Python pretty print an XML given an XML string

Question

I generated a long and ugly XML string with Python and I need to filter it through pretty printer to look nicer.

I found this post for python pretty printers, but I have to write the XML string to a file to be read back to use the tools, which I want to avoid if possible.

What python pretty tools are available that work on strings?

Which python XML library are you using? – Paul McMillan Oct 20 '10 at 00:06 — Paul McMillan, Oct 20 '10 at 00:06
@Paul : I use "from xml.dom import minidom". – prosseek Oct 20 '10 at 00:07 — prosseek, Oct 20 '10 at 00:07

score 28 · Accepted Answer · edited Dec 04 '18 at 13:26

28

Here's how to parse from a text string to the lxml structured data type.

Python 2:

from lxml import etree
xml_str = "<parent><child>text</child><child>other text</child></parent>"
root = etree.fromstring(xml_str)
print etree.tostring(root, pretty_print=True)

Python 3:

from lxml import etree
xml_str = "<parent><child>text</child><child>other text</child></parent>"
root = etree.fromstring(xml_str)
print(etree.tostring(root, pretty_print=True).decode())

Outputs:

<parent>
  <child>text</child>
  <child>other text</child>
</parent>

edited Dec 04 '18 at 13:26

Craig

548
9
24

answered Oct 20 '10 at 01:27

monkut

42,176
24
124
155

1

i was having some escaping problems and the `etree.tounicode` method resolved it – mut1na Oct 11 '17 at 11:18
3

With Python 3, use `print(etree.tostring(tree, pretty_print=True).decode())`. – orodbhen Feb 15 '18 at 14:49

Paul McMillan · Answer 2 · 2010-10-20T00:14:22.497

7

I use the lxml library, and there it's as simple as

>>> print(etree.tostring(root, pretty_print=True))

You can do that operation using any etree, which you can either generate programmatically, or read from a file.

If you're using the DOM from PyXML, it's

import xml.dom.ext
xml.dom.ext.PrettyPrint(doc)

That prints to the standard output, unless you specify an alternate stream.

http://pyxml.sourceforge.net/topics/howto/node19.html

To directly use the minidom, you want to use the toprettyxml() function.

http://docs.python.org/library/xml.dom.minidom.html#xml.dom.minidom.Node.toprettyxml

edited Oct 20 '10 at 00:14

answered Oct 20 '10 at 00:07

Paul McMillan

19,693
9
57
71

It looks like that both root and doc are structured data, not a string. Thanks for the answer. – prosseek Oct 20 '10 at 00:13
If your xml exists as a minidom node, you can use the `toprettyxml()` function. If it really only ever exists as a string, you will have to parse it in before you can pretty print it out. – Paul McMillan Oct 20 '10 at 00:16

score 2 · Answer 3 · answered Feb 12 '20 at 17:51

Here's a Python3 solution that gets rid of the ugly newline issue (tons of whitespace), and it only uses standard libraries unlike most other implementations. You mention that you have an xml string already so I am going to assume you used xml.dom.minidom.parseString()

With the following solution you can avoid writing to a file first:

import xml.dom.minidom
import os

def pretty_print_xml_given_string(input_string, output_xml):
    """
    Useful for when you are editing xml data on the fly
    """
    xml_string = input_string.toprettyxml()
    xml_string = os.linesep.join([s for s in xml_string.splitlines() if s.strip()]) # remove the weird newline issue
    with open(output_xml, "w") as file_out:
        file_out.write(xml_string)

I found how to fix the common newline issue here.

Python pretty print an XML given an XML string

3 Answers3

Linked