Module functions and constants

The pg module defines a few functions that allow to connect to a database and to define “default variables” that override the environment variables used by PostgreSQL.

These “default variables” were designed to allow you to handle general connection parameters without heavy code in your programs. You can prompt the user for a value, put it in the default variable, and forget it, without having to modify your environment. The support for default variables can be disabled by setting the -DNO_DEF_VAR option in the Python setup file. Methods relative to this are specified by the tag [DV].

All variables are set to None at module initialization, specifying that standard environment variables should be used.

connect – Open a PostgreSQL connection

pg.connect([dbname][, host][, port][, opt][, user][, passwd])

Open a pg connection

Parameters:
  • dbname – name of connected database (None = defbase)
  • host (str or None) – name of the server host (None = defhost)
  • port (int) – port used by the database server (-1 = defport)
  • opt (str or None) – connection options (None = defopt)
  • user (str or None) – PostgreSQL user (None = defuser)
  • passwd (str or None) – password for user (None = defpasswd)
Returns:

If successful, the Connection handling the connection

Return type:

Connection

Raises:
  • TypeError – bad argument type, or too many arguments
  • SyntaxError – duplicate argument definition
  • pg.InternalError – some error occurred during pg connection definition
  • Exception – (all exceptions relative to object allocation)

This function opens a connection to a specified database on a given PostgreSQL server. You can use keywords here, as described in the Python tutorial. The names of the keywords are the name of the parameters given in the syntax line. For a precise description of the parameters, please refer to the PostgreSQL user manual.

Example:

import pg

con1 = pg.connect('testdb', 'myhost', 5432, None, None, 'bob', None)
con2 = pg.connect(dbname='testdb', host='localhost', user='bob')

get/set_defhost – default server host [DV]

pg.get_defhost(host)

Get the default host

Returns:the current default host specification
Return type:str or None
Raises:TypeError – too many arguments

This method returns the current default host specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defhost(host)

Set the default host

Parameters:host (str or None) – the new default host specification
Returns:the previous default host specification
Return type:str or None
Raises:TypeError – bad argument type, or too many arguments

This methods sets the default host value for new connections. If None is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default host.

get/set_defport – default server port [DV]

pg.get_defport()

Get the default port

Returns:the current default port specification
Return type:int
Raises:TypeError – too many arguments

This method returns the current default port specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defport(port)

Set the default port

Parameters:port (int) – the new default port
Returns:previous default port specification
Return type:int or None

This methods sets the default port value for new connections. If -1 is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default port.

get/set_defopt – default connection options [DV]

pg.get_defopt()

Get the default connection options

Returns:the current default options specification
Return type:str or None
Raises:TypeError – too many arguments

This method returns the current default connection options specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defopt(options)

Set the default connection options

Parameters:options (str or None) – the new default connection options
Returns:previous default options specification
Return type:str or None
Raises:TypeError – bad argument type, or too many arguments

This methods sets the default connection options value for new connections. If None is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default options.

get/set_defbase – default database name [DV]

pg.get_defbase()

Get the default database name

Returns:the current default database name specification
Return type:str or None
Raises:TypeError – too many arguments

This method returns the current default database name specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defbase(base)

Set the default database name

Parameters:base (str or None) – the new default base name
Returns:the previous default database name specification
Return type:str or None
Raises:TypeError – bad argument type, or too many arguments

This method sets the default database name value for new connections. If None is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default host.

get/set_defuser – default database user [DV]

pg.get_defuser()

Get the default database user

Returns:the current default database user specification
Return type:str or None
Raises:TypeError – too many arguments

This method returns the current default database user specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defuser(user)

Set the default database user

Parameters:user – the new default database user
Returns:the previous default database user specification
Return type:str or None
Raises:TypeError – bad argument type, or too many arguments

This method sets the default database user name for new connections. If None is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default host.

get/set_defpasswd – default database password [DV]

pg.get_defpasswd()

Get the default database password

Returns:the current default database password specification
Return type:str or None
Raises:TypeError – too many arguments

