Mailman 3 July 2010 - lxml - The Python XML Toolkit

[lxml-dev] reading xml document
by sakshi chawla July 29, 2010

July 29, 2010

hi how can I read the attributes from xml file?

3 2

[lxml-dev] Virtual elements (python)
by donn July 29, 2010

July 29, 2010

Hi, I am hacking slowly on an SVG editor in Python and I am using lxml which is making it all possible. I would like to use many svg trees at once, but not in one big tree; rather in a 'forest' with each one rooted alone. Elements in one can refer to elements in another, by way of the <use> tag. Is there any way to make a 'virtual' element, such that it forwards its children on-to another element (in any other Tree)? * I mean something like this: Root1 has element W with children: e1 (mapped-to:Root2/e2), e5, bob, sally Root2 has element e2 (id="e2_id") e2 has children: c2, c3 Thus: >>> list(e1) [c2,c3] >>> e1.attrib.get('id') 'e2_id' >>> list(W) [e1,e5,bob,sally] >>> e1.getparent() <element W> #it 'knows' its actual Tree >>> for elem in etree.iterwalk(Root1): print elem W...e1...[now into Root2] c2,c3... [back into Root1] e5, etc. >>> etree.tostring(W) "<tags from W><tags from c2><tags from c3><tags from e5>" To top it off, can xpaths be used across such a system? So I'd say 'gimme all the elements with attrib=xyz in Root1' and it will *also* skip along into Root2 by way of that virtual element? \d (* I guess I would have to walk the trees and identify the elements that are tag=="use" and then replace them with this fancy element.)

2 2

[lxml-dev] text nodes... again
by Hervé Cauwelier July 28, 2010

July 28, 2010

Hi, I'm having difficulties working on the kind of XML used by OpenDocument. Consider that example from the tutorial: <html><body>Hello<br/>World</body></html> An application on top of OpenDocument would be to "strip" <br/> from a document. Using "Element.remove()", I lose "World". What pattern of code would you recommend to merge "Hello" and "World"? Yes, it would be glued like "HelloWorld", don't worry. The FAQ[1] mentions the issue: "A good way to deal with this is to use helper functions that copy the Element without its tail." In my case, it would be "remove the Element without its tail". Are these helper functions shipped with lxml? Is there some cookbook online? Thanks in advance, Hervé Cauwelier [1] http://codespeak.net/lxml/FAQ.html#what-about-that-trailing-text-on-seriali…

2 2

[lxml-dev] lxml 2.2.7 eggs for MacOS X 10.4 (and up?)
by "Pascal Oberndörfer" July 26, 2010

July 26, 2010

Dear all, Some lxml 2.2.7 eggs for Mac OS X have been added here: <http://www.urheberrecht.org/sw/python/lxml/10.4/lxml22/> Regards, Pascal

1 0

[lxml-dev] lxml 2.2.7 and 2.3alpha2 released
by Stefan Behnel July 25, 2010

July 25, 2010

Hi all, I just pushed lxml 2.2.7 and 2.3alpha2 to PyPI. They are bug-fix-only releases that fix a crash showing mostly in XSLT. http://pypi.python.org/pypi/lxml/2.2.7 http://codespeak.net/lxml/ http://pypi.python.org/pypi/lxml/2.3alpha2 http://codespeak.net/lxml/dev/ The 2.2 release series continues to be built with Cython 0.11.3, whereas the 2.3 release was built with the close-to-final Cython 0.13beta0. Have fun, Stefan 2.2.7 (2010-07-24) ================== Bugs fixed ---------- * Crash in XSLT when generating text-only result documents with a stylesheet created in a different thread. 2.3alpha2 (2010-07-24) ====================== Features added -------------- Bugs fixed ---------- * Crash in XSLT when generating text-only result documents with a stylesheet created in a different thread. Other changes -------------- * ``repr()`` of Element objects shows the hex ID with leading 0x (following ElementTree 1.3).

2 4

[lxml-dev] Xpath function converts int to float
by Frank Millman July 21, 2010

July 21, 2010

Hi all As the subject line states, an xpath function that should return an int actually returns a float. Here is a simple example - --------------------- from lxml import etree obj = {'one':1, 'two':2} def get_obj(context, param): return obj[param] ns = etree.FunctionNamespace(None) ns['get_obj'] = get_obj xml = "<root><obj>get_obj('one')</obj></root>" elem = etree.fromstring(xml) ob1 = elem.find('obj') ob2 = ob1.xpath(ob1.text) print ob2 --------------------- This prints 1.0 . Is this a bug? Is there a workaround? Thanks Frank Millman

2 2

[lxml-dev] lxml.etree.XMLSyntaxError reported on a valid xml
by Sylvain DUVILLARD July 19, 2010

