Mailman 3 September 2011 - lxml - The Python XML Toolkit

Fwd: Version 2.3.1 with Python 2.5 windows
by Brandon Goldfedder 28 Sep '11

28 Sep '11

All, Hitting some issues and thought I would ask. Is there a static windows build for Version 2.3.0 or 2.3.1 with Python 2.5 (need to do some Schematron stuff). easy_install doesn't seem to think so. I tried some steps to setup/build manually but they keep going boom (I tried to use the libxml2 and libxslt directly but think I'm missing a step there as well). If anyone has a pointer to a static windows build, or can point me to the step I am missing in getting it to manually setup it would greatly help. I'm trying to avoid moving from Python2.5. Thanks, Brandon Goldfedder

1 0

Multiple input documents
by Evgeny Turnaev 27 Sep '11

27 Sep '11

Hi. My question if related to XSLT document() function and processing multiple input documents in XSLT. Currently in our application fetches 3 to 7 separate xml documents merges all of them into single tree using append() or SubElement and passes merged tree into XSLT transformation. Is it possible in lxml to pass multiple trees into XSLT transformation and access them for example using document() function? If so then: will a document accessed by document() function be parsed for each access? Will i have to save document to disk to be able to load in in document() function or i can load already parsed tree from memory? Will it be faster to use document() than appending 3-4 of 40kb xml trees and 4-5 small (1kb)? One other reason why i am asking it: we have a lot of merging of the same tries (<1kb) into different documents and a few merging of 40kb tries. So i thinked: why cant lxml use the same tree using document() instead of explicitly appending it into each xml before transformation. Is there any other any other performance hints? -- -------------------------------------------- Турнаев Евгений Викторович +7 906 875 09 43 --------------------------------------------

2 1

How does lxml locate a schema for validation?
by Robert Withrow 27 Sep '11

27 Sep '11

Given a XML schema instance document like that represented by this fragment: <purchaseOrder xmlns="http://www.ibm.com" xsi:schemaLocation="http://www.ibm.com PO.xsd"> <shipTo country="US"> The XML Schema Primer suggests the processor will use the xsi:schemaLocation as a hint to find the appropriate schema document to use for validating the instance. The XML Schema specification even lays out a "Schema Document Location Strategy" for finding the schema document. Does lxml do this? The documentation seems to only describe the manual process used to pass the "schema" argument to the parser creation. That doesn't seem tractable for cases where you have multiple schemas used in the instance.

2 3

xpath() finding nodes outside current element
by Andreas Jung 20 Sep '11

20 Sep '11

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi there, I am using the following code to turn all Hx tags into DIV tags contained with all DIVs with class 'ignore-headings-for-structure'. for div in CSSSelector('div.ignore-headings-for-structure')(root):$ for heading in div.xpath(xpath_query(('h1', 'h2', 'h3', 'h4', 'h5'))):$ level = int(heading.tag[1:])$ heading.tag = 'div'$ heading.attrib['class'] = 'heading-level-%d' % level$ However div.xpath('//h1') returns a H1 *outside* the current div: (Pdb) print lxml.html.tostring(div) <div class="beschreibung ignore-headings-for-structure"><p><span class="mw-headline" id="Design_orientiert_sich_am_Menschen">Design orientiert sich am Menschen</span><span class="editsection"> </span></p> <p>Design orientiert sich am Menschen und seinen vielfältigen Bedürfnissen. Diese Bedürfnisse reichen von körperlichen und psychischen Bedürfnissen bis hin zu Anforderungen des menschlichen Verstands an die gegenständliche Umwelt. Design folgt dabei nicht allein selbst gesetzten Regeln und Intentionen, sondern muss sich vor allem mit den Interessen jener Gruppen oder Personen auseinandersetzen, denen das Design dienlich sein soll. Dadurch ist Design und sind die Entwürfe vor allem zweckorientiert. In der Designtheorie wurde dafür der Begriff der Funktionalität geprägt. Nicht zuletzt durch seine Zweckorientierung unterscheidet sich Design von der Kunst.</p> <ol><li>a</li> <li>b</li> <li>c</li> <li>d</li> </ol></div> There is no <h1> tag contained. (Pdb) print 'h1' in lxml.html.tostring(div) False However searching for an element inside returns a H1 tag outside the div element - why? (Pdb) div.xpath('//h1') [<Element h1 at ef5be20>] (Pdb) lxml.html.tostring(div.xpath('//h1')[0]) '<h1 class="title">Lastenheft</h1>\n \n\n Andreas -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQGUBAEBAgAGBQJOeLHFAAoJEADcfz7u4AZjYzoLv0BFJ7JyDCpVbsyR7rNu0iyb PYvnI5npsIRrxWteUX1vS7CF8pmLTYmd9tzj1DnQogao8I9mSqIofbiVXw9zseeN 5C02E9mJcIMQYOX1peYsZeWfjNALK/Qw4JcqiFjbNUJ/U+EFWXWxCuQkKTL/cPhm G4CvWFzT6YVDS87RfSF908D8/sOM33adVOfzvqL7mvYyFBFK6yrdNgmRwo+PAa9C gPFdYdbo9hiZ2iVCRmrLPvb8tYjo40K4FEtft+cUHIr1MOpa3kqwi9U+mN7DGqw+ geJn65LcFDHsJN2+P+KkqxGXhBnzs3l/AJuWckBDoJVC7be4jlIug9RhWVHlpfyy xnV2Cl68HVtaEtI3vBRnUoEWlGGvwI6w+AZk+E/PKATEeG/w0TW+bBmROLuVsljI 10bcbxnkrXj0Aq/Jh8cWv9q+DybAzz+rSubvYctT7c/GpyFx9yY2eQEsdOA7KTXN udoFTrnyEWJgkft4/7tkymZXo5P5Zkc= =399C -----END PGP SIGNATURE-----