This method returns the current default database password specification, or None if the environment variables should be used. Environment variables won’t be looked up.

pg.set_defpasswd(passwd)

Set the default database password

Parameters:passwd – the new default database password
Returns:the previous default database password specification
Return type:str or None
Raises:TypeError – bad argument type, or too many arguments

This method sets the default database password for new connections. If None is supplied as parameter, environment variables will be used in future connections. It returns the previous setting for default host.

escape_string – escape a string for use within SQL

pg.escape_string(string)

Escape a string for use within SQL

Parameters:string (str) – the string that is to be escaped
Returns:the escaped string
Return type:str
Raises:TypeError – bad argument type, or too many arguments

This function escapes a string for use within an SQL command. This is useful when inserting data values as literal constants in SQL commands. Certain characters (such as quotes and backslashes) must be escaped to prevent them from being interpreted specially by the SQL parser. escape_string() performs this operation. Note that there is also a Connection method with the same name which takes connection properties into account.

Note

It is especially important to do proper escaping when handling strings that were received from an untrustworthy source. Otherwise there is a security risk: you are vulnerable to “SQL injection” attacks wherein unwanted SQL commands are fed to your database.

Example:

name = input("Name? ")
phone = con.query("select phone from employees where name='%s'"
    % escape_string(name)).getresult()

escape_bytea – escape binary data for use within SQL

pg.escape_bytea(datastring)

escape binary data for use within SQL as type bytea

Parameters:datastring (str) – string containing the binary data that is to be escaped
Returns:the escaped string
Return type:str
Raises:TypeError – bad argument type, or too many arguments

Escapes binary data for use within an SQL command with the type bytea. As with escape_string(), this is only used when inserting data directly into an SQL command string.

Note that there is also a Connection method with the same name which takes connection properties into account.

Example:

picture = open('garfield.gif', 'rb').read()
con.query("update pictures set img='%s' where name='Garfield'"
    % escape_bytea(picture))

unescape_bytea – unescape data that has been retrieved as text

pg.unescape_bytea(string)

Unescape bytea data that has been retrieved as text

Parameters:datastring (str) – the bytea data string that has been retrieved as text
Returns:byte string containing the binary data
Return type:bytes
Raises:TypeError – bad argument type, or too many arguments

Converts an escaped string representation of binary data stored as bytea into the raw byte string representing the binary data – this is the reverse of escape_bytea(). Since the Query results will already return unescaped byte strings, you normally don’t have to use this method.

Note that there is also a DB method with the same name which does exactly the same.

get/set_namedresult – conversion to named tuples

pg.get_namedresult()

Get the function that converts to named tuples

This returns the function used by PyGreSQL to construct the result of the Query.namedresult() method.

New in version 4.1.

pg.set_namedresult(func)

Set a function that will convert to named tuples

Parameters:func – the function to be used to convert results to named tuples

You can use this if you want to create different kinds of named tuples returned by the Query.namedresult() method. If you set this function to None, then it will become equal to Query.getresult().

New in version 4.1.

get/set_decimal – decimal type to be used for numeric values

pg.get_decimal()

Get the decimal type to be used for numeric values

Returns:the Python class used for PostgreSQL numeric values
Return type:class

This function returns the Python class that is used by PyGreSQL to hold PostgreSQL numeric values. The default class is decimal.Decimal if available, otherwise the float type is used.

pg.set_decimal(cls)

Set a decimal type to be used for numeric values

Parameters:cls (class) – the Python class to be used for PostgreSQL numeric values

This function can be used to specify the Python class that shall be used by PyGreSQL to hold PostgreSQL numeric values. The default class is decimal.Decimal if available, otherwise the float type is used.

get/set_decimal_point – decimal mark used for monetary values

pg.get_decimal_point()

Get the decimal mark used for monetary values

Returns:string with one character representing the decimal mark
Return type:str

This function returns the decimal mark used by PyGreSQL to interpret PostgreSQL monetary values when converting them to decimal numbers. The default setting is '.' as a decimal point. This setting is not adapted automatically to the locale used by PostGreSQL, but you can use set_decimal() to set a different decimal mark manually. A return value of None means monetary values are not interpreted as decimal numbers, but returned as strings including the formatting and currency.

