jk_prettyprintobj

Introduction

This python module provides a mixin for dumping objects. This is ment for debugging purposes: Sometimes it is very convenient to have a way of writing all data of an object to STDOUT in a human readable way. This module assists in such implementations.

Information about this module can be found here:

• github.org
• pypi.python.org

How to use this module

Import the module

In order to use this module you need to import it first. So add this line to the head of your python source code file:

import jk_prettyprintobj

Use the provided mixin

To make use of the features of this module you must add a mixin to your class. Example:

class ExampleClass(jk_prettyprintobj.DumpMixin):

	def __init__(self, ...):
		...

	...

If you derive your class from a base class just add the mixin to your list of base classes. The order does not matter in this case. Here is an example how to do this:

class MyFancyException(Exception, jk_prettyprintobj.DumpMixin):

	def __init__(self, msg:str):
		super().__init__(msg)

	...

In this example we use Exception as a base class to keep this example simple. It just demonstrates the technique. You can use any base class for inheritance, it is just necessary that you somewhere in the list of base classes add jk_prettyprintobj.DumpMixin. This does not yet make use of the features provided by jk_prettyprintobj but prepares its use.

This mixin adds a regular method named dump() to the class. For all things to work it is important that you have no other method named dump() in your class that might conflict with the implementation provided by DumpMixin. This method can be called later, but some additional implementation steps need to be taken first. (See next section!)

Implement a helper method

To actually enable the class to produce output we must implement one of the helper methods. These are:

More to these options in the next sections.

Helper method _dumpVarNames()

If you implement the method _dumpVarNames() -> typing.List[str] your method needs to return a list of variable names that should be dumped to STDOUT.

Here is an example of a simple but working implementation.

class Matrix(jk_prettyprintobj.DumpMixin):

	def __init__(self, m):
		self.m = m
		self.nRows = len(m)
		self.nColumns = len(m[0])

	def _dumpVarNames(self) -> list:
		return [
			"nRows",
			"nColumns",
			"m",
		]

Now what _dumpVarNames() will do is simply returning a list of variables to access for output.

As private variables can not be accessed by mixins all variables in this example have therefore been defined as public variables. This is a general limitation of python so there is no way around this: All variables to output this way need to be non-private.

Now let's create an instance of Matrix and invoke dump():

m = Matrix([
	[	1,	2,	3 	],
	[	4,	5,	6 	],
	[	7,	8,	9.1234567	],
])

m.dump()

If dump() is invoked on an initialized instance of Matrix from this example such an object will render to something like this:

<Matrix(
	nRows = 3
	nColumns = 3
	m = [
		[ 1, 2, 3 ],
		[ 4, 5, 6 ],
		[ 7, 8, 9.1234567 ],
	]
)>

Helper method _dump(ctx)

If you implement the method _dump(ctx:jk_prettyprintobj.DumpCtx) -> None your method needs to use the provided context object to implement dumping variables to STDOUT on your own. This variant is helpful if you - for some reason - require to output private variables that an implementation of _dumpVarNames() will not be able to access.

By implementing this method you will also be able to modify the way how the contents of a variable will be written to STDOUT.

Here is an example of a simple but working implementation:

class Matrix(jk_prettyprintobj.DumpMixin):

	def __init__(self, m):
		self.__m = m
		self.__nRows = len(m)
		self.__nColumns = len(m[0])

	def _dump(self, ctx:jk_prettyprintobj.DumpCtx):
		ctx.dumpVar("nRows", self.__nRows)
		ctx.dumpVar("nColumns", self.__nColumns)
		ctx.dumpVar("m", self.__m, "float_round3")

This class is expected to represent a mathematical matrix and therefore should receive a two-dimensional field of float values during construction. During construction this data is stored in a private variable named __m. Additional private variables are created. For simplicity no other methods except dump_() are implemented in this example.

Now what _dump() will do is to invoke dumpVar() for every single variable. dumpVar() has the following signature:

• dumpVar(varName:str, value, postProcessorName:str = None) -> None

This method requires to receive up to three arguments:

• str varName: The name to use for output. In this example we use nRows as we might add a property of exactly this name. (Not implemented in this example!)
• * value: A value of any kind. This is the value that should later on be written to STDOUT.
• str processorName: This optional value can be one of several identifiers that indicate how to process the value before it is converted to a string. (See section below.)

If dump() is invoked on an initialized instance of Matrix from this example such an object will render to something like this:

<Matrix(
	nRows = 3
	nColumns = 3
	m = [
		[ 1, 2, 3 ],
		[ 4, 5, 6 ],
		[ 7, 8, 9.123 ],
	]
)>

Please note that in this case the output of the very last float in the matrix might be rounded to three digits as defined by the processor float_round3. This is different to an implementation providing _dumpVarNames().

Processors

For producing output you can apply a processor that will preprocess the output before writing it to STDOUT. This is useful to achieve a more human readable representation of data in some cases.

These are the processors you can use:

Futher Development

It is likely that future developments will add more alternatives for dumping an objects. If you have any ideas, requirements or recommendations please feel free to leave a comment.

Contact Information

This is Open Source code. That not only gives you the possibility of freely using this code it also allows you to contribute. Feel free to contact the author(s) of this software listed below, either for comments, collaboration requests, suggestions for improvement or reporting bugs:

• Jürgen Knauth: pubsrc@binary-overflow.de

License

This software is provided under the following license:

• Apache Software License 2.0