Quadratic behavior in xml.etree.ElementPath Index Predicates

[encukou]

Element.findall() and fully-consumed Element.iterfind() exhibit O(n^2) time complexity when using XPath index predicates ([1], [last()], [last()-N]) on XML documents with many same-tag siblings.
Element.find() is only affected when the first match is near the end of the sibling list, such as with [last()] or [last()-N]; .//item[1] short-circuits after the first match.

(reported privately by @DarkaMaul)

Linked PRs

• gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates #152676
• [3.14] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154810
• [3.13] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154811
• [3.15] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154812
• [3.11] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154813
• [3.10] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154814
• [3.12] gh-152674: Avoid quadratic behavior in xml.etree.ElementPath index predicates (GH-152676) #154815