host + ";"
    else:
        dsn = "DSN=" + dsn + ";"

    # attach = true is not valid against IDS. And attach is not needed for connect currently.
    #if dsn.find('attach=') == -1:
    #dsn = dsn + "attach=true;"
    if user != '' and dsn.find('UID=') == -1:
        dsn = dsn + "UID=" + user + ";"
    if password != '' and dsn.find('PWD=') == -1:
        dsn = dsn + "PWD=" + password + ";"
    try:
        conn = ibm_db.connect(dsn, '', '')
    except Exception as inst:
        message = f"An exception occurred while connecting to server: {inst}"
        LogMsg(EXCEPTION, message)
        raise _get_exception(inst)
    LogMsg(INFO, "exit _server_connect()")
    return conn


def createdb(database, dsn, user='', password='', host='', codeset='', mode=''):
    """This method creates a database by using the specified database name, code set, and mode
    """
    LogMsg(INFO, "entry createdb()")
    if database is None:
        LogMsg(ERROR, "createdb expects a not None database name value")
        raise InterfaceError("createdb expects a not None database name value")
    if (not isinstance(database, string_types)) | \
            (not isinstance(codeset, string_types)) | \
            (not isinstance(mode, string_types)):
        LogMsg(ERROR, "Arguments should be string or unicode")
        raise InterfaceError("Arguments should be string or unicode")

    conn = _server_connect(dsn, user=user, password=password, host=host)
    try:
        return_value = ibm_db.createdb(conn, database, codeset, mode)
        LogMsg(DEBUG, f"Return value from ibm_db.createdb: {return_value}")
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred during database creation: {inst}")
        raise _get_exception(inst)
    finally:
        try:
            ibm_db.close(conn)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while closing connection: {inst}")
            raise _get_exception(inst)
    LogMsg(INFO, "exit createdb()")
    return return_value


def dropdb(database, dsn, user='', password='', host=''):
    """This method drops the specified database
    """
    LogMsg(INFO, "entry dropdb()")
    if database is None:
        LogMsg(ERROR, "dropdb expects a not None database name value")
        raise InterfaceError("dropdb expects a not None database name value")
    if not isinstance(database, string_types):
        LogMsg(ERROR, "Arguments should be string or unicode")
        raise InterfaceError("Arguments should be string or unicode")

    conn = _server_connect(dsn, user=user, password=password, host=host)
    try:
        return_value = ibm_db.dropdb(conn, database)
        LogMsg(DEBUG, f"Return value from ibm_db.dropdb: {return_value}")
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred during database droping: {inst}")
        raise _get_exception(inst)
    finally:
        try:
            ibm_db.close(conn)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while closing connection: {inst}")
            raise _get_exception(inst)
    LogMsg(INFO, "exit dropdb()")
    return return_value


def recreatedb(database, dsn, user='', password='', host='', codeset='', mode=''):
    """This method drops and then recreate the database by using the specified database name, code set, and mode
    """
    LogMsg(INFO, "entry recreatedb()")
    if database is None:
        LogMsg(ERROR, "recreatedb expects a not None database name value")
        raise InterfaceError("recreatedb expects a not None database name value")
    if (not isinstance(database, string_types)) | \
            (not isinstance(codeset, string_types)) | \
            (not isinstance(mode, string_types)):
        LogMsg(ERROR, "Arguments should be string or unicode")
        raise InterfaceError("Arguments should be string or unicode")

    conn = _server_connect(dsn, user=user, password=password, host=host)
    try:
        return_value = ibm_db.recreatedb(conn, database, codeset, mode)
        LogMsg(DEBUG, f"Return value from ibm_db.recreatedb: {return_value}")
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred during database recreation: {inst}")
        raise _get_exception(inst)
    finally:
        try:
            ibm_db.close(conn)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while closing connection: {inst}")
            raise _get_exception(inst)
    LogMsg(INFO, "exit recreatedb()")
    return return_value


def createdbNX(database, dsn, user='', password='', host='', codeset='', mode=''):
    """This method creates a database if it not exist by using the specified database name, code set, and mode
    """
    LogMsg(INFO, "entry createdbNX()")
    if database is None:
        LogMsg(ERROR, "createdbNX expects a not None database name value")
        raise InterfaceError("createdbNX expects a not None database name value")
    if (not isinstance(database, string_types)) | \
            (not isinstance(codeset, string_types)) | \
            (not isinstance(mode, string_types)):
        LogMsg(ERROR, "Arguments should be string or unicode")
        raise InterfaceError("Arguments should be string or unicode")

    conn = _server_connect(dsn, user=user, password=password, host=host)
    try:
        return_value = ibm_db.createdbNX(conn, database, codeset, mode)
        LogMsg(DEBUG, f"Return value from ibm_db.createdbNX: {return_value}")
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred during database createdbNX: {inst}")
        raise _get_exception(inst)
    finally:
        try:
            ibm_db.close(conn)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while closing connection: {inst}")
            raise _get_exception(inst)
    LogMsg(INFO, "exit createdbNX()")
    return return_value


def connect(dsn, user='', password='', host='', database='', conn_options=None):
    """This method creates a non-persistent connection to the database. It returns
        a ibm_db_dbi.Connection object.
    """
    LogMsg(INFO, "entry connect()")
    try:
        message = f"dsn: {dsn}, user: {user}, host: {host}, database: {database}, conn_options: {conn_options}"
        LogMsg(DEBUG, message)

        if dsn is None:
            LogMsg(ERROR, "connect expects a not None dsn value")
            raise InterfaceError("connect expects a not None dsn value")

        if (not isinstance(dsn, string_types)) | \
                (not isinstance(user, string_types)) | \
                (not isinstance(password, string_types)) | \
                (not isinstance(host, string_types)) | \
                (not isinstance(database, string_types)):
            message = "connect expects the first five arguments to be of type string or unicode"
            LogMsg(ERROR, message)
            raise InterfaceError("connect expects the first five arguments to"
                                 " be of type string or unicode")
        if conn_options is not None:
            if not isinstance(conn_options, dict):
                message = "connect expects the sixth argument (conn_options) to be of type dict"
                LogMsg(ERROR, message)
                raise InterfaceError("connect expects the sixth argument"
                                     " (conn_options) to be of type dict")
            if not SQL_ATTR_AUTOCOMMIT in conn_options:
                conn_options[SQL_ATTR_AUTOCOMMIT] = SQL_AUTOCOMMIT_OFF
        else:
            conn_options = {SQL_ATTR_AUTOCOMMIT: SQL_AUTOCOMMIT_OFF}

