Python 3.9.0 : Testing twisted python module - part 001 .

Today I tested two python modules named: twisted and twisted[tls].
Twisted is an event-driven network programming framework written in Python and licensed under the MIT License. Twisted projects variously support TCP, UDP, SSL/TLS, IP multicast, Unix domain sockets, many protocols (including HTTP, XMPP, NNTP, IMAP, SSH, IRC, FTP, and others), and much more. Twisted is based on the event-driven programming paradigm, which means that users of Twisted write short callbacks which are called by the framework., see wikipedia webpage.
In this tutorial I will show you only some of these tests and how you can work with these python modules.
About twisted you can read more at the official webpage. In Fedora distro version 33 you can use the dnf tool to search for and install these python packages.

[root@desk mythcat]# dnf search twisted
...
python3-twisted.x86_64 : Twisted is a networking engine written in Python
python3-twisted+tls.x86_64 : Metapackage for python3-twisted: tls extras

You can also use the pip tool for installation:

[mythcat@desk ~]$ cd PythonProjects/
[mythcat@desk PythonProjects]$ pip3 install twisted
...
[mythcat@desk PythonProjects]$ pip3 install twisted[tls]
...

I used python 3.9.0 to test if this python package works:

[mythcat@desk PythonProjects]$ python3.9
Python 3.9.0 (default, Oct  6 2020, 00:00:00) 
[GCC 10.2.1 20200826 (Red Hat 10.2.1-3)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from twisted.protocols import basic

Let's test with simple example using the reactor and protocol:

from twisted.internet import reactor, protocol

class ClientEcho(protocol.Protocol):
    def connectionMade(self):
        self.transport.write("Hello, world!".encode('utf-8'))

    def dataReceived(self, data):
        print ("Server: ", data)
        self.transport.loseConnection()

class FactoryEcho(protocol.ClientFactory):
    def buildProtocol(self, addr):
        return ClientEcho()

    def clientConnectionFailed(self, connector, reason):
        print ("Connection failed")
        reactor.stop()

    def clientConnectionLost(self, connector, reason):
        print ("Connection lost")
        reactor.stop()

reactor.connectTCP("localhost", 8080, FactoryEcho())
reactor.run()

Your protocol handling class will usually subclass twisted.internet.protocol.Protocol.
The default factory class twisted.internet.protocol.Factory just instantiates each Protocol and lets every Protocol access, and possibly modify, the persistent configuration.
This protocol responds to the initial connection with a well known quote, and then terminates the connection.
The protocol never waits for an event because handles data in an asynchronous manner.
The reactor interface lets many different loops handle the networking code.
The source code have two classes each is used to show a simple echo client on port 8080 - you can use any port.
This source code is the most simple example to understand the relation between factory , protocol and reactor.
The result is this:

[mythcat@desk PythonProjects]$ python3.9 echo_client_001.py 
Server:  b'Hello, world!'
Connection lost

Python 3.9.0 : Union and in-place union operators

Python introduces two new operators for dictionaries named union used in code with pipe operator | and in-place union used in python code with this |=.

I this tutorial I will show you how can be used:

[mythcat@desk ~]$ python3.9 
Python 3.9.0 (default, Oct  6 2020, 00:00:00) 
[GCC 10.2.1 20200723 (Red Hat 10.2.1-1)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> step_one = {"worker_one":"task_one", "worker_two":"task_two"}
>>> step_two = {"worker_three":"task_one", "worker_fouw":"task_two"}
>>> merged_step_one_step_two = {**step_one, **step_two}
>>> empty = {}
>>> empty |= step_one
>>> empty
{'worker_one': 'task_one', 'worker_two': 'task_two'}
>>> step_one 
{'worker_one': 'task_one', 'worker_two': 'task_two'}
>>> step_two
{'worker_three': 'task_one', 'worker_fouw': 'task_two'}
>>> all = step_one | step_two 
>>> all 
{'worker_one': 'task_one', 'worker_two': 'task_two', 'worker_three': 'task_one', 'worker_fouw': 'task_two'} 
>>> copy_step_one = empty | step_one
>>> copy_step_one 
{'worker_one': 'task_one', 'worker_two': 'task_two'} 
>>> copy_step_one |= [("manager", "steps")]
>>> copy_step_one 
{'worker_one': 'task_one', 'worker_two': 'task_two', 'manager': 'steps'}

You can easily see how to use these operators with the dictionaries created and how to add lists to dictionaries. Operations with these operators always result in a dictionary. The order of these operations is important, see example:

>>> numbers = {0: "zero", 1: "one"}
>>> numbers_one = {0: "zero", 1: "one"}
>>> numbers_two = {0: "zero", 2: "two"}
>>> numbers_one | numbers_two
{0: 'zero', 1: 'one', 2: 'two'}
>>> numbers_two | numbers_one
{0: 'zero', 2: 'two', 1: 'one'}
>>> numbers_three = {0: 'zero', 11: 'error 1', 22: 'error 2', 33:'error 3'}
>>> numbers_three | numbers_one
{0: 'zero', 11: 'error 1', 22: 'error 2', 33: 'error 3', 1: 'one'}
>>> numbers_one | numbers_three
{0: 'zero', 1: 'one', 11: 'error 1', 22: 'error 2', 33: 'error 3'}

You can see how the dictionaries are overwritten by the union operation. These operations are implemented through dunder operations for most the dictionary objects except abstract classes. Dunder or magic methods in Python are the methods having two prefix and suffix underscores in the method name. You can find on this page.

