Convert Python ElementTree to string

Whenever I call ElementTree.tostring(e), I get the following error message:

AttributeError: 'Element' object has no attribute 'getroot'

Is there any other way to convert an ElementTree object into an XML string?

TraceBack:

Traceback (most recent call last):
  File "Development/Python/REObjectSort/REObjectResolver.py", line 145, in <module>
    cm = integrateDataWithCsv(cm, csvm)
  File "Development/Python/REObjectSort/REObjectResolver.py", line 137, in integrateDataWithCsv
    xmlstr = ElementTree.tostring(et.getroot(),encoding='utf8',method='xml')
AttributeError: 'Element' object has no attribute 'getroot'

Element objects have no .getroot() method. Drop that call, and the .tostring() call works:

xmlstr = ElementTree.tostring(et, encoding='utf8', method='xml')

You only need to use .getroot() if you have an ElementTree instance.

Other notes:

• This produces a bytestring, which in Python 3 is the bytes type.
  If you must have a str object, you have two options:

  1. Decode the resulting bytes value, from UTF-8: xmlstr.decode("utf8")

  2. Use encoding='unicode'; this avoids an encode / decode cycle:

     xmlstr = ElementTree.tostring(et, encoding='unicode', method='xml')
     

• If you wanted the UTF-8 encoded bytestring value or are using Python 2, take into account that ElementTree doesn't properly detect utf8 as the standard XML encoding, so it'll add a <?xml version='1.0' encoding='utf8'?> declaration. Use utf-8 or UTF-8 (with a dash) if you want to prevent this. When using encoding="unicode" no declaration header is added.

How do I convert ElementTree.Element to a String?

For Python 3:

xml_str = ElementTree.tostring(xml, encoding='unicode')

For Python 2:

xml_str = ElementTree.tostring(xml, encoding='utf-8')

The following is compatible with both Python 2 & 3, but only works for Latin characters:

xml_str = ElementTree.tostring(xml).decode()

---

Example usage

from xml.etree import ElementTree

xml = ElementTree.Element("Person", Name="John")
xml_str = ElementTree.tostring(xml).decode()
print(xml_str)

Output:

<Person Name="John" />

---

Explanation

Despite what the name implies, ElementTree.tostring() returns a bytestring by default in Python 2 & 3. This is an issue in Python 3, which uses Unicode for strings.

> In Python 2 you could use the str type for both text and binary data. > Unfortunately this confluence of two different concepts could lead to > brittle code which sometimes worked for either kind of data, sometimes > not. [...] > > To make the distinction between text and binary data clearer and more pronounced, [Python 3] made text and binary data distinct types that cannot blindly be mixed together.

^{Source: Porting Python 2 Code to Python 3}

If we know what version of Python is being used, we can specify the encoding as unicode or utf-8. Otherwise, if we need compatibility with both Python 2 & 3, we can use decode() to convert into the correct type.

For reference, I've included a comparison of .tostring() results between Python 2 and Python 3.

ElementTree.tostring(xml)
# Python 3: b'<Person Name="John" />'
# Python 2: <Person Name="John" />

ElementTree.tostring(xml, encoding='unicode')
# Python 3: <Person Name="John" />
# Python 2: LookupError: unknown encoding: unicode

ElementTree.tostring(xml, encoding='utf-8')
# Python 3: b'<Person Name="John" />'
# Python 2: <Person Name="John" />

ElementTree.tostring(xml).decode()
# Python 3: <Person Name="John" />
# Python 2: <Person Name="John" />

Thanks to Martijn Peters for pointing out that the str datatype changed between Python 2 and 3.

---

Why not use str()?

In most scenarios, using str() would be the "cannonical" way to convert an object to a string. Unfortunately, using this with Element returns the object's location in memory as a hexstring, rather than a string representation of the object's data.

from xml.etree import ElementTree

xml = ElementTree.Element("Person", Name="John")
print(str(xml))  # <Element 'Person' at 0x00497A80>

Non-Latin Answer Extension

Extension to @Stevoisiak's answer and dealing with non-Latin characters. Only one way will display the non-Latin characters to you. The one method is different on both Python 3 and Python 2.

Input

xml = ElementTree.fromstring('<Person Name="크리스" />')
xml = ElementTree.Element("Person", Name="크리스")  # Read Note about Python 2

> NOTE: In Python 2, when calling the toString(...) code, assigning xml with ElementTree.Element("Person", Name="크리스")will raise an error... > >UnicodeDecodeError: 'ascii' codec can't decode byte 0xed in position 0: ordinal not in range(128)

Output

ElementTree.tostring(xml)
# Python 3 (크리스): b'<Person Name="&#53356;&#47532;&#49828;" />'
# Python 3 (John): b'<Person Name="John" />'

# Python 2 (크리스): <Person Name="&#53356;&#47532;&#49828;" />
# Python 2 (John): <Person Name="John" />


ElementTree.tostring(xml, encoding='unicode')
# Python 3 (크리스): <Person Name="크리스" />             <-------- Python 3
# Python 3 (John): <Person Name="John" />

# Python 2 (크리스): LookupError: unknown encoding: unicode
# Python 2 (John): LookupError: unknown encoding: unicode

ElementTree.tostring(xml, encoding='utf-8')
# Python 3 (크리스): b'<Person Name="\xed\x81\xac\xeb\xa6\xac\xec\x8a\xa4" />'
# Python 3 (John): b'<Person Name="John" />'

# Python 2 (크리스): <Person Name="크리스" />             <-------- Python 2
# Python 2 (John): <Person Name="John" />

ElementTree.tostring(xml).decode()
# Python 3 (크리스): <Person Name="&#53356;&#47532;&#49828;" />
# Python 3 (John): <Person Name="John" />

# Python 2 (크리스): <Person Name="&#53356;&#47532;&#49828;" />
# Python 2 (John): <Person Name="John" />

I had the same problem in Python 3.8 and none of the previous answers solved it. The issue is that ElementTree is both the name of a module and of a class within it. Using an alias makes it clear:

from xml.etree.ElementTree import ElementTree
import xml.etree.ElementTree as XET
...
ElementTree.tostring(...)  # Attribute-error
XET.tostring(...)          # Works

If you just need this for debugging to see how the XML looks like, then instead of print(xml.etree.ElementTree.tostring(e)) you can use dump like this:

xml.etree.ElementTree.dump(e)

And this works both with Element and ElementTree objects as e, so there should be no need for getroot.

The documentation of dump says: > xml.etree.ElementTree.dump(elem) > > Writes an element tree or element structure to sys.stdout. This function should be used for debugging only. > > The exact output format is implementation dependent. In this version, it’s written as an ordinary XML file. > > elem is an element tree or an individual element. > > Changed in version 3.8: The dump() function now preserves the attribute order specified by the user.