New in version 4.1.1.

pg.set_decimal_point(string)

Specify which decimal mark is used for interpreting monetary values

Parameters:string (str) – string with one character representing the decimal mark

This function can be used to specify the decimal mark used by PyGreSQL to interpret PostgreSQL monetary values. The default value is ‘.’ as a decimal point. This value is not adapted automatically to the locale used by PostGreSQL, so if you are dealing with a database set to a locale that uses a ',' instead of '.' as the decimal point, then you need to call set_decimal(',') to have PyGreSQL interpret monetary values correctly. If you don’t want money values to be converted to decimal numbers, then you can call set_decimal(None), which will cause PyGreSQL to return monetary values as strings including their formatting and currency.

New in version 4.1.1.

get/set_bool – whether boolean values are returned as bool objects

pg.get_bool()

Check whether boolean values are returned as bool objects

Returns:whether or not bool objects will be returned
Return type:bool

This function checks whether PyGreSQL returns PostgreSQL boolean values converted to Python bool objects, or as 'f' and 't' strings which are the values used internally by PostgreSQL. By default, conversion to bool objects is activated, but you can disable this with the set_bool() function.

New in version 4.2.

pg.set_bool(on)

Set whether boolean values are returned as bool objects

Parameters:on – whether or not bool objects shall be returned

This function can be used to specify whether PyGreSQL shall return PostgreSQL boolean values converted to Python bool objects, or as 'f' and 't' strings which are the values used internally by PostgreSQL. By default, conversion to bool objects is activated, but you can disable this by calling set_bool(True).

New in version 4.2.

Changed in version 5.0: Boolean values had been returned as string by default in earlier versions.

get/set_array – whether arrays are returned as list objects

pg.get_array()

Check whether arrays are returned as list objects

Returns:whether or not list objects will be returned
Return type:bool

This function checks whether PyGreSQL returns PostgreSQL arrays converted to Python list objects, or simply as text in the internal special output syntax of PostgreSQL. By default, conversion to list objects is activated, but you can disable this with the set_array() function.

New in version 5.0.

pg.set_array(on)

Set whether arrays are returned as list objects

Parameters:on – whether or not list objects shall be returned

This function can be used to specify whether PyGreSQL shall return PostgreSQL arrays converted to Python list objects, or simply as text in the internal special output syntax of PostgreSQL. By default, conversion to list objects is activated, but you can disable this by calling set_array(False).

New in version 5.0.

Changed in version 5.0: Arrays had been always returned as text strings only in earlier versions.

get/set_bytea_escaped – whether bytea data is returned escaped

pg.get_bytea_escaped()

Check whether bytea values are returned as escaped strings

Returns:whether or not bytea objects will be returned escaped
Return type:bool

This function checks whether PyGreSQL returns PostgreSQL bytea values in escaped form or in unescaped from as byte strings. By default, bytea values will be returned unescaped as byte strings, but you can change this with the set_bytea_escaped() function.

New in version 5.0.

pg.set_bytea_escaped(on)

Set whether bytea values are returned as escaped strings

Parameters:on – whether or not bytea objects shall be returned escaped

This function can be used to specify whether PyGreSQL shall return PostgreSQL bytea values in escaped form or in unescaped from as byte strings. By default, bytea values will be returned unescaped as byte strings, but you can change this by calling set_bytea_escaped(True).

New in version 5.0.

