5.2. XML Module lxml

5.2. XML Module lxml — Python

pip install lxml

5.2.1. Creating elements

Creating elements:

from lxml.etree import tostring, Element


root = Element("iris")

print(tostring(root))
# b'<iris/>'

Adding elements using list interface:

from lxml.etree import tostring, Element


root = Element('iris')
root.append(Element('setosa'))
root.append(Element('versicolor'))
root.append(Element('virginica'))

print(tostring(root))
# b'<iris><setosa/><versicolor/><virginica/></iris>'

5.2.2. Length of a subtree

Length of a subtree:

from lxml.etree import Element


root = Element('iris')
root.append(Element('setosa'))
root.append(Element('versicolor'))
root.append(Element('virginica'))

print(len(root))
# 3

5.2.3. Selecting subtree

Selecting subtree:

from lxml.etree import Element


root = Element('iris')
root.append(Element('setosa'))
root.append(Element('versicolor'))
root.append(Element('virginica'))

selected = root[2]
print(selected.tag)
# virginica

Where is selected element:

from lxml.etree import Element


root = Element('iris')
root.append(Element('setosa'))
root.append(Element('versicolor'))
root.append(Element('virginica'))

selected = root[1]
root.index(selected)
# 1

selected = root[2]
root.index(selected)
# 2

5.2.4. Element tree as a lists

Elements are lists:

from lxml.etree import tostring, Element


root = Element('iris")
root.append(Element('setosa"))
root.append(Element('versicolor"))
root.append(Element('virginica"))

children = list(root)
print(children)
# [
#     <Element setosa at 0x113cd4048>,
#     <Element versicolor at 0x113cd4188>,
#     <Element virginica at 0x113cd41c8>
# ]

Iterating over elements:

from lxml.etree import Element


root = Element("iris")
root.append(Element("setosa"))
root.append(Element("versicolor"))
root.append(Element("virginica"))

for child in root:
    print(child.tag)

# setosa
# versicolor
# virginica

Slicing elements:

from lxml.etree import Element


root = Element("iris")
root.append(Element("setosa"))
root.append(Element("versicolor"))
root.append(Element("virginica"))

root.insert(0, Element("arctica"))

start = root[:1]
end = root[-1:]

print(start[0].tag)  # arctica
print(end[0].tag)    # virginica

5.2.5. Elements as a dict

Create element using dict interface:

from lxml.etree import tostring, Element


tag = Element("iris", kingdom="plantae")

print(tostring(tag))
# b'<iris kingdom="plantae"/>'

Get element attributes and values:

from lxml.etree import tostring, Element


tag = Element("iris", kingdom="plantae")

print(tag.get("kingdom"))          # plantae
print(tag.get("not-existing"))     # None

Set element attributes and values:

from lxml.etree import tostring, Element


tag = Element("iris", kingdom="plantae")
tag.set("kind", "flower")

print(tag.get("kind"))
# flower

print(tostring(tag))
# b'<iris kingdom="plantae" kind="flower"/>'

Elements carry attributes as a dict:

from lxml.etree import Element


tag = Element("iris", kingdom="plantae")
tag.set("kind", "flower")

tag.keys()
# ['kind', 'kingdom']

tag.values()
# ['plantae', 'flower']

tag.items()
# [('kingdom', 'plantae'), ('kind', 'flower')]

Iterating over element attributes and values:

from lxml.etree import Element


tag = Element("iris", kingdom="plantae")
tag.set("kind", "flower")

for key, value in tag.items():
    print(f'{key} -> {value}')

# kingdom -> plantae
# kind -> flower

Elements carry attributes as a dict:

from lxml.etree import Element


tag = Element("iris", kingdom="plantae")
tag.set("kind", "flower")

tag.attrib['kingdom']
# 'plantae'

tag.attrib['not-existing']
# Traceback (most recent call last):
# KeyError: 'not-existing'

tag.attrib['species'] = 'Setosa'
tag.attrib.get('species')
# 'Setosa'

tag.attrib
# {'kingdom': 'plantae', 'kind': 'flower'}

tag.attrib.items()
# [('kingdom', 'plantae'), ('kind', 'flower'), ('species', 'Setosa')]

5.2.6. Elements contain text

from lxml.etree import tostring, Element

tag = Element("iris")
tag.text = "Setosa"

tag.text
# 'Setosa'

tostring(tag)
# b'<iris>Setosa</iris>'

5.2.7. Tree iteration

from lxml.etree import tostring, Element, SubElement

root = Element("iris")
SubElement(root, "species").text = "Setosa"
SubElement(root, "species").text = "Virginica"
SubElement(root, "flower").text = "Versicolor"

print(tostring(root, pretty_print=True))
# b'<iris>
#       <species>Setosa</species>
#       <species>Virginica</species>
#       <flower>Versicolor</flower>
# </iris>'


for element in root.iter():
    print(f'{element.tag} -> {element.text}')

# iris -> None
# species -> Setosa
# species -> Virginica
# flower -> Versicolor


for element in root.iter("species"):
    print(f'{element.tag} -> {element.text}')

# species -> Setosa
# species -> Virginica


for element in root.iter("species", "flower"):
    print(f'{element.tag} -> {element.text}')

# species -> Setosa
# species -> Virginica
# flower -> Versicolor

5.2.8. Entities

from lxml.etree import tostring, Element, SubElement, Entity

root = Element("iris")
print(tostring(root))
# b'<iris/>'

root.append(Entity("#234"))
print(tostring(root))
# b'<iris>&#234;</iris>'

5.2.10. Serialization

from lxml.etree import tostring, XML


root = XML('<root><a><b/></a></root>')

tostring(root)
# b'<root><a><b/></a></root>'

print(tostring(root, xml_declaration=True))
# b"<?xml version='1.0' encoding='ASCII'?>\n<root><a><b/></a></root>"

print(tostring(root, encoding='utf-8'))
# b'<root><a><b/></a></root>'

print(tostring(root, encoding='iso-8859-2'))
# b"<?xml version='1.0' encoding='iso-8859-2'?>\n<root><a><b/></a></root>"

print(tostring(root, pretty_print=True))
# b'<root>\n  <a>\n    <b/>\n  </a>\n</root>\n'

print(tostring(root, pretty_print=True).decode())
# <root>
#   <a>
#     <b/>
#   </a>
# </root>

from lxml.etree import tostring, XML

root = XML('<html><head/><body><p>Hello<br/>World</p></body></html>')

# default: method = 'xml'
tostring(root)
# b'<html><head/><body><p>Hello<br/>World</p></body></html>'

tostring(root, method='xml')
# b'<html><head/><body><p>Hello<br/>World</p></body></html>'

tostring(root, method='html')
# b'<html><head></head><body><p>Hello<br>World</p></body></html>'

print(tostring(root, method='html', pretty_print=True))
# b'<html>\n<head></head>\n<body><p>Hello<br>World</p></body>\n</html>\n'

print(tostring(root, method='html', pretty_print=True).decode())
# <html>
# <head></head>
# <body><p>Hello<br>World</p></body>
# </html>

tostring(root, method='text')
# b'HelloWorld'