Argthtjtr

I'm trying to print an ElementTree using python 3.6. Here is a reproducible example of my code:

from xml.etree import ElementTree as ET

root = ET.Element('gpx')

el = ET.SubElement(root, 'test')

el.text = 0.3

print(ET.dump(root))

Error message is:

Traceback (most recent call last):

File "<stdin>", line 1, in <module>

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 1177, in dump

    elem.write(sys.stdout, encoding="unicode")

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 776, in write

    short_empty_elements=short_empty_elements)

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 941, in _serialize_xml

    short_empty_elements=short_empty_elements)

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 938, in _serialize_xml

    write(_escape_cdata(text))

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 1074, in _escape_cdata

    _raise_serialization_error(text)

File "/usr/local/lib/python3.6/xml/etree/ElementTree.py", line 1057, in _raise_serialization_error

    "cannot serialize %r (type %s)" % (text, type(text).__name__)

TypeError: cannot serialize 0.3 (type float)

Serializing a float type must be something very commonly done, buut I can't find a satisfactory answer on how to do this- what is the standard method?

Research:

I can find one question on stack overflow about this, but it suggests cooecing the float to a string,I need the output to be numeric.

There's an old discussion regarding this on the google forums, but this is 10 years old, and involves using the simplejson library- an extra library seems like overkill for this, especially when there is potentially a more modern solution

Interesting. I see that it is not going to work with a float because the _escape_cdata function uses the in operator (if "&" in text).

Also, the docstring for the text attribute indicates that it is either a string or None. The documentation, however, says that the "values are usually strings but may be any application-specific object", which I find misleading.

If you need to get other types when parsing an XML document, I recommend you use lxml.objectify

You must set text to a text value:

>>> from xml.etree import ElementTree as ET

... root = ET.Element('gpx')

... el = ET.SubElement(root, 'test')

... el.text = '0.3'

... print(ET.dump(root))

<gpx><test>0.3</test></gpx>

Simply put, text can only store text - there is no way to signal that a text node stores another type. Any other types must be serialised for storing, and de-serialised for usage.

Note that the float, being a primitive type, is still stored verbatim as the literal '0.3'. You can convert it back easily:

>>> print(el.text)

'0.3'

>>> print(float(el.text))

0.3

Keep in mind that XML is a text format - it can only store text. If you want other data types, you must define how the text is to be interpreted as the target type. For example, you can use an XML Schema Definition (XSD) to define a field as xs:integer. You can also embed a type marker in your texts - e.g. storing f0.3 or i7331 to mark floats and integers. However, all of these need a parser that is aware of such conventions, even for XSD.