What is the difference between a ElementTree and an Element? (python xml)

Question

from xml.etree.ElementTree import ElementTree, Element, SubElement, dump

elem = Element('1')
sub = SubElement(elem, '2')
tree = ElementTree(elem)

dump(tree)
dump(elem)

In the code above, dumping tree (which is an ElementTree) and dumping elem (which is an Element) results in the same thing. Therefore I am having trouble determining what the difference is between the two.

Answer 1

dumping tree (which is an ElementTree) and dumping elem (which is an Element) results in the same thing.

dump() function works the same for ElementTree and Element because it was intentionally made to behave this way:

def dump(elem):
    # debugging
    if not isinstance(elem, ElementTree):
        elem = ElementTree(elem)
    elem.write(sys.stdout)
    ...

I am having trouble determining what the difference is between the two.

ElementTree is a wrapper class that corresponds to the "entire element hierarchy" providing serialization functionality - dumping and loading the tree. Element, on the other hand, is a much "bigger" class that defines the Element interface.

Answer 2

The ElementTree wrapper class is used to read and write XML files [ref]. Most ElementTree apis are simple wrappers around the root Element [ref]. Simply put, ElementTree wraps the root Element (for convenience) and provides methods to serialize/deserialize the entire tree. Hence parse() belongs to ElementTree where iter() is a simple wrapper.

Then there are helper functions like iterparse and dump() in the xml.etree.ElementTree namespace. dump() writes a full xml doc to stdout [ref] whereas iterparse spits out Elements iteratively. Contrast parse(), which returns an xml.etree.ElementTree.ElementTree object and hence a complete hierarchy, to iterparse(), which returns an iterator[1].

1 There might be some confusion between xml.etree.ElementTree package namespace and xml.etree.ElementTree.ElementTree class name.