Changed in version 5.0: Bytea data had been returned in escaped form by default in earlier versions.

get/set_jsondecode – decoding JSON format

pg.get_jsondecode()

Get the function that deserializes JSON formatted strings

This returns the function used by PyGreSQL to construct Python objects from JSON formatted strings.

pg.set_jsondecode(func)

Set a function that will deserialize JSON formatted strings

Parameters:func – the function to be used for deserializing JSON strings

You can use this if you do not want to deserialize JSON strings coming in from the database, or if want to use a different function than the standard function json.loads() or if you want to use it with parameters different from the default ones. If you set this function to None, then the automatic deserialization of JSON strings will be deactivated.

New in version 5.0.

Changed in version 5.0: JSON data had been always returned as text strings in earlier versions.

get/set_cast_hook – fallback typecast function

pg.get_cast_hook()

Get the function that handles all external typecasting

This returns the callback function used by PyGreSQL to provide plug-in Python typecast functions.

pg.set_cast_hook(func)

Set a function that will handle all external typecasting

Parameters:func – the function to be used as a callback

If you set this function to None, then only the typecast functions implemented in the C extension module are enabled. You normally would not want to change this. Instead, you can use get_typecast() and set_typecast() to add or change the plug-in Python typecast functions.

New in version 5.0.

get/set_datestyle – assume a fixed date style

pg.get_datestyle()

Get the assumed date style for typecasting

This returns the PostgreSQL date style that is silently assumed when typecasting dates or None if no fixed date style is assumed, in which case the date style is requested from the database when necessary (this is the default). Note that this method will not get the date style that is currently set in the session or in the database. You can get the current setting with the methods DB.get_parameter() and Connection.parameter(). You can also get the date format corresponding to the current date style by calling Connection.date_format().

New in version 5.0.

pg.set_datestyle(datestyle)

Set a fixed date style that shall be assumed when typecasting

Parameters:datestyle (str) – the date style that shall be assumed, or None if no fixed dat style shall be assumed

PyGreSQL is able to automatically pick up the right date style for typecasting date values from the database, even if you change it for the current session with a SET DateStyle command. This is happens very effectively without an additional database request being involved. If you still want to have PyGreSQL always assume a fixed date style instead, then you can set one with this function. Note that calling this function will not alter the date style of the database or the current session. You can do that by calling the method DB.set_parameter() instead.

New in version 5.0.

get/set_typecast – custom typecasting

PyGreSQL uses typecast functions to cast the raw data coming from the database to Python objects suitable for the particular database type. These functions take a single string argument that represents the data to be casted and must return the casted value.

PyGreSQL provides through its C extension module basic typecast functions for the common database types, but if you want to add more typecast functions, you can set these using the following functions.

pg.get_typecast(typ)

Get the global cast function for the given database type

Parameters:typ (str) – PostgreSQL type name
Returns:the typecast function for the specified type
Return type:function or None

New in version 5.0.

pg.set_typecast(typ, cast)

Set a global typecast function for the given database type(s)

Parameters:
  • typ (str or int) – PostgreSQL type name or list of type names
  • cast – the typecast function to be set for the specified type(s)

The typecast function must take one string object as argument and return a Python object into which the PostgreSQL type shall be casted. If the function takes another parameter named connection, then the current database connection will also be passed to the typecast function. This may sometimes be necessary to look up certain database settings.

New in version 5.0.

Note that database connections cache types and their cast functions using connection specific DbTypes objects. You can also get, set and reset typecast functions on the connection level using the methods DbTypes.get_typecast(), DbTypes.set_typecast() and DbTypes.reset_typecast() of the DB.dbtypes object. This will not affect other connections or future connections. In order to be sure a global change is picked up by a running connection, you must reopen it or call DbTypes.reset_typecast() on the DB.dbtypes object.

