Python advanced 3.the python std lib by example – application building blocks

THE PYTHON STD LIB BY EXAMPLE
– APPLICATION BUILDING BLOCKS
John
Saturday, December 21, 2013

Brief introduction
• This lesson covers some of the more frequently
reused building blocks that solve problems
common to many applications
• Command-line argument parsing: module
getopt,optparse,argparse.
• Interactive programs: readline, cmd,
shlex,fileinput
• Manage application configuration: ConfigParser
• Manage log file: logging
• Others: atexit, sched etc

Module getopt – option parsing
• The getopt() take 3 arguments:
1. The sequeence of arguments to be parsed.
Usually it is sys.argv[1:]
2. Single-character option, following a colon(:)
mean it require an argument. E.g. ab:c:, it mean
a is -a simple flag, while –b and –c require an
argument.
3. It is optional. If used, it is a list of long-style
option names. Having suffix “=“ means ruquire
an argument. E.g [‘noarg’,’witharg=‘]

Quick example 1
# file test.py
import getopt
import sys
opts,args = getopt.getopt(sys.argv[1:],’ab:c:’)
for opt in opts:
print opt
•Then run command line “python test.py –a –b valb –c valc”, the output
should be
('-a', '')
('-b', 'val')
('-c', 'val')

quick example 2
# file test.py
import getopt
import sys
opts,args = getopt.getopt(sys.argv[1:],’’,[‘opta’,’optb=‘,’optc=‘])
for opt in opts:
print opt
•run command line “python test.py --opta --optb val --optc val”
('--opta', '')
('--optb', 'val')
('--optc', 'val')

Module optparse
• it is a modern alternative for module getopt

More options in add_option
method
• option dest: the attribute name
• option default: the default value. also can be set in
set_defaults method.
• option type: convert the argument string to specific type,
e.g. int, float,string
• option choices: validation use a list of candidate strings. e.g.
choices=[‘a’,’b’,’c’] . the valid value should be ‘a’, or ‘b’ or ‘c’
• option help: help message
• option action: store, store_const,store_true,append,acount

Module argparse
• argparse added to python 2.7 as a
replacement of optparse.
• Change the OptionParser by ArgumentParser,
and add_option by add_argument in
previous example.

More Options of ArgumentParser
• add help automatically:

parser = argparse.ArgumentParser(add_help=True)

• version control: version=‘1.0’ (-v or --version
will show the version number)

More options of add_argument
method
• All options in optparse.add_option method.
• option nargs: number of expected
arguments. (? mean 0 or 1 arguments, *
means 0 or all, + means 1 or all).
example
parser.add_argument(‘-c’,nargs=3)
we need write command line “perl -c 1 2 3”

Module getpass: handle passowrd
prompts securely
• print a promt and then read input from the user.

• We can change the prompt:
getpass(prompt=‘what is your favorite color?’)

MODULE CMD:
BUILD COMMAND
LINE PROCESSORS

brief introduction
• the Cmd class is used as base class for
interactive shells and other command
interpreters.
• It can provide interactive prompt, commandline editing, and command completion.

quick example

• method do_greet() connect with command “greet”
• method help_greet() connect with command “help greet”

auto-completion
• auto-completion is already enabled in previous
example.
• if user want to do user-defined on auto-completion,
use complete_<comand> (e.g. complete_greet() )

Overrite Base Class method
• Method cmploop(intro) is the main
processing loop. Each iteration in cmdloop
call parseline and onecmd.
• Method emptyline: behavior if emptyline.
Default behavior is that rerun previous
command.
• Method default: default method if command
is not found.

Running shell command
• An exclamation point(!) map to the do_shell()
method and is intended “shelling out” to run other
commands.
• First import subprocess module.
• Second, define do_shell() method

Brief introduction
• The logging module define standard API for
reporting error and status information.
• Both application developer and module
author benefit from this module, but has
different considerations.

Logging to file
• Use basicConfig set the log file name

The level of the events logging
tracks
Level

Numeric
value

When it’s used

DEBU Detailed information, typically of interest only when
G
diagnosing problems.
INFO Confirmation that things are working as expected.
An indication that something unexpected happened, or
WAR
indicative of some problem in the near future (e.g. ‘disk space
NING
low’). The software is still working as expected.
ERRO Due to a more serious problem, the software has not been
R
able to perform some function.
CRITIC A serious error, indicating that the program itself may be
AL
unable to continue running.

10
20
30
40
50

Best practice: logging in single file
import logging
logging.basicConfig(level=logging.WARNING) # Only the level
equal or alove this level can be displayed.
logging.debug('This message should go to the log file')
logging.info('So should this')
logging.warning('And this, too')
•Write the log to file:
logging.basicConfig(filename='example.log',level=logging.DEBU
G)

Best practice: Logging from
multiple modules
• Define logging config
in main file

• Write log in other files

Python advanced 3.the python std lib by example – application building blocks

Recommended

More Related Content

What's hot (20)

Similar to Python advanced 3.the python std lib by example – application building blocks (20)

More from John(Qiang) Zhang (7)

Recently uploaded (20)

Python advanced 3.the python std lib by example – application building blocks