Question

I've seen a fair share of ungainly XML->JSON code on the web, and having interacted with Stack's users for a bit, I'm convinced that this crowd can help more than the first few pages of Google results can.

So, we're parsing a weather feed, and we need to populate weather widgets on a multitude of web sites. We're looking now into Python-based solutions.

This public weather.com RSS feed is a good example of what we'd be parsing (our actual weather.com feed contains additional information because of a partnership w/them).

In a nutshell, how should we convert XML to JSON using Python?

Was it helpful?

Solution

There is no "one-to-one" mapping between XML and JSON, so converting one to the other necessarily requires some understanding of what you want to do with the results.

That being said, Python's standard library has several modules for parsing XML (including DOM, SAX, and ElementTree). As of Python 2.6, support for converting Python data structures to and from JSON is included in the json module.

So the infrastructure is there.

OTHER TIPS

xmltodict (full disclosure: I wrote it) can help you convert your XML to a dict+list+string structure, following this "standard". It is Expat-based, so it's very fast and doesn't need to load the whole XML tree in memory.

Once you have that data structure, you can serialize it to JSON:

import xmltodict, json

o = xmltodict.parse('<e> <a>text</a> <a>text</a> </e>')
json.dumps(o) # '{"e": {"a": ["text", "text"]}}'

You can use the xmljson library to convert using different XML JSON conventions.

For example, this XML:

<p id="1">text</p>

translates via the BadgerFish convention into this:

{
  'p': {
    '@id': 1,
    '$': 'text'
  }
}

and via the GData convention into this (attributes are not supported):

{
  'p': {
    '$t': 'text'
  }
}

... and via the Parker convention into this (attributes are not supported):

{
  'p': 'text'
}

It's possible to convert from XML to JSON and from JSON to XML using the same conventions:

>>> import json, xmljson
>>> from lxml.etree import fromstring, tostring
>>> xml = fromstring('<p id="1">text</p>')
>>> json.dumps(xmljson.badgerfish.data(xml))
'{"p": {"@id": 1, "$": "text"}}'
>>> xmljson.parker.etree({'ul': {'li': [1, 2]}})
# Creates [<ul><li>1</li><li>2</li></ul>]

Disclosure: I wrote this library. Hope it helps future searchers.

Here's the code I built for that. There's no parsing of the contents, just plain conversion.

from xml.dom import minidom
import simplejson as json
def parse_element(element):
    dict_data = dict()
    if element.nodeType == element.TEXT_NODE:
        dict_data['data'] = element.data
    if element.nodeType not in [element.TEXT_NODE, element.DOCUMENT_NODE, 
                                element.DOCUMENT_TYPE_NODE]:
        for item in element.attributes.items():
            dict_data[item[0]] = item[1]
    if element.nodeType not in [element.TEXT_NODE, element.DOCUMENT_TYPE_NODE]:
        for child in element.childNodes:
            child_name, child_dict = parse_element(child)
            if child_name in dict_data:
                try:
                    dict_data[child_name].append(child_dict)
                except AttributeError:
                    dict_data[child_name] = [dict_data[child_name], child_dict]
            else:
                dict_data[child_name] = child_dict 
    return element.nodeName, dict_data

if __name__ == '__main__':
    dom = minidom.parse('data.xml')
    f = open('data.json', 'w')
    f.write(json.dumps(parse_element(dom), sort_keys=True, indent=4))
    f.close()

If some time you get only response code instead of all data then error like json parse will be there so u need to convert it as text

import xmltodict

data = requests.get(url)
xpars = xmltodict.parse(data.text)
json = json.dumps(xpars)
print json 

There is a method to transport XML-based markup as JSON which allows it to be losslessly converted back to its original form. See http://jsonml.org/.