        # If the dsn does not contain port and protocol adding database
        # and hostname is no good.  Add these when required, that is,
        # if there is a '=' in the dsn.  Else the dsn string is taken to be
        # a DSN entry.
        if dsn.find('=') != -1:
            if dsn[len(dsn) - 1] != ';':
                dsn = dsn + ";"
            if database != '' and dsn.find('DATABASE=') == -1:
                dsn = dsn + "DATABASE=" + database + ";"
            if host != '' and dsn.find('HOSTNAME=') == -1:
                dsn = dsn + "HOSTNAME=" + host + ";"
        else:
            dsn = "DSN=" + dsn + ";"

        if user != '' and dsn.find('UID=') == -1:
            dsn = dsn + "UID=" + user + ";"
        if password != '' and dsn.find('PWD=') == -1:
            dsn = dsn + "PWD=" + password + ";"

        LogMsg(DEBUG, f"Connection string: {dsn}")

        conn = ibm_db.connect(dsn, '', '', conn_options)
        conn_object = Connection(conn)
        conn_object.set_current_schema(_retrieve_current_schema(dsn) or user)
        LogMsg(INFO, "Connection successful.")
        LogMsg(INFO, "exit connect()")
        return conn_object
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred while connecting: {inst}")
        raise _get_exception(inst)


def pconnect(dsn, user='', password='', host='', database='', conn_options=None):
    """This method creates persistent connection to the database. It returns
        a ibm_db_dbi.Connection object.
    """
    LogMsg(INFO, "entry pconnect()")
    if dsn is None:
        LogMsg(ERROR, "connect expects a not None dsn value")
        raise InterfaceError("connect expects a not None dsn value")

    if (not isinstance(dsn, string_types)) | \
            (not isinstance(user, string_types)) | \
            (not isinstance(password, string_types)) | \
            (not isinstance(host, string_types)) | \
            (not isinstance(database, string_types)):
        LogMsg(ERROR, "connect expects the first five arguments to be of type string or unicode")
        raise InterfaceError("connect expects the first five arguments to"
                             " be of type string or unicode")
    if conn_options is not None:
        if not isinstance(conn_options, dict):
            LogMsg(ERROR, "connect expects the sixth argument (conn_options) to be of type dict")
            raise InterfaceError("connect expects the sixth argument"
                                 " (conn_options) to be of type dict")
        if not SQL_ATTR_AUTOCOMMIT in conn_options:
            conn_options[SQL_ATTR_AUTOCOMMIT] = SQL_AUTOCOMMIT_OFF
    else:
        conn_options = {SQL_ATTR_AUTOCOMMIT: SQL_AUTOCOMMIT_OFF}

    # If the dsn does not contain port and protocal adding database
    # and hostname is no good.  Add these when required, that is,
    # if there is a '=' in the dsn.  Else the dsn string is taken to be
    # a DSN entry.
    if dsn.find('=') != -1:
        if dsn[len(dsn) - 1] != ';':
            dsn = dsn + ";"
        if database != '' and dsn.find('DATABASE=') == -1:
            dsn = dsn + "DATABASE=" + database + ";"
        if host != '' and dsn.find('HOSTNAME=') == -1:
            dsn = dsn + "HOSTNAME=" + host + ";"
    else:
        dsn = "DSN=" + dsn + ";"

    if user != '' and dsn.find('UID=') == -1:
        dsn = dsn + "UID=" + user + ";"
    if password != '' and dsn.find('PWD=') == -1:
        dsn = dsn + "PWD=" + password + ";"
    try:
        LogMsg(DEBUG, f"Connecting to database with DSN: {dsn}")
        conn = ibm_db.pconnect(dsn, '', '', conn_options)
        LogMsg(INFO, "Connection established successfully")
        conn_object = Connection(conn)
        conn_object.set_current_schema(_retrieve_current_schema(dsn) or user)
        LogMsg(INFO, "exit pconnect()")
        return conn_object
    except Exception as inst:
        LogMsg(EXCEPTION, f"An exception occurred while connecting: {inst}")
        raise _get_exception(inst)


class Connection(object):
    """This class object represents a connection between the database
    and the application.

    """

    def __init__(self, conn_handler):
        """Constructor for Connection object. It takes ibm_db
        connection handler as an argument.

        """
        self.conn_handler = conn_handler

        # Used to identify close cursors for generating exceptions
        # after the connection is closed.
        self._cursor_list = []
        self.__dbms_name = ibm_db.get_db_info(conn_handler, SQL_DBMS_NAME)
        self.__dbms_ver = ibm_db.get_db_info(conn_handler, SQL_DBMS_VER)
        self.FIX_RETURN_TYPE = 1

    # This method is used to get the DBMS_NAME
    def __get_dbms_name(self):
        return self.__dbms_name

    # This attribute specifies the DBMS_NAME
    # It is a read only attribute.
    dbms_name = property(__get_dbms_name, None, None, "")

    # This method is used to get the DBMS_ver
    def __get_dbms_ver(self):
        return self.__dbms_ver

    # This attribute specifies the DBMS_ver
    # It is a read only attribute.
    dbms_ver = property(__get_dbms_ver, None, None, "")

    def close(self):
        """This method closes the Database connection associated with
        the Connection object.  It takes no arguments.

        """
        LogMsg(INFO, "entry close()")
        self.rollback()
        try:
            if self.conn_handler is None:
                LogMsg(ERROR, "Connection cannot be closed; connection is no longer active.")
                raise ProgrammingError("Connection cannot be closed; "
                                       "connection is no longer active.")
            else:
                return_value = ibm_db.close(self.conn_handler)
                LogMsg(INFO, "Connection closed.")
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while closing connection: {inst}")
            raise _get_exception(inst)
        self.conn_handler = None
        for index in range(len(self._cursor_list)):
            if (self._cursor_list[index]() != None):
                tmp_cursor = self._cursor_list[index]()
                tmp_cursor.conn_handler = None
                tmp_cursor.stmt_handler = None
                tmp_cursor._all_stmt_handlers = None
        self._cursor_list = []
        LogMsg(INFO, "exit close()")
        return return_value

    def commit(self):
        """This method commits the transaction associated with the
        Connection object.  It takes no arguments.

        """
        LogMsg(INFO, "entry commit()")
        try:
            return_value = ibm_db.commit(self.conn_handler)
            LogMsg(INFO, "Transaction committed.")
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while committing transaction: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit commit()")
        return return_value

    def rollback(self):
        """This method rollbacks the transaction associated with the
        Connection object.  It takes no arguments.

        """
        LogMsg(INFO, "entry rollback()")
        try:
            return_value = ibm_db.rollback(self.conn_handler)
            LogMsg(INFO, "Transaction rolled back.")
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while rolling back transaction: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit rollback()")
        return return_value

    def cursor(self):
        """This method returns a Cursor object associated with the
        Connection.  It takes no arguments.

