def test_update_sentences(self):
xml_content = self.xml_content1
sentences = [
Sentence('7e985ffb692bb6f617f25619ecca39a9'),
Sentence('7e985ffb692bb6f617f25619ecca3910')
]
for s in sentences:
s.pos_tags = 'nn nn'
s.significance = 3
s.sem_orient = 1
xml = XMLContent(xml_content)
print(xml.get_xml_document())
for sentence in xml.sentences:
print(sentence.md5sum, sentence.value, sentence.significance)
xml.sentences = sentences
xml_out = xml.get_xml_document()
for sentence in xml.sentences:
assert sentence.significance == 3
assert sentence.sem_orient == 1
assert 'CDATA' in xml_out
def test_update_sentences(self):
xml_content = self.xml_content1
sentences = [Sentence('7e985ffb692bb6f617f25619ecca39a9'),
Sentence('7e985ffb692bb6f617f25619ecca3910')]
for s in sentences:
s.pos_tags = 'nn nn'
s.significance = 3
s.sem_orient = 1
xml = XMLContent(xml_content)
print(xml.get_xml_document())
for sentence in xml.sentences:
print(sentence.md5sum, sentence.value, sentence.significance)
xml.sentences = sentences
xml_out = xml.get_xml_document()
for sentence in xml.sentences:
assert sentence.significance == 3
assert sentence.sem_orient == 1
assert 'CDATA' in xml_out
def from_api_dict(cls, api_dict):
'''
Parses a dict with a structure analoguous to the JSON format defined
in the API specification.
:param api_dict: The document to parse.
:type api_dict: dict
:returns: The parsed document as XMLContent object.
:rtype: :py:class:`weblyzard_api.model.xml_content.XMLContent`
'''
cls._check_document_format(api_dict, strict=True)
# This basically creates an empty XMLContent object
xml_content = XMLContent(xml_content=None, remove_duplicates=True)
# add all items in api_dict unless they need special handling
xml_content.update_attributes({
key: value
for key, value in api_dict.items()
if key not in ('sentences', 'annotations', 'language_id',
'features', 'relations', 'content')
})
# parse sentences
sentences = [
JSON10ParserSentence.from_api_dict(sentence_dict)
for sentence_dict in api_dict.get('sentences', [])
]
xml_content.sentences = sentences
# parse annotations
annotations = [
JSON10ParserAnnotation.from_api_dict(annotation_dict)
for annotation_dict in api_dict.get('annotations', [])
]
xml_content.body_annotations = annotations
# add relations and features
xml_content.relations = api_dict.get('relations', {})
xml_content.features = api_dict.get('features', {})
# map the language_id to XMLContent.lang
if 'language_id' in api_dict:
xml_content.attributes['lang'] = api_dict['language_id']
# removed this: title is already set via attributes
if 'title' in api_dict:
for sentence in sentences:
if sentence.is_title and sentence.value != api_dict['title']:
raise MalformedJSONException(
'The sentence marked with "is_title": "True" must ' +
'match the "title" attribute.')
else:
for sentence in sentences:
if sentence.is_title:
api_dict['title'] = sentence.value
return xml_content
def from_api_dict(cls, api_dict):
'''
Parses a dict with a structure analoguous to the JSON format defined
in the API specification.
:param api_dict: The document to parse.
:type api_dict: dict
:returns: The parsed document as XMLContent object.
:rtype: :py:class:`weblyzard_api.model.xml_content.XMLContent`
'''
cls._check_document_format(api_dict, strict=True)
# This basically creates an empty XMLContent object
xml_content = XMLContent(xml_content=None, remove_duplicates=True)
# add all items in api_dict unless they need special handling
xml_content.update_attributes({key: value for key, value in api_dict.iteritems() if
key not in ('sentences', 'annotations',
'language_id', 'features',
'relations', 'content')})
# parse sentences
sentences = [JSON10ParserSentence.from_api_dict(sentence_dict) for
sentence_dict in api_dict.get('sentences', [])]
xml_content.sentences = sentences
# parse annotations
annotations = [JSON10ParserAnnotation.from_api_dict(annotation_dict) for
annotation_dict in api_dict.get('annotations', [])]
xml_content.body_annotations = annotations
# add relations and features
xml_content.relations = api_dict.get('relations', {})
xml_content.features = api_dict.get('features', {})
# map the language_id to XMLContent.lang
if 'language_id' in api_dict:
xml_content.attributes['lang'] = api_dict['language_id']
# removed this: title is already set via attributes
if 'title' in api_dict:
for sentence in sentences:
if sentence.is_title and sentence.value != api_dict['title']:
raise MalformedJSONException('The sentence marked with "is_title": "True" must ' +
'match the "title" attribute.')
else:
for sentence in sentences:
if sentence.is_title:
api_dict['title'] = sentence.value
return xml_content
