XML | Ucode

Appearance

On this page

XML ​

Although XML is more complex than JSON and not as commonly used in web applications today, it's still important to understand how to work with XML.

DOM vs SAX ​

There are two primary methods for parsing XML: DOM and SAX.

  • DOM loads the entire XML document into memory, creating a tree structure. This consumes more memory and can be slower, but allows for easy traversal of nodes.
  • SAX operates in a streaming fashion, parsing the XML as it reads it. This method is more memory-efficient and faster, but requires the programmer to handle events manually.

In most cases, it's preferable to use SAX due to its lower memory usage.

SAX Example ​

Using SAX to parse XML in Python is straightforward. You typically define three key event handlers: start_element, end_element, and char_data. Here's a simple example:

python
from xml.parsers.expat import ParserCreate

class DefaultSaxHandler(object):
    def start_element(self, name, attrs):
        print('sax:start_element: %s, attrs: %s' % (name, str(attrs)))

    def end_element(self, name):
        print('sax:end_element: %s' % name)

    def char_data(self, text):
        print('sax:char_data: %s' % text)

xml = r'''<?xml version="1.0"?>
<ol>
    <li><a href="/python">Python</a></li>
    <li><a href="/ruby">Ruby</a></li>
</ol>
'''

handler = DefaultSaxHandler()
parser = ParserCreate()
parser.StartElementHandler = handler.start_element
parser.EndElementHandler = handler.end_element
parser.CharacterDataHandler = handler.char_data
parser.Parse(xml)

Note: When reading large strings, the CharacterDataHandler might be called multiple times, so you'll need to concatenate the data and process it in the EndElementHandler.

Generating XML ​

Most of the time, generated XML structures are simple. The most straightforward way to create XML is by concatenating strings:

python
L = []
L.append(r'<?xml version="1.0"?>')
L.append(r'<root>')
L.append(encode('some & data'))
L.append(r'</root>')
return ''.join(L)

For more complex XML, it's advisable to use JSON instead.

Summary ​

When parsing XML, focus on the nodes of interest, save the data during event handling, and process it after parsing is complete.

Exercise ​

Here's an exercise to parse the XML format of weather forecasts from WeatherAPI using SAX:

python
from xml.parsers.expat import ParserCreate
from urllib import request

class WeatherSaxHandler(object):
    def __init__(self):
        self.city = ''
        self.condition = ''
        self.temperature = ''
        self.wind = ''
        self.current_tag = ''

    def start_element(self, name, attrs):
        self.current_tag = name

    def end_element(self, name):
        if name == 'location':
            self.city = self.current_data
        elif name == 'current':
            pass  # You could process the current weather here

    def char_data(self, text):
        if self.current_tag == 'name':
            self.current_data = text.strip()
        elif self.current_tag == 'condition':
            self.condition = text.strip()
        elif self.current_tag == 'temp_c':
            self.temperature = text.strip()
        elif self.current_tag == 'wind_kph':
            self.wind = text.strip()

def parseXml(xml_str):
    handler = WeatherSaxHandler()
    parser = ParserCreate()
    parser.StartElementHandler = handler.start_element
    parser.EndElementHandler = handler.end_element
    parser.CharacterDataHandler = handler.char_data
    parser.Parse(xml_str)
    
    return {
        'city': handler.city,
        'weather': {
            'condition': handler.condition,
            'temperature': handler.temperature,
            'wind': handler.wind,
        }
    }

# Test
URL = 'https://api.weatherapi.com/v1/current.xml?key=b4e8f86b44654e6b86885330242207&q=Beijing&aqi=no'

with request.urlopen(URL, timeout=4) as f:
    data = f.read()

result = parseXml(data.decode('utf-8'))
assert result['city'] == 'Beijing'
print(result)

This code defines a SAX handler for weather data and verifies that the parsed city matches "Beijing."