        """
        LogMsg(INFO, "entry cursor()")
        if self.conn_handler is None:
            LogMsg(ERROR, "Cursor cannot be returned; connection is no longer active.")
            raise ProgrammingError("Cursor cannot be returned; "
                                   "connection is no longer active.")
        cursor = Cursor(self.conn_handler, self)
        self._cursor_list.append(weakref.ref(cursor))
        LogMsg(INFO, "exit cursor()")
        return cursor

    # Sets connection attribute values
    def set_option(self, attr_dict):
        """Input: connection attribute dictionary
           Return: True on success or False on failure
        """
        LogMsg(INFO, "entry set_option()")
        LogMsg(DEBUG, f"Setting connection options: {attr_dict}")
        LogMsg(INFO, "exit set_option()")
        return ibm_db.set_option(self.conn_handler, attr_dict, 1)

    # Retrieves connection attributes values
    def get_option(self, attr_key):
        """Input: connection attribute key
           Return: current setting of the resource attribute requested
        """
        LogMsg(INFO, "entry get_option()")
        LogMsg(DEBUG, f"Getting connection option: {attr_key}")
        LogMsg(INFO, "exit get_option()")
        return ibm_db.get_option(self.conn_handler, attr_key, 1)

    # Sets FIX_RETURN_TYPE. Added for performance improvement
    def set_fix_return_type(self, is_on):
        LogMsg(INFO, "entry set_fix_return_type()")
        try:
            LogMsg(DEBUG, f"Setting fix return type to: {is_on}")
            if is_on:
                self.FIX_RETURN_TYPE = 1
            else:
                self.FIX_RETURN_TYPE = 0
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while setting fix return type: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit set_fix_return_type()")
        return self.FIX_RETURN_TYPE

    # Sets connection AUTOCOMMIT attribute
    def set_autocommit(self, is_on):
        """Input: connection attribute: true if AUTOCOMMIT ON, false otherwise (i.e. OFF)
           Return: True on success or False on failure
        """
        LogMsg(INFO, "entry set_autocommit()")
        try:
            LogMsg(DEBUG, f"Setting autocommit to: {is_on}")
            if is_on:
                is_set = ibm_db.set_option(self.conn_handler, {SQL_ATTR_AUTOCOMMIT: SQL_AUTOCOMMIT_ON}, 1)
            else:
                is_set = ibm_db.set_option(self.conn_handler, {SQL_ATTR_AUTOCOMMIT: SQL_AUTOCOMMIT_OFF}, 1)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while setting autocommit: {inst}")
            raise _get_exception(inst)
        LogMsg(DEBUG, f"set_autocommit returns: {is_set}")
        LogMsg(INFO, "exit set_autocommit()")
        return is_set

    # Sets connection attribute values
    def set_current_schema(self, schema_name):
        """Input: connection attribute dictionary
           Return: True on success or False on failure
        """
        LogMsg(INFO, "entry set_current_schema()")
        self.current_schema = schema_name
        try:
            LogMsg(DEBUG, f"Setting current schema to: {schema_name}")
            is_set = ibm_db.set_option(self.conn_handler, {SQL_ATTR_CURRENT_SCHEMA: schema_name}, 1)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred setting current schema: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit set_current_schema()")
        return is_set

    # Retrieves connection attributes values
    def get_current_schema(self):
        """Return: current setting of the schema attribute
        """
        LogMsg(INFO, "entry get_current_schema()")
        try:
            conn_schema = ibm_db.get_option(self.conn_handler, SQL_ATTR_CURRENT_SCHEMA, 1)
            if conn_schema is not None and conn_schema != '':
                self.current_schema = conn_schema
            LogMsg(DEBUG, f"Current schema: {self.current_schema}")
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while getting current schema: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit get_current_schema()")
        return self.current_schema

    # Retrieves the IBM Data Server version for a given Connection object
    def server_info(self):
        """Return: tuple (DBMS_NAME, DBMS_VER)
        """
        LogMsg(INFO, "entry server_info()")
        try:
            server_info = []
            server_info.append(self.dbms_name)
            server_info.append(self.dbms_ver)
            LogMsg(INFO, f"Server info: {server_info}")
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while getting server info: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit server_info()")
        return tuple(server_info)

    def set_case(self, server_type, str_value):
        LogMsg(INFO, "entry set_case()")
        LogMsg(DEBUG, f"set case to {str_value}")
        LogMsg(INFO, "exit set_case()")
        return str_value.upper()

    # Retrieves the tables for a specified schema (and/or given table name)
    def tables(self, schema_name=None, table_name=None):
        """Input: connection - ibm_db.IBM_DBConnection object
           Return: sequence of table metadata dicts for the specified schema
        """
        LogMsg(INFO, "entry tables()")
        LogMsg(INFO, f"Retrieving the tables for a specified schema")
        result = []
        if schema_name is not None:
            schema_name = self.set_case("DB2_LUW", schema_name)
            LogMsg(INFO, f"Schema name: {schema_name}")
        if table_name is not None:
            table_name = self.set_case("DB2_LUW", table_name)
            LogMsg(INFO, f"Table name: {table_name}")

        try:
            stmt = ibm_db.tables(self.conn_handler, None, schema_name, table_name)
            LogMsg(DEBUG, f"Statement to retrieve table: {stmt}")
            row = ibm_db.fetch_assoc(stmt)
            i = 0
            while (row):
                result.append(row)
                i += 1
                row = ibm_db.fetch_assoc(stmt)
            ibm_db.free_result(stmt)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while retrieving the tables: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit tables()")
        return result

    # Retrieves metadata pertaining to index for specified schema (and/or table name)
    def indexes(self, unique=True, schema_name=None, table_name=None):
        """Input: connection - ibm_db.IBM_DBConnection object
           Return: sequence of index metadata dicts for the specified table
        Example:
           Index metadata retrieved from schema 'PYTHONIC.TEST_TABLE' table
           {
           'TABLE_SCHEM':       'PYTHONIC',              'TABLE_CAT':          None,
           'TABLE_NAME':        'ENGINE_USERS',          'PAGES':              None,
           'COLUMN_NAME':       'USER_ID'                'FILTER_CONDITION':   None,
           'INDEX_NAME':        'SQL071201150750170',    'CARDINALITY':        None,
           'ORDINAL_POSITION':   1,                      'INDEX_QUALIFIER':   'SYSIBM',
           'TYPE':               3,
           'NON_UNIQUE':         0,
           'ASC_OR_DESC':       'A'
           }
        """
        LogMsg(INFO, "entry indexes()")
        LogMsg(INFO, f"Retrieving metadata pertaining to index for specified schema/Table")
        result = []
        if schema_name is not None:
            schema_name = self.set_case("DB2_LUW", schema_name)
            LogMsg(INFO, f"Schema name: {schema_name}")
        if table_name is not None:
            table_name = self.set_case("DB2_LUW", table_name)
            LogMsg(INFO, f"Table name: {table_name}")

        try:
            stmt = ibm_db.statistics(self.conn_handler, None, schema_name, table_name, unique)
            LogMsg(INFO, f"Statement to retrieving metadata pertaining to index: {stmt}")
            row = ibm_db.fetch_assoc(stmt)
            i = 0
            while (row):
                if row['TYPE'] == SQL_INDEX_OTHER:
                    result.append(row)
                i += 1
                row = ibm_db.fetch_assoc(stmt)
            ibm_db.free_result(stmt)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while retrieving metadata pertaining to index: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit indexes()")
        return result

