Pip3line

Python 2.7 & 3 plugins

The short version (for the busy)

Create a usual python file using your favorite text editor/IDE, with this skeleton:

#!/usr/bin/env python
#-*- coding: UTF-8 -*-

def pip3line_transform(inputData):
	""" Take a bytearray as input and returns a bytearray"""
	# if you need to convert to a Python 'string'
	string = inputData.decode('utf-8')

	# Some code ......

	# just remember to convert it back to a bytearray
	return bytearray(string, 'utf-8')

The pip3line_transform function will be called automatically by Pip3line

Now you just need to load it using the Python 2.7/3 module transform (by default in the available list)

If there is no Python transform check the prequisites below.

Prequisites

The two Python plugins need respectively the Python 2.7 and Python 3.x libraries to be installed, if the plugin cannot found them, it will fail to load.

The Python 3 plugin is now compiling against the latest Python 3.x, although it should still works fine with older version

Files locations

The plugins looks for python files in the following locations

python 2.7

• $HOME/.pip3line/python27
• $APP_DIR/plugins/python27
• $PREFIX/share/pip3line/python27 (UNIX only)

python 3

• $HOME/.pip3line/python3
• $APP_DIR/plugins/python3
• $PREFIX/share/pip3line/python3 (UNIX only)

They will load every Python files found and register them into the transformation list based on their file name.

Additionally you can add your own files outside these directories, and make them "persistent" within Pip3line. See the transformation called "Python script 2.7/3" for this.

A few notes on the Python interpreters states and the module reload feature

Both plugins load and initialize one and only one python interpreter, so there are two loaded in memory if both plugins are loaded properly.

The auto-reload feature will reinitialise the loaded module every time a transformation is applied, so in this case there can be no persistence in the module data

However if the auto-reload feature is disable the plugin module will be able to keep data accross transformation, but be aware that this applies to all instances of the same module.

The Code

It is relatively simple, as shown in the extras directory.

You just need to implement the function pip3line_transform somewhere:

#!/usr/bin/env python
#-*- coding: UTF-8 -*-

def pip3line_transform(inputData):
	""" Take a bytearray as input and returns a bytearray"""
	# if you need to convert to a Python 'string'
	string = inputData.decode('utf-8')

	# Some code ......

	# just remember to convert it back to a bytearray
	return bytearray(string, 'utf-8')

If you want to implement a bi-directional transformation set the Pip3line_is_two_ways to true

Then check the Pip3line_INBOUND value

Use the gzip and bz2 examples as a starting point:

#!/usr/bin/env python3
#-*- coding: UTF-8 -*-

import gzip

Pip3line_is_two_ways = True

def pip3line_transform(inputData):
	ret = None
	if (Pip3line_INBOUND):
		ret = bytearray(gzip.compress(inputData))
	else:
		ret = bytearray(gzip.decompress(inputData))
	return ret

Taking parameters from the gui

You can also send parameters to the scripts via the gui setting tab which are accessible as a Python dictionary Pip3line_params

you can define default parameters names by filling the Pip3line_params_names string list

Pip3line_params_names = ["param1", "param2"]

They are then accessible in the gui in the settings tab:

The transformation can then access the values via normal dictionary access

Pip3line_params["param1"]
Pip3line_params["param2"]