Collective.vdexvocabulary with Metadata

h2o · April 10, 2020, 3:10am

According to the specifications, Vdex XML files may include metadata. I understand this to mean that terms may contain attributes. Is my understanding correct? If so, does anyone have a sample vdex file with metadata or a project/module that uses it?

The c.vdexvocabulary module relies on imsvdex which has methods (getVocabMetadata and getTermMetadataById) for getting the metadata. These methods return lists of lxml.etree._Element types. However, I get 'None' using lxml find and xpath.

My vdex:

<?xml version="1.0" encoding="utf-8"?>
<vdex xmlns="http://www.imsglobal.org/xsd/imsvdex_v1p0" orderSignificant="false" language="en">
    <vocabIdentifier>collective.vdexvocabulary.languages</vocabIdentifier>

<term>
    <termIdentifier>english</termIdentifier>
    <caption>
        <langstring language="en">English</langstring>
        <langstring language="it">Inglese</langstring>
    </caption>
    <metadata>
        <attribute1>One</attribute1>
        <attribute2>Two</attribute2>
    </metadata>
</term>

My console code (vdex registered via c.vdexvocabulary vocabulary type):

>>>from zope.component import getUtility
>>>from zope.schema.interfaces import IVocabularyFactory
>>> factory = getUtility(IVocabularyFactory, 'collective.vdexvocabulary.languages')
>>>mdata = factory.vdex.getTermMetadataById('english')
>>>mdata
[<Element {http://www.imsglobal.org/xsd/imsvdex_v1p0}metadata at 0x7f8b064e0a00>]
>>>element = mdata[0]
>>>type(element)
<type 'lxml.etree._Element'>
>>>element.find('attribute1').text
File "<console>", line 1, in <module>
AttributeError: 'NoneType' object has no attribute ''text'

BTW, the vocabulary that I want to create is a standard geographic code. I want to add metadata like longitude and latitude, This is hardly updated so I figured it might be better if it is implemented as a vocabulary defined in a module.

Any leads would be highly appreciated.

dieter · April 10, 2020, 4:57am

Your XML document has an (implicitly) associated (XML) namespace (http://www.imsglobal.org/xsd/imsvdex_v1p0). Therefore, you cannot refer to elements with their local name. Try element.find('{http://www.imsglobal.org/xsd/imsvdex_v1p0}attribute1').text.

h2o · April 10, 2020, 6:31am

Thanks.

This is basic but I did not know.

jensens · April 12, 2020, 8:12am

The underlying imsvdex package supports fetching metadata from vocab, or term

github.com

collective/imsvdex/blob/master/imsvdex/vdex.py#L160




def getVocabName(self, lang=None):
    """
    returns the vocabulary name in the given (or default) language.
    If lang is '*', returns a dict of all translations keyed by language
    """
    xpath = "%s/%s" % (self.vdexTag("vocabName"), self.vdexTag("langstring"))
    captions = self.tree.findall(xpath)
    return self.getLangstring(captions, lang, default="(unnamed vocabulary)")


def getVocabMetadata(self):
    """
    returns the VDEX metadata element(s) for the Vocabulary
    """
    xpath = self.vdexTag("metadata")
    return self.tree.getroot().findall(xpath)


def getVocabWildcard(self, foreign_ns, tagname):
    """
    returns 'wildcard' element(s) (with a foreign namespace) for the Vocabulary
    """

github.com

collective/imsvdex/blob/master/imsvdex/vdex.py#L301


    term = self.getTermById(identifier)
    return self.getTermCaption(term, lang)


def getTermDescriptionById(self, identifier, lang=None):
    """
    returns the description(s) for a given term identifier
    """
    term = self.getTermById(identifier)
    return self.getTermDescription(term, lang)


def getTermMetadataById(self, identifier):
    """
    returns the VDEX metadata element(s) inside a term
      for a given term identifier
    """
    term = self.getTermById(identifier)
    xpath = self.vdexTag("metadata")
    return term.findall(xpath)


def getTermWildcardById(self, identifier, namespace, tagname):
    """

but AFAIK the zope.schema.vocabulary and its terms nor the collective.vdexvocabulary and it's vdexterm does not support metadata.

github.com

collective/collective.vdexvocabulary/blob/master/src/collective/vdexvocabulary/term.py

from zope.interface import directlyProvides
from zope.interface import implements
from zope.schema.interfaces import ITitledTokenizedTerm
from zope.schema.interfaces import ITokenizedTerm
from zope.schema.vocabulary import getVocabularyRegistry


class VdexTerm(object):
    """Vdex tokenized term used by VdexVocabulary."""

    implements(ITokenizedTerm)

    def __init__(self, value, token=None, title=None, description=None, related=[]):
        self.value = value
        if token is None:
            token = value
        if isinstance(token, unicode):
            self.token = token.encode("utf-8")
        else:
            self.token = str(token)

This file has been truncated. show original

So, this is something to be added. Pull Requests appreciated.