# Load document
context = etree.iterparse(file_name,
events=('end',),
remove_blank_text=True)
annotations_tag = '{http://ilk.uvt.nl/folia}annotations'
sentence_tag = '{http://ilk.uvt.nl/folia}s'
word_tag = '{http://ilk.uvt.nl/folia}w'
text_content_tag = '{http://ilk.uvt.nl/folia}t'
for event, elem in context:
if elem.tag == annotations_tag:
# add entity-annotation for liwc
annotation_attrs = {
'annotator': 'liwc',
'annotatortype': 'auto',
'datetime': datetime.now().isoformat(),
'set': 'liwc-set'
}
etree.SubElement(elem, 'entity-annotation', annotation_attrs)
if elem.tag == sentence_tag:
words = elem.findall(word_tag)
for word in words:
w = word.find(text_content_tag).text
if w in liwc_dict.keys():
for cat in liwc_dict[w]:
cat_label = 'liwc-{}'.format(liwc_categories[cat])
add_entity(elem, cat_label, [word], text_content_tag)
write_folia_file(context, file_name, dir_out, 'liwc')
entity_annotations = elem.findall(entity_annotations_tag)
for element in entity_annotations:
if element.attrib.get("annotator") == annotator_name:
add_annotation_tag = False
# add entity-annotation for embodied emotions
if add_annotation_tag:
annotation_attrs = {
"annotator": annotator_name,
"annotatortype": "manual",
"datetime": datetime.now().isoformat(),
"set": annotator_set,
}
etree.SubElement(elem, "entity-annotation", annotation_attrs)
if elem.tag == sentence_tag:
words = elem.findall(word_tag)
for word in words:
w_id = word.attrib.get(id_tag)
if w_id in word_id2annotations.keys():
# print w_id
for annotation in word_id2annotations[w_id]:
cat_label = "EmbodiedEmotions-{}".format(annotation.folia_entity_class())
add_entity(elem, cat_label, None, None, annotation)
num_annotations_added += 1
print " Added {} entities".format(num_annotations_added)
with open(folia_file, "w") as f:
f.write(etree.tostring(context.root, encoding="utf8", xml_declaration=True, pretty_print=True))
