Source code for skosprovider.utils

# -*- coding: utf-8 -*-
This module contains utility functions for dealing with skos providers.

from __future__ import unicode_literals

import sys

from skosprovider.skos import (

from xml.dom.minidom import Node, Element
import html5lib

PY3 = sys.version_info[0] == 3

if PY3:  # pragma: no cover
    binary_type = bytes
else:  # pragma: no cover
    binary_type = str

[docs]def dict_dumper(provider): ''' Dump a provider to a format that can be passed to a :class:`skosprovider.providers.DictionaryProvider`. :param skosprovider.providers.VocabularyProvider provider: The provider that wil be turned into a `dict`. :rtype: A list of dicts. .. versionadded:: 0.2.0 ''' ret = [] for stuff in provider.get_all(): c = provider.get_by_id(stuff['id']) labels = [l.__dict__ for l in c.labels] notes = [n.__dict__ for n in c.notes] sources = [s.__dict__ for s in c.sources] if isinstance(c, Concept): ret.append({ 'id':, 'uri': c.uri, 'type': c.type, 'labels': labels, 'notes': notes, 'sources': sources, 'narrower': c.narrower, 'broader': c.broader, 'related': c.related, 'member_of': c.member_of, 'subordinate_arrays': c.subordinate_arrays, 'matches': c.matches }) elif isinstance(c, Collection): ret.append({ 'id':, 'uri': c.uri, 'type': c.type, 'labels': labels, 'notes': notes, 'sources': sources, 'members': c.members, 'member_of': c.member_of, 'superordinates': c.superordinates, 'infer_concept_relations': c.infer_concept_relations }) return ret
[docs]def extract_language(lang): ''' Turn a language in our domain model into a IANA tag. .. versionadded:: 0.7.0 ''' if lang is None: lang = 'und' # return undefined code when no language else: lang = text_(lang, encoding="UTF-8") return lang
[docs]def text_(s, encoding='latin-1', errors='strict'): """ If ``s`` is an instance of ``binary_type``, return ``s.decode(encoding, errors)``, otherwise return ``s``""" if isinstance(s, binary_type): return s.decode(encoding, errors) return s
[docs]def add_lang_to_html(htmltext, lang): ''' Take a piece of HTML and add an xml:lang attribute to it. .. versionadded:: 0.7.0 ''' if lang == 'und': return htmltext parser = html5lib.HTMLParser( tree=html5lib.treebuilders.getTreeBuilder("dom") ) html = parser.parseFragment(htmltext) html.normalize() if len(html.childNodes) == 0: return '<div xml:lang="%s"></div>' % lang elif len(html.childNodes) == 1: node = html.firstChild if node.nodeType == Node.TEXT_NODE: div = Element('div') div.ownerDocument = html div.setAttribute('xml:lang', lang) div.childNodes = [node] html.childNodes = [div] else: node.setAttribute('xml:lang', lang) else: #add a single encompassing div div = Element('div') div.ownerDocument = html div.setAttribute('xml:lang', lang) div.childNodes = html.childNodes html.childNodes = [div] return html.toxml()
def _df_writexml(self, writer, indent="", addindent="", newl=""): ''' Monkeypatch method for unexisting `writexml` in :class:`xml.dom.minidom.DocumentFragment`. .. versionadded:: 0.7.0 ''' # indent = current indentation # addindent = indentation to add to higher levels # newl = newline string if self.childNodes: if (len(self.childNodes) == 1 and self.childNodes[0].nodeType == Node.TEXT_NODE): self.childNodes[0].writexml(writer, '', '', '') else: for node in self.childNodes: node.writexml(writer, indent+addindent, addindent, newl) from xml.dom.minidom import DocumentFragment DocumentFragment.writexml = _df_writexml