    # Retrieves metadata pertaining to primary keys for specified schema (and/or table name)
    def primary_keys(self, unique=True, schema_name=None, table_name=None):
        """Input: connection - ibm_db.IBM_DBConnection object
           Return: sequence of PK metadata dicts for the specified table
        Example:
           PK metadata retrieved from 'PYTHONIC.ORDERS' table
           {
           'TABLE_SCHEM':  'PYTHONIC',                 'TABLE_CAT': None,
           'TABLE_NAME':   'ORDERS',
           'COLUMN_NAME':  'ORDER_ID'
           'PK_NAME':      'SQL071128122038680',
           'KEY_SEQ':       1
           }
        """
        LogMsg(INFO, "entry primary_keys()")
        LogMsg(INFO, f"Retrieving metadata pertaining to primary keys for specified schema/Table")
        result = []
        if schema_name is not None:
            schema_name = self.set_case("DB2_LUW", schema_name)
            LogMsg(INFO, f"Schema name: {schema_name}")
        if table_name is not None:
            table_name = self.set_case("DB2_LUW", table_name)
            LogMsg(INFO, f"Table name: {table_name}")

        try:
            stmt = ibm_db.primary_keys(self.conn_handler, None, schema_name, table_name)
            LogMsg(DEBUG, f"Statement to retrieving metadata pertaining to primary keys: {stmt}")
            row = ibm_db.fetch_assoc(stmt)
            i = 0
            while (row):
                result.append(row)
                i += 1
                row = ibm_db.fetch_assoc(stmt)
            ibm_db.free_result(stmt)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while retrieving metadata pertaining to primary keys: {inst}")
            raise _get_exception(inst)
        LogMsg(INFO, "exit primary_keys()")
        return result

    # Retrieves metadata pertaining to foreign keys for specified schema (and/or table name)
    def foreign_keys(self, unique=True, schema_name=None, table_name=None):
        """Input: connection - ibm_db.IBM_DBConnection object
           Return: sequence of FK metadata dicts for the specified table
        Example:
           FK metadata retrieved from 'PYTHONIC.ENGINE_EMAIL_ADDRESSES' table
           {
           'PKTABLE_SCHEM': 'PYTHONIC',                 'PKTABLE_CAT':    None,
           'PKTABLE_NAME':  'ENGINE_USERS',             'FKTABLE_CAT':    None,
           'PKCOLUMN_NAME': 'USER_ID',                  'UPDATE_RULE':    3,
           'PK_NAME':       'SQL071205090958680',       'DELETE_RULE':    3
           'KEY_SEQ':        1,                         'DEFERRABILITY':  7,
           'FK_NAME':       'SQL071205091000160',
           'FKCOLUMN_NAME': 'REMOTE_USER_ID',
           'FKTABLE_NAME':  'ENGINE_EMAIL_ADDRESSES',
           'FKTABLE_SCHEM': 'PYTHONIC'
           }
        """
        LogMsg(INFO, "entry foreign_keys()")
        LogMsg(INFO, f"Retrieving metadata pertaining to foreign keys for specified schema/Table")
        result = []
        if schema_name is not None:
            schema_name = self.set_case("DB2_LUW", schema_name)
            LogMsg(INFO, f"Schema name: {schema_name}")
        if table_name is not None:
            table_name = self.set_case("DB2_LUW", table_name)
            LogMsg(INFO, f"Table name: {table_name}")

        try:
            stmt = ibm_db.foreign_keys(self.conn_handler, None, None, None, None, schema_name, table_name)
            LogMsg(DEBUG, f"Statement to retrieving metadata pertaining to foreign keys: {stmt}")
            row = ibm_db.fetch_assoc(stmt)
            i = 0
            while (row):
                result.append(row)
                i += 1
                row = ibm_db.fetch_assoc(stmt)
            ibm_db.free_result(stmt)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while retrieving metadata pertaining foreign keys: {inst}")
            raise _get_exception(inst)

        LogMsg(INFO, "exit foreign_keys()")
        return result

    # Retrieves the columns for a specified schema (and/or table name and column name)
    def columns(self, schema_name=None, table_name=None, column_names=None):
        """Input: connection - ibm_db.IBM_DBConnection object
           Return: sequence of column metadata dicts for the specified schema
        Example:
           Column metadata retrieved from schema 'PYTHONIC.FOO' table, column 'A'
           {
           'TABLE_NAME':        'FOO',        'NULLABLE':           1,
           'ORDINAL_POSITION':   2L,          'REMARKS':            None,
           'COLUMN_NAME':       'A',          'BUFFER_LENGTH':      30L,
           'TYPE_NAME':         'VARCHAR',    'SQL_DATETIME_SUB':   None,
           'COLUMN_DEF':         None,        'DATA_TYPE':          12,
           'IS_NULLABLE':       'YES',        'SQL_DATA_TYPE':      12,
           'COLUMN_SIZE':        30L,         'TABLE_CAT':          None,
           'CHAR_OCTET_LENGTH':  30L,         'TABLE_SCHEM':       'PYTHONIC',
           'NUM_PREC_RADIX':     None,
           'DECIMAL_DIGITS':     None
           }
        """
        LogMsg(INFO, "entry columns()")
        LogMsg(INFO, f"Retrieving the columns for a specified schema/Table")
        result = []
        if schema_name is not None:
            schema_name = self.set_case("DB2_LUW", schema_name)
            LogMsg(INFO, f"Schema name: {schema_name}")
        if table_name is not None:
            table_name = self.set_case("DB2_LUW", table_name)
            LogMsg(INFO, f"Table name: {table_name}")

        try:
            stmt = ibm_db.columns(self.conn_handler, None, schema_name, table_name)
            LogMsg(DEBUG, f"Statement to retrieving the columns: {stmt}")
            row = ibm_db.fetch_assoc(stmt)
            i = 0
            while (row):
                result.append(row)
                i += 1
                row = ibm_db.fetch_assoc(stmt)
            ibm_db.free_result(stmt)

            col_names_lower = []
            if column_names is not None:
                for name in column_names:
                    col_names_lower.append(name.lower())
                include_columns = []
                if column_names and column_names != '':
                    for column in result:
                        if column['COLUMN_NAME'].lower() in col_names_lower:
                            column['COLUMN_NAME'] = column['COLUMN_NAME'].lower()
                            include_columns.append(column)
                    result = include_columns
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred while retrieving the columns: {inst}")
            raise _get_exception(inst)