Also note that the typecasting for all of the basic types happens already in the C extension module. The typecast functions that can be set with the above methods are only called for the types that are not already supported by the C extension module.

cast_array/record – fast parsers for arrays and records

PosgreSQL returns arrays and records (composite types) using a special output syntax with several quirks that cannot easily and quickly be parsed in Python. Therefore the C extension module provides two fast parsers that allow quickly turning these text representations into Python objects: Arrays will be converted to Python lists, and records to Python tuples. These fast parsers are used automatically by PyGreSQL in order to return arrays and records from database queries as lists and tuples, so you normally don’t need to call them directly. You may only need them for typecasting arrays of data types that are not supported by default in PostgreSQL.

pg.cast_array(string[, cast][, delim])

Cast a string representing a PostgreSQL array to a Python list

Parameters:
  • string (str) – the string with the text representation of the array
  • cast (callable or None) – a typecast function for the elements of the array
  • delim – delimiter character between adjacent elements
Returns:

a list representing the PostgreSQL array in Python

Return type:

list

Raises:
  • TypeError – invalid argument types
  • ValueError – error in the syntax of the given array

This function takes a string containing the text representation of a PostgreSQL array (which may look like '{{1,2}{3,4}}' for a two-dimensional array), a typecast function cast that is called for every element, and an optional delimiter character delim (usually a comma), and returns a Python list representing the array (which may be nested like [[1, 2], [3, 4]] in this example). The cast function must take a single argument which will be the text representation of the element and must output the corresponding Python object that shall be put into the list. If you don’t pass a cast function or set it to None, then unprocessed text strings will be returned as elements of the array. If you don’t pass a delimiter character, then a comma will be used by default.

New in version 5.0.

pg.cast_record(string[, cast][, delim])

Cast a string representing a PostgreSQL record to a Python tuple

Parameters:
  • string (str) – the string with the text representation of the record
  • cast (callable, list or tuple of callables, or None) – typecast function(s) for the elements of the record
  • delim – delimiter character between adjacent elements
Returns:

a tuple representing the PostgreSQL record in Python

Return type:

tuple

Raises:
  • TypeError – invalid argument types
  • ValueError – error in the syntax of the given array

This function takes a string containing the text representation of a PostgreSQL record (which may look like '(1,a,2,b)' for a record composed of four fields), a typecast function cast that is called for every element, or a list or tuple of such functions corresponding to the individual fields of the record, and an optional delimiter character delim (usually a comma), and returns a Python tuple representing the record (which may be inhomogeneous like (1, 'a', 2, 'b') in this example). The cast function(s) must take a single argument which will be the text representation of the element and must output the corresponding Python object that shall be put into the tuple. If you don’t pass cast function(s) or pass None instead, then unprocessed text strings will be returned as elements of the tuple. If you don’t pass a delimiter character, then a comma will be used by default.

New in version 5.0.

Note that besides using parentheses instead of braces, there are other subtle differences in escaping special characters and NULL values between the syntax used for arrays and the one used for composite types, which these functions take into account.

Type helpers

The module provides the following type helper functions. You can wrap parameters with these functions when passing them to DB.query() or DB.query_formatted() in order to give PyGreSQL a hint about the type of the parameters, if it cannot be derived from the context.

pg.Bytea(bytes)

A wrapper for holding a bytea value

New in version 5.0.

pg.HStore(dict)

A wrapper for holding an hstore dictionary

New in version 5.0.

pg.Json(obj)

A wrapper for holding an object serializable to JSON

New in version 5.0.

The following additional type helper is only meaningful when used with DB.query_formatted(). It marks a parameter as text that shall be literally included into the SQL. This is useful for passing table names for instance.

pg.Literal(sql)

A wrapper for holding a literal SQL string

New in version 5.0.

Module constants

Some constants are defined in the module dictionary. They are intended to be used as parameters for methods calls. You should refer to the libpq description in the PostgreSQL user manual for more information about them. These constants are:

pg.version
pg.__version__

constants that give the current version

pg.INV_READ
pg.INV_WRITE

large objects access modes, used by Connection.locreate() and LargeObject.open()

pg.SEEK_SET
pg.SEEK_CUR
pg.SEEK_END

positional flags, used by LargeObject.seek()

pg.TRANS_IDLE
pg.TRANS_ACTIVE
pg.TRANS_INTRANS
pg.TRANS_INERROR
pg.TRANS_UNKNOWN

transaction states, used by Connection.transaction()