def documents(start_index, end_index):
for i in range(start_index, end_index):
doc = Document()
doc.text = 'this is text'
doc.tags['id'] = 'id in tags'
doc.tags['inner_dict'] = {'id': 'id in inner_dict'}
chunk = Document()
chunk.text = 'text in chunk'
chunk.tags['id'] = 'id in chunk tags'
doc.chunks.append(chunk)
yield doc
def random_docs(num_docs):
for j in range(1, num_docs + 1):
doc = Document()
doc.text = f'i\'m dummy doc {j}'
doc.offset = 1000
doc.tags['id'] = 1000 # this will be ignored
yield doc
def test_data_type_builder_doc(builder, input_data_type, output_data_type):
a = Document()
a.id = 'a236cbb0eda62d58'
a.text = 'text test'
d, t = _new_doc_from_data(builder(a), input_data_type)
if input_data_type != DataInputType.CONTENT:
assert d.id == a.id
assert d.text == a.text
assert t == output_data_type
def set_hello(d: Document):
d.text = 'hello'
return d
def bar(d: Document):
d.text = 'hello'
return d