        LogMsg(INFO, "exit columns()")
        return result

    def __enter__(self):
        return self

    def __exit__(self, exc_type, exc_val, exc_tb):
        return self.close()


# Defines a cursor for the driver connection
class Cursor(object):
    """This class represents a cursor of the connection.  It can be
    used to process an SQL statement.
    """

    # This method is used to get the description attribute.
    def __get_description(self):
        """ If this method has already been called, after executing a select statement,
            return the stored information in the self.__description.
        """
        LogMsg(INFO, "entry __get_description()")
        if self.__description is not None:
            return self.__description

        if self.stmt_handler is None:
            return None
        self.__description = []

        try:
            num_columns = ibm_db.num_fields(self.stmt_handler)
            LogMsg(INFO, "Fetching column descriptions...")
            LogMsg(DEBUG, f"Number of columns: {num_columns}")

            """ If the execute statement did not produce a result set return None.
            """
            if num_columns == False:
                self.__description = None
                return None
            for column_index in range(num_columns):
                column_desc = []
                column_desc.append(ibm_db.field_name(self.stmt_handler,
                                                     column_index))
                type = ibm_db.field_type(self.stmt_handler, column_index)
                type = type.upper()
                LogMsg(INFO, f"Processing column")
                LogMsg(DEBUG, f"Column type: {type}")
                if STRING == type:
                    column_desc.append(STRING)
                elif TEXT == type:
                    column_desc.append(TEXT)
                elif XML == type:
                    column_desc.append(XML)
                elif BINARY == type:
                    column_desc.append(BINARY)
                elif NUMBER == type:
                    column_desc.append(NUMBER)
                elif BIGINT == type:
                    column_desc.append(BIGINT)
                elif FLOAT == type:
                    column_desc.append(FLOAT)
                elif DECIMAL == type:
                    column_desc.append(DECIMAL)
                elif DATE == type:
                    column_desc.append(DATE)
                elif TIME == type:
                    column_desc.append(TIME)
                elif DATETIME == type:
                    column_desc.append(DATETIME)
                elif ROWID == type:
                    column_desc.append(ROWID)
                elif BOOLEAN == type:
                    column_desc.append(BOOLEAN)

                column_desc.append(ibm_db.field_display_size(
                    self.stmt_handler, column_index))

                column_desc.append(ibm_db.field_display_size(
                    self.stmt_handler, column_index))

                column_desc.append(ibm_db.field_precision(
                    self.stmt_handler, column_index))

                column_desc.append(ibm_db.field_scale(self.stmt_handler,
                                                      column_index))

                column_desc.append(ibm_db.field_nullable(
                    self.stmt_handler, column_index))

                self.__description.append(column_desc)
        except Exception as inst:
            LogMsg(EXCEPTION, f"An exception occurred: {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]

        LogMsg(INFO, "exit __get_description()")
        return self.__description

    # This attribute provides the metadata information of the columns
    # in the result set produced by the last execute function.  It is
    # a read only attribute.
    description = property(fget=__get_description)

    # This method is used to get the rowcount attribute.
    def __get_rowcount(self):
        return self.__rowcount

    def __iter__(self):
        return self

    def __next__(self):
        row = self.fetchone()
        if row == None:
            raise StopIteration
        return row

    if PY2:
        next = __next__

    # This attribute specifies the number of rows the last executeXXX()
    # produced or affected.  It is a read only attribute.
    rowcount = property(__get_rowcount, None, None, "")

    # This method is used to get the Connection object
    def __get_connection(self):
        return self.__connection

    # This attribute specifies the connection object.
    # It is a read only attribute.
    connection = property(__get_connection, None, None, "")

    def __init__(self, conn_handler, conn_object=None):
        """Constructor for Cursor object. It takes ibm_db connection
        handler as an argument.
        """

        # This attribute is used to determine the fetch size for fetchmany
        # operation. It is a read/write attribute
        self.arraysize = 1
        self.__rowcount = -1
        self._result_set_produced = False
        self.__description = None
        self.conn_handler = conn_handler
        self.stmt_handler = None
        self._is_scrollable_cursor = False
        self.__connection = conn_object
        self.messages = []
        self.FIX_RETURN_TYPE = conn_object.FIX_RETURN_TYPE
        LogMsg(INFO, "Cursor object initialized.")

    # This method closes the statemente associated with the cursor object.
    # It takes no argument.
    def close(self):
        """This method closes the cursor object.  After this method is
        called the cursor object is no longer usable.  It takes no
        arguments.

        """
        LogMsg(INFO, "entry close()")
        messages = []
        if self.conn_handler is None:
            '''
            Changes for django
            '''
            #self.messages.append(ProgrammingError("Cursor cannot be closed; connection is no longer active."))
            #raise self.messages[len(self.messages) - 1]
            LogMsg(WARNING, "Cursor cannot be closed; connection is no longer active.")
            return None
        try:
            return_value = ibm_db.free_stmt(self.stmt_handler)
        except Exception as inst:
            LogMsg(ERROR, f"Error occurred while closing cursor: {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]
        self.stmt_handler = None
        self.conn_handler = None
        self._all_stmt_handlers = None
        if self.__connection is not None:
            try:
                self.__connection._cursor_list.remove(weakref.ref(self))
            except:
                pass
        LogMsg(INFO, "Cursor closed successfully.")
        LogMsg(INFO, "exit close()")
        return return_value

    # helper for calling procedure
    def _callproc_helper(self, procname, parameters=None):
        LogMsg(INFO, "entry _callproc_helper()")
        LogMsg(DEBUG, f"Calling procedure: {procname}")
        if parameters is not None:
            buff = []
            CONVERT_STR = (buffer)
            # Convert date/time and binary objects to string for
            # inserting into the database.
            for param in parameters:
                if isinstance(param, CONVERT_STR):
                    param = str(param)
                buff.append(param)
            parameters = tuple(buff)
            LogMsg(DEBUG, f"Procedure parameters: {parameters}")

            try:
                result = ibm_db.callproc(self.conn_handler, procname, parameters)
            except Exception as inst:
                LogMsg(ERROR, f"Error procedure '{procname}': {_get_exception(inst)}")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
        else:
            LogMsg(DEBUG, "Calling procedure without parameters.")
            try:
                result = ibm_db.callproc(self.conn_handler, procname)
            except Exception as inst:
                LogMsg(ERROR, f"Error calling procedure '{procname}': {_get_exception(inst)}")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
        LogMsg(INFO, f"Procedure '{procname}' called successfully.")
        LogMsg(INFO, "exit _callproc_helper()")
        return result

    def callproc(self, procname, parameters=None):
        """This method can be used to execute a stored procedure.
        It takes the name of the stored procedure and the parameters to
        the stored procedure as arguments.