It's a kind of XSLT of JSON. I hope you find it helpful

You may want to have a look at http://designtheory.org/library/extrep/designdb-1.0.pdf. This project starts off with an XML to JSON conversion of a large library of XML files. There was much research done in the conversion, and the most simple intuitive XML -> JSON mapping was produced (it is described early in the document). In summary, convert everything to a JSON object, and put repeating blocks as a list of objects.

objects meaning key/value pairs (dictionary in Python, hashmap in Java, object in JavaScript)

There is no mapping back to XML to get an identical document, the reason is, it is unknown whether a key/value pair was an attribute or an <key>value</key>, therefore that information is lost.

If you ask me, attributes are a hack to start; then again they worked well for HTML.

Well, probably the simplest way is just parse the XML into dictionaries and then serialize that with simplejson.

I'd suggest not going for a direct conversion. Convert XML to an object, then from the object to JSON.

In my opinion, this gives a cleaner definition of how the XML and JSON correspond.

It takes time to get right and you may even write tools to help you with generating some of it, but it would look roughly like this:

class Channel:
  def __init__(self)
    self.items = []
    self.title = ""

  def from_xml( self, xml_node ):
    self.title = xml_node.xpath("title/text()")[0]
    for x in xml_node.xpath("item"):
      item = Item()
      item.from_xml( x )
      self.items.append( item )

  def to_json( self ):
    retval = {}
    retval['title'] = title
    retval['items'] = []
    for x in items:
      retval.append( x.to_json() )
    return retval

class Item:
  def __init__(self):
    ...

  def from_xml( self, xml_node ):
    ...

  def to_json( self ):
    ...

I found for simple XML snips, use regular expression would save troubles. For example:

# <user><name>Happy Man</name>...</user>
import re
names = re.findall(r'<name>(\w+)<\/name>', xml_string)
# do some thing to names

To do it by XML parsing, as @Dan said, there is not one-for-all solution because the data is different. My suggestion is to use lxml. Although not finished to json, lxml.objectify give quiet good results:

>>> from lxml import objectify
>>> root = objectify.fromstring("""
... <root xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
...   <a attr1="foo" attr2="bar">1</a>
...   <a>1.2</a>
...   <b>1</b>
...   <b>true</b>
...   <c>what?</c>
...   <d xsi:nil="true"/>
... </root>
... """)

>>> print(str(root))
root = None [ObjectifiedElement]
    a = 1 [IntElement]
      * attr1 = 'foo'
      * attr2 = 'bar'
    a = 1.2 [FloatElement]
    b = 1 [IntElement]
    b = True [BoolElement]
    c = 'what?' [StringElement]
    d = None [NoneElement]
      * xsi:nil = 'true'

While the built-in libs for XML parsing are quite good I am partial to lxml.

But for parsing RSS feeds, I'd recommend Universal Feed Parser, which can also parse Atom. Its main advantage is that it can digest even most malformed feeds.

Python 2.6 already includes a JSON parser, but a newer version with improved speed is available as simplejson.

With these tools building your app shouldn't be that difficult.

When I do anything with XML in python I almost always use the lxml package. I suspect that most people use lxml. You could use xmltodict but you will have to pay the penalty of parsing the XML again.

To convert XML to json with lxml you:

  1. Parse XML document with lxml
  2. Convert lxml to a dict
  3. Convert list to json

I use the following class in my projects. Use the toJson method.

from lxml import etree 
import json


class Element:
    '''
    Wrapper on the etree.Element class.  Extends functionality to output element
    as a dictionary.
    '''

    def __init__(self, element):
        '''
        :param: element a normal etree.Element instance
        '''
        self.element = element

    def toDict(self):
        '''
        Returns the element as a dictionary.  This includes all child elements.
        '''
        rval = {
            self.element.tag: {
                'attributes': dict(self.element.items()),
            },
        }
        for child in self.element:
            rval[self.element.tag].update(Element(child).toDict())
        return rval


class XmlDocument:
    '''
    Wraps lxml to provide:
        - cleaner access to some common lxml.etree functions
        - converter from XML to dict
        - converter from XML to json
    '''
    def __init__(self, xml = '<empty/>', filename=None):
        '''
        There are two ways to initialize the XmlDocument contents:
            - String
            - File

        You don't have to initialize the XmlDocument during instantiation
        though.  You can do it later with the 'set' method.  If you choose to
        initialize later XmlDocument will be initialized with "<empty/>".

