This Python 2 and 3 project (tested with CPython 2.7 and 3.3 as well as PyPy 2) allows for easy creation of XML. The hierarchical structure of XML is easy to spot and the code to create XML is much shorter than using SAX, DOM or similar APIs. There is also functionality to efficiently validate and transform XML while it is being created.
If you want to jump right in, have a look at the examples...
The most convenient usage - if you can live with defining allowed element names beforehand, want to define a vocabulary anyway or just want to create HTML5:
XML-String Creation
ElementTree Creation
DOM Creation
ecoxipy.pyxom Examples
See this example of how to create a simple HTML5 document template function:
# In the function it is applied to the "html5" decorator creates the variable
# "_b" being an instance of "ecoxipy.MarkupBuilder" with
# "ecoxipy.string_output.StringOutput" for XML creation. It also creates for
# each HTML5 element a variable being a method of "_b", with the name of
# element, variable and method all being equal.
from ecoxipy.html import html5
@html5
def create_testdoc(title, subtitle, *content):
# Slicing without start argument on the builder creates a document, the
# stop argument defines the document type declaration and the step
# argument defines if the XML declaration should be omitted.
return _b[:'html':True](
# Method calls on a MarkupBuilder instance create elements with the
# name equal to the method name.
html(
# Child dictionary entries become attributes, especially useful
# for non-identifier attribute names:
{'data-info': 'Created by Ecoxipy'},
head(
_b.title(
# Children which are not of the XML representation
# and are either "str" or "unicode" instances or are
# neither iterables, generators nor callables, become text
# nodes:
title
)
),
body(
article(
# Child iterables and generators are unpacked
# automatically:
[h1(title), h2(subtitle)], # Iterable
(p(item) for item in content), # Generator
# Child callables are executed:
hr,
# Calling a MarkupBuilder creates a XML fragment from the
# arguments, here strings are regarded as raw XML.
_b(
# Explicitly create text node:
_b & '<THE END>',
'<footer>Copyright 2013</footer>' # raw XML
)
)
),
# You can also create comments:
_b | "This is a comment.",
# Slicing with a start argument creates a processing instruction:
_b['pi-target':'PI content.'],
# Named arguments of element method-calls become attributes:
xmlns='http://www.w3.org/1999/xhtml/'
)
)
It could be used like this:
>>> create_testdoc('Test', 'A Simple Test Document', 'Hello World & Universe!', 'How are you?')
'<!DOCTYPE html><html data-info="Created by Ecoxipy" xmlns="http://www.w3.org/1999/xhtml/"><head><title>Test</title></head><body><article><h1>Test</h1><h2>A Simple Test Document</h2><p>Hello World & Universe!</p><p>How are you?</p><hr/><THE END><footer>Copyright 2013</footer><!--This is a comment.--><?pi-target PI content.?></article></body></html>'
Pretty-printing the result yields the following HTML:
<!DOCTYPE html>
<html data-info="Created by Ecoxipy" xmlns="http://www.w3.org/1999/xhtml/">
<head>
<title>Test</title>
</head>
<body>
<article>
<h1>Test</h1>
<h2>A Simple Test Document</h2>
<p>Hello World & Universe!</p>
<p>How are you?</p>
<hr/>
<THE END>
<footer>Copyright 2013</footer>
</article>
</body>
<!--This is a comment.-->
<?pi-target PI content.?>
</html>