Python 3.9.0 : Introduction to release 3.9.0.

This is a short introduction to release 3.9.0. 

Five days ago, a new release of version 3.9 appeared with a series of improvements and new python packages, see the official website. 

You can install easily on Fedora 32 with dnf tool.

[root@desk mythcat]# dnf install  python39.x86_64
...
Installing:
 python39                x86_64                3.9.0-1.fc32
...
Installed:
  python39-3.9.0-1.fc32.x86_64                                                                             

Complete!

A new article written about this release shows some of the advantages of this programming language, you can read it here. 

You can run easily with this command;

[root@desk mythcat]# python3.9
Python 3.9.0 (default, Oct  6 2020, 00:00:00) 
[GCC 10.2.1 20200723 (Red Hat 10.2.1-1)] on linux
Type "help", "copyright", "credits" or "license" for more information. 

You can configure python with this command, see an example:

[root@desk mythcat]# python3.9-x86_64-config --libs 
 -lcrypt -lpthread -ldl  -lutil -lm -lm 

This release of Python 3.9 uses a new parser, based on parsing expression grammar (PEG) instead of LL(1) know as LL parser (Left-to-right, Leftmost derivation). PEG parsers are more powerful than LL(1) parsers and avoid the need for special hacks. This the same abstract syntax tree (AST) as the old LL(1) parser. The PEG parser is the default, but you can run your program using the old parser, see the next example:

[mythcat@desk ~]$ python3.9 -X oldparser my_script_name.py
...

One of the most important improvements for me is PEP 614 -- Relaxing Grammar Restrictions On Decorators. In Python 3.9, these restrictions are lifted and you can now use any expression. 

For example including one accessing items in a dictionary, like this simple example:

buttons = {
  "hello": QPushButton("hello word!"),
  "bye": QPushButton("bye word!"),
  "buy": QPushButton("buy!!"),
}
...
@buttons["hello"].clicked.connect
def speak_hello():
... 

Yes, comes with another Python Enhancement Proposals - PEP 615 with support for the IANA Time Zone Database in the Standard Library. This acronym is similar to I.A.N.A known as the Internet Assigned Numbers Authority, but it is not the same. 

In this case, the zoneinfo module provides a concrete time zone implementation to support the IANA time zone database. 

This support contains code and data that represent the history of local time for many representative locations around the globe. Many features for math, strings, union operator for the dictionary, HTTP codes, and more. 

Completion of variable and module names is automatically enabled at interpreter startup so that the Tab key invokes the completion function. 

You can see a simple example with zoneinfo python module and completion feature.

[mythcat@desk ~]$ python3.9
Python 3.9.0 (default, Oct  6 2020, 00:00:00) 
[GCC 10.2.1 20200723 (Red Hat 10.2.1-1)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from zoneinfo import ZoneInfo
>>> import zoneinfo
>>> zoneinfo.available_timezones()
{'Hongkong', 'America/Iqaluit', 'America/Indianapolis', 'America/Louisville', 'America/New_York', 
'America/Mazatlan', 'Australia/Yancowinna', 'Africa/Ndjamena', 'Portugal', 'Africa/Bujumbura', 
'America/Rosario', 'America/Antigua','America/Indiana/Tell_City', 'America/Managua', 
'Europe/Paris', 'Europe/Oslo', 
...
>>> tz = ZoneInfo("Europe/Bucharest")
>>> tz.
tz.clear_cache(  tz.from_file(    tz.key           tz.tzname(       
tz.dst(          tz.fromutc(      tz.no_cache(     tz.utcoffset(    
>>> tz. 

On my desktop I got this result:

[mythcat@desk ~]$ python var_access_benchmark.py 
Variable and attribute read access:
   6.0 ns	read_local
   6.5 ns	read_nonlocal
  10.7 ns	read_global
  10.7 ns	read_builtin
  25.2 ns	read_classvar_from_class
  23.4 ns	read_classvar_from_instance
  34.3 ns	read_instancevar
  29.4 ns	read_instancevar_slots
  26.8 ns	read_namedtuple
  42.4 ns	read_boundmethod

Variable and attribute write access:
   6.6 ns	write_local
   7.0 ns	write_nonlocal
  23.3 ns	write_global
  54.1 ns	write_classvar
  46.4 ns	write_instancevar
  39.4 ns	write_instancevar_slots

Data structure read access:
  26.1 ns	read_list
  28.1 ns	read_deque
  27.5 ns	read_dict
  25.8 ns	read_strdict

Data structure write access:
  29.6 ns	write_list
  31.7 ns	write_deque
  34.4 ns	write_dict
  31.7 ns	write_strdict

Stack (or queue) operations:
  55.5 ns	list_append_pop
  49.7 ns	deque_append_pop
  51.0 ns	deque_append_popleft

Timing loop overhead:
   0.4 ns	loop_overhead