        :param: xml Set this argument if you want to parse from a string.
        :param: filename Set this argument if you want to parse from a file.
        '''
        self.set(xml, filename) 

    def set(self, xml=None, filename=None):
        '''
        Use this to set or reset the contents of the XmlDocument.

        :param: xml Set this argument if you want to parse from a string.
        :param: filename Set this argument if you want to parse from a file.
        '''
        if filename is not None:
            self.tree = etree.parse(filename)
            self.root = self.tree.getroot()
        else:
            self.root = etree.fromstring(xml)
            self.tree = etree.ElementTree(self.root)


    def dump(self):
        etree.dump(self.root)

    def getXml(self):
        '''
        return document as a string
        '''
        return etree.tostring(self.root)

    def xpath(self, xpath):
        '''
        Return elements that match the given xpath.

        :param: xpath
        '''
        return self.tree.xpath(xpath);

    def nodes(self):
        '''
        Return all elements
        '''
        return self.root.iter('*')

    def toDict(self):
        '''
        Convert to a python dictionary
        '''
        return Element(self.root).toDict()

    def toJson(self, indent=None):
        '''
        Convert to JSON
        '''
        return json.dumps(self.toDict(), indent=indent)


if __name__ == "__main__":
    xml='''<system>
    <product>
        <demod>
            <frequency value='2.215' units='MHz'>
                <blah value='1'/>
            </frequency>
        </demod>
    </product>
</system>
'''
    doc = XmlDocument(xml)
    print doc.toJson(indent=4)

The output from the built in main is:

{
    "system": {
        "attributes": {}, 
        "product": {
            "attributes": {}, 
            "demod": {
                "attributes": {}, 
                "frequency": {
                    "attributes": {
                        "units": "MHz", 
                        "value": "2.215"
                    }, 
                    "blah": {
                        "attributes": {
                            "value": "1"
                        }
                    }
                }
            }
        }
    }
}

Which is a transformation of this xml:

<system>
    <product>
        <demod>
            <frequency value='2.215' units='MHz'>
                <blah value='1'/>
            </frequency>
        </demod>
    </product>
</system>

jsonpickle or if you're using feedparser, you can try feed_parser_to_json.py

My answer addresses the specific (and somewhat common) case where you don't really need to convert the entire xml to json, but what you need is to traverse/access specific parts of the xml, and you need it to be fast, and simple (using json/dict-like operations).

Approach

For this, it is important to note that parsing an xml to etree using lxml is super fast. The slow part in most of the other answers is the second pass: traversing the etree structure (usually in python-land), converting it to json.

Which leads me to the approach I found best for this case: parsing the xml using lxml, and then wrapping the etree nodes (lazily), providing them with a dict-like interface.

Code

Here's the code:

from collections import Mapping
import lxml.etree

class ETreeDictWrapper(Mapping):

    def __init__(self, elem, attr_prefix = '@', list_tags = ()):
        self.elem = elem
        self.attr_prefix = attr_prefix
        self.list_tags = list_tags

    def _wrap(self, e):
        if isinstance(e, basestring):
            return e
        if len(e) == 0 and len(e.attrib) == 0:
            return e.text
        return type(self)(
            e,
            attr_prefix = self.attr_prefix,
            list_tags = self.list_tags,
        )

    def __getitem__(self, key):
        if key.startswith(self.attr_prefix):
            return self.elem.attrib[key[len(self.attr_prefix):]]
        else:
            subelems = [ e for e in self.elem.iterchildren() if e.tag == key ]
            if len(subelems) > 1 or key in self.list_tags:
                return [ self._wrap(x) for x in subelems ]
            elif len(subelems) == 1:
                return self._wrap(subelems[0])
            else:
                raise KeyError(key)

    def __iter__(self):
        return iter(set( k.tag for k in self.elem) |
                    set( self.attr_prefix + k for k in self.elem.attrib ))

    def __len__(self):
        return len(self.elem) + len(self.elem.attrib)