        """
        LogMsg(INFO, "entry callproc()")
        LogMsg(DEBUG, f"Calling callproc with procname={procname}, parameters={parameters}")
        self.messages = []
        if not isinstance(procname, string_types):
            LogMsg(ERROR, "callproc expects the first argument to be of type String or Unicode.")
            self.messages.append(InterfaceError("callproc expects the first argument to be of type String or Unicode."))
            raise self.messages[len(self.messages) - 1]
        if parameters is not None:
            if not isinstance(parameters, (list, tuple)):
                LogMsg(ERROR, "callproc expects the second argument to be of type list or tuple.")
                self.messages.append(
                    InterfaceError("callproc expects the second argument to be of type list or tuple."))
                raise self.messages[len(self.messages) - 1]
        result = self._callproc_helper(procname, parameters)
        LogMsg(DEBUG, f"Result received from callproc helper: {result}")
        return_value = None
        self.__description = None
        self._all_stmt_handlers = []
        if isinstance(result, tuple):
            self.stmt_handler = result[0]
            return_value = result[1:]
        else:
            self.stmt_handler = result
        self._result_set_produced = True
        LogMsg(DEBUG, "callproc executed successfully.")
        LogMsg(INFO, "exit callproc()")
        return return_value

    # Helper for preparing an SQL statement.
    def _prepare_helper(self, operation, parameters=None):
        LogMsg(INFO, "entry _prepare_helper()")
        try:
            ibm_db.free_stmt(self.stmt_handler)
            LogMsg(DEBUG, "Successfully freed existing statement handler.")
        except:
            pass

        try:
            self.stmt_handler = ibm_db.prepare(self.conn_handler, operation)
            LogMsg(DEBUG, f"Successfully prepared statement with operation: {operation}")
            LogMsg(INFO, "exit _prepare_helper()")
        except Exception as inst:
            LogMsg(ERROR, f"Error preparing statement with operation '{operation}': {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]

    # Helper for preparing an SQL statement.
    def _set_cursor_helper(self):
        LogMsg(INFO, "entry _set_cursor_helper()")
        if (ibm_db.get_option(self.stmt_handler, ibm_db.SQL_ATTR_CURSOR_TYPE, 0) != ibm_db.SQL_CURSOR_FORWARD_ONLY):
            self._is_scrollable_cursor = True
            LogMsg(INFO, "Cursor type is scrollable.")
        else:
            self._is_scrollable_cursor = False
            LogMsg(INFO, "Cursor type is forward-only.")
        self._result_set_produced = False
        try:
            num_columns = ibm_db.num_fields(self.stmt_handler)
            LogMsg(DEBUG, f"Number of columns retrieved: {num_columns}")
        except Exception as inst:
            LogMsg(ERROR, f"Error getting number of columns: {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]
        if not num_columns:
            LogMsg(INFO, "exit _set_cursor_helper()")
            return False
        self._result_set_produced = True

        LogMsg(INFO, "exit _set_cursor_helper()")
        return True

    # Helper for executing an SQL statement.
    def _execute_helper(self, parameters=None):
        LogMsg(INFO, "entry  _execute_helper()")
        if parameters is not None:
            buff = []
            CONVERT_STR = (buffer)
            # Convert date/time and binary objects to string for
            # inserting into the database.
            for param in parameters:
                if isinstance(param, memoryview):
                    param = param.tobytes()
                elif isinstance(param, CONVERT_STR):
                    param = str(param)
                buff.append(param)
            parameters = tuple(buff)
            LogMsg(DEBUG, f"Executing statement with parameters: {parameters}")
            try:
                return_value = ibm_db.execute(self.stmt_handler, parameters)
                if not return_value:
                    if ibm_db.conn_errormsg() is not None:
                        error_msg = f"Connection error: {str(ibm_db.conn_errormsg())}"
                        LogMsg(ERROR, error_msg)
                        self.messages.append(Error(str(ibm_db.conn_errormsg())))
                        raise self.messages[len(self.messages) - 1]
                    if ibm_db.stmt_errormsg() is not None:
                        error_msg = f"Statement error: {str(ibm_db.conn_errormsg())}"
                        LogMsg(ERROR, error_msg)
                        self.messages.append(Error(str(ibm_db.stmt_errormsg())))
                        raise self.messages[len(self.messages) - 1]
            except Exception as inst:
                LogMsg(ERROR, f"Error executing statement with parameters: {_get_exception(inst)} ")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
        else:
            LogMsg(DEBUG, f"Executing statement without parameters")
            try:
                return_value = ibm_db.execute(self.stmt_handler)
                if not return_value:
                    if ibm_db.conn_errormsg() is not None:
                        error_msg = f"Connection error: {str(ibm_db.conn_errormsg())}"
                        LogMsg(ERROR, error_msg)
                        self.messages.append(Error(str(ibm_db.conn_errormsg())))
                        raise self.messages[len(self.messages) - 1]
                    if ibm_db.stmt_errormsg() is not None:
                        error_msg = f"Statement error: {str(ibm_db.conn_errormsg())}"
                        LogMsg(ERROR, error_msg)
                        self.messages.append(Error(str(ibm_db.stmt_errormsg())))
                        raise self.messages[len(self.messages) - 1]
            except Exception as inst:
                LogMsg(ERROR, f"Error executing statement without parameters: {_get_exception(inst)} ")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
        LogMsg(INFO, "exit  _execute_helper()")
        return return_value

    # This method is used to set the rowcount after executing an SQL
    # statement.
    def _set_rowcount(self):
        LogMsg(INFO, "entry  _set_rowcount()")
        self.__rowcount = -1
        if not self._result_set_produced:
            try:
                counter = ibm_db.num_rows(self.stmt_handler)
                LogMsg(DEBUG, f"Number of rows retrieved: {counter}")
            except Exception as inst:
                LogMsg(ERROR, f"Error getting row count: {_get_exception(inst)}")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
            self.__rowcount = counter
        elif self._is_scrollable_cursor:
            try:
                counter = ibm_db.get_num_result(self.stmt_handler)
                LogMsg(DEBUG, f"Number of rows retrieved: {counter}")
            except Exception as inst:
                LogMsg(ERROR, f"Error getting row count: {_get_exception(inst)}")
                self.messages.append(_get_exception(inst))
                raise self.messages[len(self.messages) - 1]
            if counter >= 0:
                self.__rowcount = counter
        LogMsg(INFO, "exit  _set_rowcount()")
        return True

    # Retrieves the last generated identity value from the DB2 catalog
    def _get_last_identity_val(self):
        """
        The result of the IDENTITY_VAL_LOCAL function is not affected by the following:
         - A single row INSERT statement with a VALUES clause for a table without an
        identity column
         - A multiple row INSERT statement with a VALUES clause
         - An INSERT statement with a fullselect

