Mailman 3 August 2005 - lxml - The Python XML Toolkit

[lxml-dev] lxml relaxng support
by Carlos Pita 09 Aug '05

09 Aug '05

Hi all! I started using lxml because of its relaxng support. But: 1) does it parse relaxng compact syntax? 2) is there any way to get better error report (besides ok/error)? Btw, I've built win32 lxml binaries (which uses the libxml binaries located at http://www.zlatkovic.com/libxml.en.html). If anyone of you needs them just email me. Thank you in advance. Regards, Carlos ___________________________________________________________ 1GB gratis, Antivirus y Antispam Correo Yahoo!, el mejor correo web del mundo http://correo.yahoo.com.ar

1 0

[lxml-dev] .find() evals to false
by dharana 09 Aug '05

09 Aug '05

-- code snippet ------------------------------------------------------ ~ $ python Python 2.4.1 (#1, Jun 2 2005, 01:24:02) [GCC 2.95.4 20011002 (Debian prerelease)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> class foo(object): ... pass ... >>> obj1 = foo() >>> if obj1: ... print 'obj1 doesn\'t evals to False' ... obj1 doesn't evals to False >>> import lxml.etree >>> mystr = '<?xml version="1.0"?><root><child /></root>' >>> xmldoc = lxml.etree.fromstring(mystr) >>> xmldoc.find('child') <Element child at -481b6620> >>> if xmldoc.find('child'): ... print 'Element child doesn\'t evals to False' ... >>> ---------------------------------------------------------------------- I've been hit by this several times by now. I think that an element like a node shouldn't eval to false. What do you think about this? -- dharana

3 5

[lxml-dev] Suggestion: adding has_key() method to etree._Element.attrib
by dharana 09 Aug '05

09 Aug '05

I wanted to know if a given element had a given attribute. I first tried with: if node.has_key(attrib_node): but I found out that _Element don't have this method so I tried with: if attrib_name in node.attrib: and it worked. Would it be ok to also have a has_key() method? -- dharana

3 4

[lxml-dev] test_xpath_number fails
by Hamish Lawson 08 Aug '05

08 Aug '05

When I run "make test" I get the following error: **** 270/300 ( 90.0%): Doctest: lxml.tests.test_xpathevaluator.xpath ====================================================================== FAIL: test_xpath_number (lxml.tests.test_xpathevaluator.ETreeXPathTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/local/python/lib/python2.4/unittest.py", line 260, in run testMethod() File "/webapps/local/src/python-modules/lxml/src/lxml/tests/test_xpathevaluator.py", line 20, in test_xpath_number self.assertEquals('nan', str(tree.xpath('number(/a)'))) File "/usr/local/python/lib/python2.4/unittest.py", line 333, in failUnlessEqual raise self.failureException, \ AssertionError: 'nan' != 'NaN' ---------------------------------------------------------------------- Ran 270 tests in 0.605s FAILED (failures=1) **** I'm using lxml 0.7, libxml2 2.6.16, libxslt 1.1.2, Python 2.4, Solaris 9. Hamish Lawson

3 5

[lxml-dev] Trying to copy.copy() a node causes a segmentation fault
by dharana 05 Aug '05

05 Aug '05

Trying to copy an xml node causes a segfault: >> import copy >> >> xmlsection xmlnode.find('sections').findall('section')[0]: >> original_section = copy.copy(xmlsection) Segmentation fault -- dharana

3 4

Re: [lxml-dev] No output for tree.write
by Hamish Lawson 05 Aug '05

05 Aug '05

Hello Martijn > Very strange indeed.. Since I do not have access to a Solaris machine > I'm going to > need some help debugging this one. Can you write to a StringIO()? What > does it do if you use etree.tostring()? I can successfully write to a StringIO() and to a disk file. etree.tostring() also returns the text of the document as expected. Hamish

2 2

[lxml-dev] API incompleteness, tree modification
by Slou 04 Aug '05

04 Aug '05

Talking about incompleteness There is no easy way to insert element relative to the given one. There are two ways to do it: 1) use insert(idx, elmnt) method, which needs integer argument idx, then _Element should have index() method 2) expose libxml2 API functions as addNextSibling (or smth like that) and add functions insert_before(base, elmt), insert_after(base, elmt), replace(base, elmt) second way is preferred cause it's a lot more pythonic (it doesn't involve entity 'numerical index' in this operation). my Pyrex skills are not that good, that's why I implemented the first way. here is my implementation: === line 455 === # start, stop parameters make no sense, cause all elements are unique def index(self, _Element x): cdef int k cdef xmlNode* c_child k = 0 c_child = self._c_node.children while c_child is not NULL: if _isElement(c_child): if c_child is x._c_node: return k else: k = k + 1 c_child = c_child.next raise ValueError === (I can not make normal patch to SVN because Windows version of SVN client can not do checkout, I do not know why). I will be happy to see second way implemented.

2 5

[lxml-dev] Named entities ignored
by Hamish Lawson 04 Aug '05

04 Aug '05

Are named entities ignored by lxml? Even with a suitable DTD declared, the following program does not output the copyright sign; it does, however, if it's provided as a numeric entity. **** from lxml import etree from StringIO import StringIO import sys source = """\ <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html> <p>©2005</p> </html> """ infile = StringIO(source) tree = etree.parse(infile) tree.write(sys.stdout) **** Hamish Lawson

2 3

[lxml-dev] lxml on windows action
by Martijn Faassen 04 Aug '05

04 Aug '05

Hi there, Before the lxml 0.8 release, I'd like to get the windows build and release process in place. So far I've heard about 3 or 4 people who got it built on Windows, but I'd like an official version that can be hosted on the lxml website that we can update for each release. I'm not able to maintain this myself, so this is a request for a regular maintainer. Here is what I think are the requirements: * should be easy to build. That means the setup.py in lxml should be adjusted so that typing in the right command will build the right thing on windows. * should include libxml2 + libxslt so that users do not have to go out and install the right versions of that. These versions should ideally not conflict with other versions of those libraries that may already be installed. Does not need to contain the included libxml2/libxslt bindings though; lxml is supposed to replace these. * should be easy to install on Windows. * it should be easy to run the tests on Windows (I heard there were some problems with getting it going). * whenever I pull a release, I need a specific person who I can contact so that I can make sure the windows package for this release will be created as well. This way, I can put that online as well on the lxml website at release time or shortly after release, next to the source release. Opinions? Volunteers to work on aspects of this? Regards, Martijn

4 4

[lxml-dev] Typo on installation Web page?
by Gavrie Philipson 04 Aug '05

04 Aug '05

Hi, On the installation Web page at http://codespeak.net/lxml/installation.html, the requirements mention: "libxml 2.2.16 (newer versions should work)". I think this should be 2.6.16 and not 2.2.16, since I had no success compiling with 2.5.10, but it works fine with 2.6.19. -- Gavrie

2 1