Mailman 3 March 2026 - lxml - The Python XML Toolkit

baffled by namespace problem in xpath calls
by Martin Mueller March 11, 2026

March 11, 2026

I am probably missing something trivial, but I am completely baffled by the use of namespaces when using xpath with lxml. I have a template that creates a tree from an XML document and passes it to a function for processing at the w level. If I change the root element of a TEI document into ABC (turning it into a generic XML document),the following code works: for w in tree.xpath('//w'): print(etree.tostring(w)) This shouldn't work in a TEI document, because the xpath method requires a qualified name or namespace. I assumed that the full tag name {http://www.tei-c.org/ns/1.0} might work: for w in tree.xpath('http://www.tei-c.org/ns/1.0}w'): print(etree.tostring(w)) But it doesn't work. Neither does a special xmlns:tei declaration. So I am totally baffled. For ordinary lxml processing the ordinary mapping of the URI to tei would work. But it fails for xpath. The rather terse lxml manual was no help. I'll be grateful for any help.

3 3

specifying conditions for removing the tags of a particular elemenet
by Martin Mueller March 9, 2026

March 9, 2026

I'm having trouble with the etree.strip_tags function. I have a large TEI corpus with <w> elements for every word token. There are a lot of short <hi> element, and I would like to get rid of them and move the <hi> markers into @rend attributes, which simplifies processing at the level beyond <w> elements. I would like to 1. mark all the <w> children of <hi> elements with less than four children with appropriate @rend attributes 2. use the strip_tags function to get rid of <hi> elements with less than four children I know how to get rid of all <hi> elements with something like def process_tree(tree,filename, item): for div in tree.iter(tei +'div'): for el in div.iter('*'): etree.strip_tags(el, tei + 'hi') newname = os.path.join(allerlei, item) tree.write(newname, encoding='utf-8') But something like for el in div.iter('*'): if len(el) < 4: etree.strip_tags(el, tei + 'hi') does not work. Is it not possible to specify constraints on the strip_tags function or do I not know what I am doing? The latter very likely, and I will be grateful for any advice Martin Mueller Professor emeritus of English and Classics Northwestern University

3 2

Questions about XSLT (XSL Transformations)
by Schimon Jehudah March 5, 2026

March 5, 2026

Greetings. I am interested to know more of the XSLT component of the module LXML. Is this XSLT component an own XSLT implementation, or is it a so called "binding" with another software (e.g. libxslt)? I ask because, the XSLT component of Python appears to fail, whereas Saxon JS suceeds. So, I am interested to improve the XSLT component of LXML. Regards, Schimon

2 2