        """
        LogMsg(INFO, "entry  _get_last_identity_val()")
        operation = 'SELECT IDENTITY_VAL_LOCAL() FROM SYSIBM.SYSDUMMY1'
        try:
            stmt_handler = ibm_db.prepare(self.conn_handler, operation)
            LogMsg(DEBUG, f"Preparing statement with operation: {operation}")
            if ibm_db.execute(stmt_handler):
                row = ibm_db.fetch_tuple(stmt_handler)
                if row[0] is not None:
                    identity_val = int(row[0])
                    LogMsg(DEBUG, f"Identity value retrieved: {identity_val}")
                else:
                    identity_val = None
                    LogMsg(DEBUG, "Identity value is None")
            else:
                if ibm_db.conn_errormsg() is not None:
                    error_msg = f"Connection error: {str(ibm_db.conn_errormsg())}"
                    LogMsg(ERROR, error_msg)
                    self.messages.append(Error(str(ibm_db.conn_errormsg())))
                    raise self.messages[len(self.messages) - 1]
                if ibm_db.stmt_errormsg() is not None:
                    error_msg = f"Statement error: {str(ibm_db.stmt_errormsg())}"
                    LogMsg(ERROR, error_msg)
                    self.messages.append(Error(str(ibm_db.stmt_errormsg())))
                    raise self.messages[len(self.messages) - 1]
        except Exception as inst:
            LogMsg(ERROR, f"Error occured in getting identity value: {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]
        LogMsg(INFO, "exit  _get_last_identity_val()")
        return identity_val

    last_identity_val = property(_get_last_identity_val, None, None, "")

    def execute(self, operation, parameters=None):
        """
        This method can be used to prepare and execute an SQL
        statement.  It takes the SQL statement(operation) and a
        sequence of values to substitute for the parameter markers in
        the SQL statement as arguments.
        """
        LogMsg(INFO, "entry execute()")
        LogMsg(DEBUG, f"Executing SQL operation: {operation}")
        if parameters is not None:
            LogMsg(DEBUG, f"SQL parameters: {parameters}")
        self.messages = []
        if not isinstance(operation, string_types):
            err_msg = "execute expects the first argument [%s] to be of type String or Unicode." % operation
            LogMsg(ERROR, err_msg)
            self.messages.append(
                InterfaceError("execute expects the first argument [%s] to be of type String or Unicode." % operation))
            raise self.messages[len(self.messages) - 1]
        if parameters is not None:
            if not isinstance(parameters, (list, tuple, dict)):
                LogMsg(ERROR, "execute parameters argument should be sequence.")
                self.messages.append(InterfaceError("execute parameters argument should be sequence."))
                raise self.messages[len(self.messages) - 1]
        self.__description = None
        self._all_stmt_handlers = []
        self._prepare_helper(operation)
        self._execute_helper(parameters)
        self._set_cursor_helper()
        LogMsg(INFO, "SQL operation executed successfully.")
        LogMsg(INFO, "exit execute()")
        return self._set_rowcount()

    def executemany(self, operation, seq_parameters):
        """
        This method can be used to prepare, and then execute an SQL
        statement many times.  It takes the SQL statement(operation)
        and sequence of sequence of values to substitute for the
        parameter markers in the SQL statement as its argument.
        """
        LogMsg(INFO, "entry executemany()")
        LogMsg(DEBUG, f"Executing SQL operation in executemany: {operation}")
        LogMsg(DEBUG, f"Number of parameter sets: {len(seq_parameters)}")
        self.messages = []
        if not isinstance(operation, string_types):
            LogMsg(ERROR, f"executemany expects the first argument to be of type String or Unicode.")
            self.messages.append(
                InterfaceError("executemany expects the first argument to be of type String or Unicode."))
            raise self.messages[len(self.messages) - 1]
        if seq_parameters is None:
            LogMsg(ERROR, f"executemany expects a not None seq_parameters value")
            self.messages.append(InterfaceError("executemany expects a not None seq_parameters value"))
            raise self.messages[len(self.messages) - 1]

        if not isinstance(seq_parameters, (list, tuple)):
            LogMsg(ERROR, f"executemany expects the second argument to be of type list or tuple of sequence.")
            self.messages.append(
                InterfaceError("executemany expects the second argument to be of type list or tuple of sequence."))
            raise self.messages[len(self.messages) - 1]

        CONVERT_STR = (buffer)
        # Convert date/time and binary objects to string for
        # inserting into the database.
        buff = []
        seq_buff = []
        for index in range(len(seq_parameters)):
            buff = []
            for param in seq_parameters[index]:
                if isinstance(param, CONVERT_STR):
                    param = str(param)
                buff.append(param)
            seq_buff.append(tuple(buff))
        seq_parameters = tuple(seq_buff)
        self.__description = None
        self._all_stmt_handlers = []
        self.__rowcount = -1
        self._prepare_helper(operation)
        try:
            autocommit = ibm_db.autocommit(self.conn_handler)
            if autocommit != 0:
                ibm_db.autocommit(self.conn_handler, 0)
            self.__rowcount = ibm_db.execute_many(self.stmt_handler, seq_parameters)
            if autocommit != 0:
                ibm_db.commit(self.conn_handler)
                ibm_db.autocommit(self.conn_handler, autocommit)
            if self.__rowcount == -1:
                if ibm_db.conn_errormsg() is not None:
                    error_msg = f"Connection error: {str(ibm_db.conn_errormsg())}"
                    LogMsg(ERROR, error_msg)
                    self.messages.append(Error(str(ibm_db.conn_errormsg())))
                    raise self.messages[len(self.messages) - 1]
                if ibm_db.stmt_errormsg() is not None:
                    error_msg = f"Statement error: {str(ibm_db.stmt_errormsg())}"
                    LogMsg(ERROR, error_msg)
                    self.messages.append(Error(str(ibm_db.stmt_errormsg())))
                    raise self.messages[len(self.messages) - 1]
        except Exception as inst:
            self._set_rowcount()
            self.messages.append(Error(inst))
            LogMsg(ERROR, f"Error in executemany: {inst}")
            raise self.messages[len(self.messages) - 1]
        LogMsg(INFO, "SQL operation executemany successfully.")
        LogMsg(INFO, "exit executemany()")
        return True