    # defining __contains__ is not necessary, but improves speed
    def __contains__(self, key):
        if key.startswith(self.attr_prefix):
            return key[len(self.attr_prefix):] in self.elem.attrib
        else:
            return any( e.tag == key for e in self.elem.iterchildren() )


def xml_to_dictlike(xmlstr, attr_prefix = '@', list_tags = ()):
    t = lxml.etree.fromstring(xmlstr)
    return ETreeDictWrapper(
        t,
        attr_prefix = '@',
        list_tags = set(list_tags),
    )

This implementation is not complete, e.g., it doesn't cleanly support cases where an element has both text and attributes, or both text and children (only because I didn't need it when I wrote it...) It should be easy to improve it, though.

Speed

In my specific use case, where I needed to only process specific elements of the xml, this approach gave a suprising and striking speedup by a factor of 70 (!) compared to using @Martin Blech's xmltodict and then traversing the dict directly.

Bonus

As a bonus, since our structure is already dict-like, we get another alternative implementation of xml2json for free. We just need to pass our dict-like structure to json.dumps. Something like:

def xml_to_json(xmlstr, **kwargs):
    x = xml_to_dictlike(xmlstr, **kwargs)
    return json.dumps(x)

If your xml includes attributes, you'd need to use some alphanumeric attr_prefix (e.g. "ATTR_"), to ensure the keys are valid json keys.

I haven't benchmarked this part.

This stuff here is actively maintained and so far is my favorite: xml2json in python

To anyone that may still need this. Here's a newer, simple code to do this conversion.

from xml.etree import ElementTree as ET

xml    = ET.parse('FILE_NAME.xml')
parsed = parseXmlToJson(xml)


def parseXmlToJson(xml):
  response = {}

  for child in list(xml):
    if len(list(child)) > 0:
      response[child.tag] = parseXmlToJson(child)
    else:
      response[child.tag] = child.text or ''

    # one-liner equivalent
    # response[child.tag] = parseXmlToJson(child) if len(list(child)) > 0 else child.text or ''

  return response

check out lxml2json (disclosure: I wrote it)

https://github.com/rparelius/lxml2json

it's very fast, lightweight (only requires lxml), and one advantage is that you have control over whether certain elements are converted to lists or dicts

You can use declxml. It has advanced features like multi attributes and complex nested support. You just need to write a simple processor for it. Also with the same code, you can convert back to JSON as well. It is fairly straightforward and the documentation is awesome.

Link: https://declxml.readthedocs.io/en/latest/index.html

Prepare data in Python: To create JSON first you need to prepare data in python. We can use List and Dictionary in Python to prepare the data.

Python List <==> JSON Array

Python Dictionary <==> JSON Object (Key Value Format) Check this for more details

https://devstudioonline.com/article/create-json-and-xml-in-python

To represent data in JSON format

name=John
age=20
gender=male
address=Sector 12 Greater Kailash, New Delhi
Jobs=Noida,Developer | Gurugram,Tester |Faridabad,Designer

In json we repesent a data in key and value format

{
    "name":"john",
    "age":20,
    "gender":"male",
    "address":["New kP college","Greater Kailash","New Delhi"],
    "jobs":[
               {"Place":"Noida","Title":"Developer "},
               {"Place":"Gurugram","Title":"Tester "},
               {"Place":"Faridabad","Title":"Designer"}
           ]
}

To represent data in XML format

<!-- In xml we write a code under a key you can take any key -->
<info> <!-- key open -->

<name> john </name> 
<age> 20 </age>
<gender> male </gender>

<address> 
<item> New kP college </item>
<item> Greater Kailash </item>
<item> New Delhi </item>
</address>

<jobs>
 <item>
  <title>Developer </title>
  <place>Noida</place>
 </item>

 <item>
  <title>Designer</title>
  <place>Gurugram</place>
 </item>
 
 <item>
  <title>Developer </title>
  <place>Faridabad</place>
 </item>
</jobs>

</info> <!-- key close-->

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top