July 19, 2010

Hi ! I get the following error when parsing an xml file with objectify on this call : objectify.fromstring(etree.tostring(raw_doc), parser) ... File "lxml.objectify.pyx", line 1826, in lxml.objectify.fromstring (src/lxml/lxml.objectify.c:18625) File "lxml.etree.pyx", line 2532, in lxml.etree.fromstring (src/lxml/lxml.etree.c:48634) File "parser.pxi", line 1545, in lxml.etree._parseMemoryDocument (src/lxml/lxml.etree.c:72245) File "parser.pxi", line 1424, in lxml.etree._parseDoc (src/lxml/lxml.etree.c:71106) File "parser.pxi", line 938, in lxml.etree._BaseParser._parseDoc (src/lxml/lxml.etree.c:67875) File "parser.pxi", line 539, in lxml.etree._ParserContext._handleParseResultDoc (src/lxml/lxml.etree.c:64257) File "parser.pxi", line 625, in lxml.etree._handleParseResult (src/lxml/lxml.etree.c:65178) File "parser.pxi", line 565, in lxml.etree._raiseParseError (src/lxml/lxml.etree.c:64521) lxml.etree.XMLSyntaxError: Element '{http://www.spiritconsortium.org/XMLSchema/SPIRIT/1.4}remapPort': No match found for key-sequence ['phantom_remap'] of keyref '{http://www.spiritconsortium.org/XMLSchema/SPIRIT/1.4}remapStatePortRef'. The parser was made using a schema : component_schema = etree.XMLSchema(file='component.xsd') component_parser = objectify.makeparser(schema = component_schema, remove_blank_text = True) Now, when I use xmllint stand alone like this, it works : xmllint --noout --schema component.xsd Vendor_Library_adec_1.0.xml Vendor_Library_adec_1.0.xml validates I'm using the latest version of all released modules : >>> lxml.etree: (2, 2, 6, 0) >>> libxml used: (2, 7, 7) >>> libxml compiled: (2, 7, 7) >>> libxslt used: (1, 1, 26) >>> libxslt compiled: (1, 1, 26) Should I enter a bug or do I do something obviously wrong here ? I also could validate the file against the schema using xmlspy Thanks in advance Sylvain

1 0

[lxml-dev] Quick question about objectify
by Frank Millman July 17, 2010

July 17, 2010

Hi all Sometimes an element may or may not exist in a document. If it exists, I need to perform some processing on it. Using etree and xpath, the following is convenient - for obj in root.xpath('bp:object', namespaces=nsmap): [do something] If there are no 'object's, an empty list is returned, so it works. I tried the same with objectify - for obj in root.object: [do something] If there are no 'object's, AttributeError is raised. This means that I must either precede this with 'if hasattr(...)', or wrap it in a try/except. I have to do this quite a lot, so the code does not flow as nicely. Is there any way to use objectify and retain the convenience of xpath? Thanks Frank Millman

2 1

[lxml-dev] LXML Windows builds for Python 2.7
by Jason R. Coombs July 13, 2010

July 13, 2010

No sooner does Sidnei get the builds out for Python 2.6 (thanks!) as I need a build for Python 2.7. I was able to build lxml for Python 2.7 on a 32-bit platform using this script: http://dl.dropbox.com/u/54081/linked/lxml-build.py Comments are welcome. The script also has support for building on a 64-bit platform using binaries I found built for PHP, though I haven't yet got that to compile (it seems they've used optimizations for VS2008 SP1, so I'm applying that SP now). I'm pleased to report that thusfar, the tentative builds for Python 2.6 are working for me. Please consider also building against Python 2.7 when releasing to PyPI. Thanks, Jason

1 0

Re: [lxml-dev] building lxml from source using Mingw32
by Christian Heimes July 12, 2010

July 12, 2010

> Ok, thank for you suggestion, I don't know that. I have used it because > it was the only download that I have found on the MS website. > My first idea was to built lxml from source using Mingw, but, it was > too complicated for my skills (if I have well understand the procedure I > must convert all DLL in the required packages using a program called > pexports). It's tricky to compile extensions with MinGW32. You have to use an unsupported version in order to get proper C++ support and event that isn't ABI compatible with MSVC++ (AFAIK). > Do you know if is it possible to have a free version of the Visual C++ 2008? > Sandro Here you are: http://www.microsoft.com/express/downloads/#2008-All I hightly recommand the DVD ISO. It's larger but it will work after Microsoft has removed VS EE 2008 from its site. MS tends to remove old Visual Studio stuff after the new version has been released. The express edition can't build X64 binaries but it's possible to integrate the 64bit compiler from the Windows 7 SDK. Please don't ask how. I haven't tried it because I'm in the lucky position to have a free MSDN membership for my Python open source work. ;) Christian

3 2