    def _fetch_helper(self, fetch_size=-1):
        """
        This method is a helper function for fetching fetch_size number of
        rows, after executing an SQL statement which produces a result set.
        It takes the number of rows to fetch as an argument.
        If this is not provided it fetches all the remaining rows.
        """
        LogMsg(INFO, "entry _fetch_helper()")
        if self.stmt_handler is None:
            LogMsg(ERROR, "Please execute an SQL statement in order to get a row from result set.")
            self.messages.append(
                ProgrammingError("Please execute an SQL statement in order to get a row from result set."))
            raise self.messages[len(self.messages) - 1]
        if not self._result_set_produced:
            LogMsg(ERROR, "The last call to execute did not produce any result set.")
            self.messages.append(ProgrammingError("The last call to execute did not produce any result set."))
            raise self.messages[len(self.messages) - 1]
        row_list = []
        rows_fetched = 0
        while (fetch_size == -1) or \
                (fetch_size != -1 and rows_fetched < fetch_size):
            try:
                row = ibm_db.fetch_tuple(self.stmt_handler)
            except Exception as inst:
                if ibm_db.stmt_errormsg() is not None:
                    error_msg = f"Statement error: {str(ibm_db.stmt_errormsg())}"
                    LogMsg(ERROR, error_msg)
                    self.messages.append(Error(str(ibm_db.stmt_errormsg())))
                    raise self.messages[len(self.messages) - 1]
                else:
                    LogMsg(ERROR, f"Error occured : {_get_exception(inst)}")
                    self.messages.append(_get_exception(inst))
                    raise self.messages[len(self.messages) - 1]

            if row != False:
                if self.FIX_RETURN_TYPE == 1:
                    row_list.append(self._fix_return_data_type(row))
                else:
                    row_list.append(row)
            else:
                LogMsg(DEBUG, f"Returning {row_list} from _fetch_helper()")
                LogMsg(INFO, "exit _fetch_helper()")
                return row_list
            rows_fetched = rows_fetched + 1
        LogMsg(DEBUG, f"Returning {row_list} from _fetch_helper()")
        LogMsg(INFO, "exit _fetch_helper()")
        return row_list

    def fetchone(self):
        """This method fetches one row from the database, after
        executing an SQL statement which produces a result set.

        """
        LogMsg(INFO, "entry fetchone()")
        LogMsg(INFO, "Fetching one row from the database.")
        row_list = self._fetch_helper(1)
        if len(row_list) == 0:
            LogMsg(DEBUG, "No rows fetched.")
        else:
            LogMsg(DEBUG, "Row fetched successfully.")
        if len(row_list) == 0:
            LogMsg(INFO, "exit fetchone()")
            return None
        else:
            LogMsg(INFO, "exit fetchone()")
            return row_list[0]

    def fetchmany(self, size=0):
        """This method fetches size number of rows from the database,
        after executing an SQL statement which produces a result set.
        It takes the number of rows to fetch as an argument.  If this
        is not provided it fetches self.arraysize number of rows.
        """
        LogMsg(INFO, "entry fetchmany()")
        message = f"Fetching {size} rows from the database."
        LogMsg(DEBUG, message)
        if not isinstance(size, int_types):
            LogMsg(EXCEPTION, "fetchmany expects argument type int or long.")
            self.messages.append(InterfaceError("fetchmany expects argument type int or long."))
            raise self.messages[len(self.messages) - 1]
        if size == 0:
            size = self.arraysize
        if size < -1:
            LogMsg(ERROR, "fetchmany argument size expected to be positive")
            self.messages.append(ProgrammingError("fetchmany argument size expected to be positive."))
            raise self.messages[len(self.messages) - 1]

        fetch_nrows = self._fetch_helper(size)
        nrows = len(fetch_nrows)
        message = f"Fetched {nrows} rows successfully."
        LogMsg(DEBUG, message)
        LogMsg(INFO, "exit fetchmany()")
        return fetch_nrows

    def fetchall(self):
        """This method fetches all remaining rows from the database,
        after executing an SQL statement which produces a result set.
        """
        LogMsg(INFO, "entry fetchall()")
        LogMsg(INFO, "Fetching all remaining rows from the database.")
        rows_fetched = self._fetch_helper()
        nrows = len(rows_fetched)
        LogMsg(DEBUG, f"Fetched {nrows} rows successfully.")
        LogMsg(INFO, "exit fetchall()")
        return rows_fetched

    def nextset(self):
        """This method can be used to get the next result set after
        executing a stored procedure, which produces multiple result sets.
        """
        LogMsg(INFO, "entry nextset()")
        LogMsg(INFO, "Attempting to retrieve next result set.")
        self.messages = []
        if self.stmt_handler is None:
            LogMsg(ERROR, "Please execute an SQL statement in order to get result sets.")
            self.messages.append(ProgrammingError("Please execute an SQL statement in order to get result sets."))
            raise self.messages[len(self.messages) - 1]
        if not self._result_set_produced:
            LogMsg(ERROR, "The last call to execute did not produce any result set.")
            self.messages.append(ProgrammingError("The last call to execute did not produce any result set."))
            raise self.messages[len(self.messages) - 1]
        try:
            # Store all the stmt handler that were created.  The
            # handler was the one created by the execute method.  It
            # should be used to get next result set.
            self.__description = None
            self._all_stmt_handlers.append(self.stmt_handler)
            self.stmt_handler = ibm_db.next_result(self._all_stmt_handlers[0])
        except Exception as inst:
            LogMsg(ERROR, f"Error while retrieving next result set: {_get_exception(inst)}")
            self.messages.append(_get_exception(inst))
            raise self.messages[len(self.messages) - 1]

        if not self.stmt_handler:
            self.stmt_handler = None
        if self.stmt_handler is None:
            return None

        LogMsg(INFO, "Successfully retrieved next result set.")
        LogMsg(INFO, "exit nextset()")
        return True

    def setinputsizes(self, sizes):
        """This method currently does nothing."""
        pass

    def setoutputsize(self, size, column=-1):
        """This method currently does nothing."""
        pass

    # This method is used to convert a string representing decimal
    # and binary data in a row tuple fetched from the database
    # to decimal and binary objects, for returning it to the user.
    def _fix_return_data_type(self, row):
        LogMsg(INFO, "entry _fix_return_data_type()")
        message = f"Fixing return data types for row: {row}"
        LogMsg(DEBUG, message)
        row_list = None
        for index in range(len(row)):
            if row[index] is not None:
                type = ibm_db.field_type(self.stmt_handler, index)
                type = type.upper()

                try:
                    if type == 'BLOB':
                        if row_list is None:
                            row_list = list(row)
                        row_list[index] = buffer(row[index])

                    elif type == 'DECIMAL':
                        if row_list is None:
                            row_list = list(row)
                        row_list[index] = decimal.Decimal(str(row[index]).replace(",", "."))

                except Exception as inst:
                    LogMsg(ERROR, f"Data type format error: {str(inst)}")
                    self.messages.append(DataError("Data type format error: " + str(inst)))
                    raise self.messages[len(self.messages) - 1]
        if row_list is None:
            LogMsg(INFO, "exit _fix_return_data_type()")
            return row
        else:
            LogMsg(DEBUG, f"Fixed return data types: {row_list}")
            LogMsg(INFO, "exit _fix_return_data_type()")
            return tuple(row_list)

    def __enter__(self):
        return self

    def __exit__(self, exc_type, exc_val, exc_tb):
        return self.close()