2 1

Custom Element interface, proxy problem
by Marc van Dijk 20 Sep '11

20 Sep '11

Dear all, I recently adopted lxml as the core of the flexible database for scientific data I'm developing. It uses custom element classes using the etree.AttributeBasedElementClassLookup. Although this work great I do have some problems with checking for the existence of class objects that get returned using this method. I will try to explain by example: Lets say I have a function that performs some queries on an XML database and returns the results for each query as a list of custom element classes. I want to collect all the individual results in one list checking for duplicates in the process. I can do this in two ways: 1) finallist = [] for element in otherlistst: if not element in finallist: finallist.append(element) 2) finallist = [] for list in otherlists: finallist.extend(list) finallist = set(finallist) The result for method 1 is a list that contains only one custom element class instance. This is not wat is expected. Method two works and results in a set (which can be converted to list) of unique element class instances. I would like to know why method 1 does not work and method 2 does. And if there is anything I can do to have method 1 work. Thanks, Marc

2 3

UnicodeDecodeError
by mbutubuntu 19 Sep '11

19 Sep '11

Hello folks, I'm relatively new to LXML but I've encountered a problem that I can't solve... I can't be sure if it's a BUG of mine or not. The error comes when I try to parse an HTML webpage to extract its title, on github gists you can found the HTML source of the page, the simple program and the Traceback, this is the link https://gist.github.com/1227169 I hope you've enough information to discover my problem, Best Regards, Fabio Buda Web Developer/Designer @ netdesign http://www.netd.it

1 0

lxml import error from django - apache with mod_wsgi on windows
by Ανδρέας Πορευόπουλος 19 Sep '11

19 Sep '11

Hi everybody, I don't know if you can help me, but I am facing a very peculiar problem. I am developing an application with django, and I am consuming a SOAP web service with suds and processing all XML input with lxml. Development is done on a linux machine and everything works fine, either with django's server or with apache and mod_wsgi. But the deployment target is a Windows machine. At the target I have installed django, python, apache and mod_wsgi. Also all my needed libs such as suds, lxml, south. When I am serving with apache I am getting an import error at my line with: from lxml import etree This only happens with apache and mod_wsgi. When i use the development server of django all is ok When I start a python shell import lxml gives no errors. This my print out of your script: Python : (2, 6, 6, 'final', 0) lxml.etree : (2, 3, 0, 0) libxml used : (2, 7, 7) libxml compiled : (2, 7, 7) libxslt used : (1, 1, 26) libxslt compiled : (1, 1, 26) Mind you that I have also tried python 2.7 with the same results. lxml was installed using easy_install My PYTHONPATH is correct, everything else (suds, south, django) gets imported with no problem (all of them are under my site-packages dir, and all was installed with easy_install) Any help would be much appreciated. Thank you Andreas

2 1

lxml 2.3.1 release?
by Laurence Rowe 13 Sep '11

13 Sep '11

When is the next lxml release planned? There are a couple of changes on trunk which would be helpful for Plone 4.2 which should hit final in the next couple of months. Laurence

2 1

img tag rendered with text although not contained with the img node
by Andreas Jung 11 Sep '11

11 Sep '11

I using lxml.html to find img nodes within a HTML document. In one case I have an image that is rendered with text outside <img> tag although the img node has no 'text' attribute set and has not children. Anything I am missing? -aj (Pdb) img <Element img at 1680e870> (Pdb) print img.text None (Pdb) lxml.html.tostring(img) '<img src="html.jpg" title="HTML ist die Grundlage des Internet" alt="Grafik: HTML-Männchen" style="float:left;padding:5px;" id="de023d5b-a636-4aee-b6c1-2f1e06e4f560">Eine weit verbreitete Meinung ist, dass es sehr schwierig sein muss, eine Homepage zu erstellen. Aber das stimmt nicht. ' (Pdb) img.text_content() '' (Pdb) img.text (Pdb) for x in img.iterchildren(): print x

2 1

Problem with undeclared namespaces
by Isaac Jurado 07 Sep '11

07 Sep '11

Hello, I have a script that parses FreeSWITCH XML-CDR records, and it is implemented as a very basic SAX parser: class CDRParser(): def __init__(self): ... def start(self, tag, attrib, nsmap): ... def data(self, text): ... def end(self, tag): ... def close(self): ... return a_dictionary parser = XMLParser(target=CDRParser(), remove_comments=True) cdrdict = XML(xmltext, parser) This has been working very nicely so far. However, we recently made some changes that resulted in some XML tags looking like this: <nolocal:originator>....</nolocal:originator> The problem is that thath "nolocal" namespace is not declared anywhere in the whole document, so lxml seems to rip it off from the tag names when invoking the sax events. We discovered that secret nsmap argument by looking briefly at the lxml code but the object does not seem to contain useful data. Is there a way to detect the presence of that "nolocal" namespace from the SAX events? It should be some solution that does not require any modification from the original XML text. It could be either some flag to disable namespace stripping or a way to declare the "nolocal" namespace from the parser code. Thanks in advance. -- Isaac Jurado Internet Busines Solutions eConcept

2 2