def test_docs_serialization_format(self):
import json
from eWRT.util.module_path import get_resource
DOCS = [{'id': 7,
'body': 'Ehre sei Gott.',
'title': '',
'format': 'text/html',
'header': {'test': 'testvalue'}},
{'id': 8,
'body': '',
'title': 'Guten Tag!',
'format': 'text/html',
'header': {}}]
REFERENCE_MULTI = json.load(
open(get_resource(__file__, 'data/jeremia_reference_output_documents.json')))
REFERENCE_SINGLE = json.load(open(get_resource(
__file__, 'data/jeremia_reference_output_single_document.json')))
# document list
j = Jeremia()
result = j.submit_documents(DOCS)
result.sort()
REFERENCE_MULTI.sort()
assert REFERENCE_MULTI == result
# single document
result = j.submit_document(DOCS[0])
assert REFERENCE_SINGLE == result
def test_single_document_with_annotations(self):
'''
Tests the handling of single document annotations.
'''
DOC = {'id': 12,
'body': 'UBS has finally succeeded. They obtained a 10% share of CS.',
'title': 'UBS versus Credit Suisse.',
'format': 'text/html',
'title_annotation': [{'start': 0, 'end': 3, 'surfaceForm': 'UBS', 'key': 'http://dbpedia.org/UBS'},
{'start': 11, 'end': 24, 'surfaceForm': 'Credit Suisse', 'key': 'http://dbpedia.org/Credit Suisse'}],
'body_annotation': [{'start': 0, 'end': 3, 'surfaceForm': 'UBS', 'key': 'http://dbpedia.org/UBS'},
{'start': 56, 'end': 58, 'surfaceForm': 'CS', 'key': 'http://dbpedia.org/Credit Suisse'}],
'header': {},
}
j = Jeremia()
# this test requires Jeremia version 0.0.4+
if j.version() < "0.0.4":
return
print('submitting document with annotations...')
result = j.submit_document(DOC)
# check: all annotations have been preserved
print(result)
assert len(result['annotation']) == 4
# check: annotations
for annotation in result['annotation']:
# title
if annotation['md5sum'] == '8e3f3deac5e6c01dab521c07e3a60d7b':
assert annotation['start'] == 0 or annotation['start'] == 11
assert annotation['end'] == 3 or annotation['end'] == 24
# first body sentence
elif annotation['md5sum'] == 'ffafdc744dcda3d58ab6eafc86ad99b1':
assert annotation['start'] == 0
assert annotation['end'] == 3
# second body sentence with adjusted indices
elif annotation['md5sum'] == '25faaf0960a68ae741125ca436b330ee':
assert annotation['start'] == 29
assert annotation['end'] == 31
def test_docs_serialization_format(self):
import json
from eWRT.util.module_path import get_resource
DOCS = [{
'id': 7,
'body': 'Ehre sei Gott.',
'title': '',
'format': 'text/html',
'header': {
'test': 'testvalue'
}
}, {
'id': 8,
'body': '',
'title': 'Guten Tag!',
'format': 'text/html',
'header': {}
}]
REFERENCE_MULTI = json.load(
open(
get_resource(__file__,
'data/jeremia_reference_output_documents.json')))
REFERENCE_SINGLE = json.load(
open(
get_resource(
__file__,
'data/jeremia_reference_output_single_document.json')))
# document list
j = Jeremia()
result = j.submit_documents(DOCS)
result.sort()
REFERENCE_MULTI.sort()
assert REFERENCE_MULTI == result
# single document
result = j.submit_document(DOCS[0])
assert REFERENCE_SINGLE == result
def test_single_document_with_annotations(self):
'''
Tests the handling of single document annotations.
'''
DOC = {
'id':
12,
'body':
'UBS has finally succeeded. They obtained a 10% share of CS.',
'title':
'UBS versus Credit Suisse.',
'format':
'text/html',
'title_annotation': [{
'start': 0,
'end': 3,
'surfaceForm': 'UBS',
'key': 'http://dbpedia.org/UBS'
}, {
'start': 11,
'end': 24,
'surfaceForm': 'Credit Suisse',
'key': 'http://dbpedia.org/Credit Suisse'
}],
'body_annotation': [{
'start': 0,
'end': 3,
'surfaceForm': 'UBS',
'key': 'http://dbpedia.org/UBS'
}, {
'start': 56,
'end': 58,
'surfaceForm': 'CS',
'key': 'http://dbpedia.org/Credit Suisse'
}],
'header': {},
}
j = Jeremia()
# this test requires Jeremia version 0.0.4+
if j.version() < b"0.0.4":
return
print('submitting document with annotations...')
result = j.submit_document(DOC)
# check: all annotations have been preserved
print(result)
assert len(result['annotation']) == 4
# check: annotations
for annotation in result['annotation']:
# title
if annotation['md5sum'] == '8e3f3deac5e6c01dab521c07e3a60d7b':
assert annotation['start'] == 0 or annotation['start'] == 11
assert annotation['end'] == 3 or annotation['end'] == 24
# first body sentence
elif annotation['md5sum'] == 'ffafdc744dcda3d58ab6eafc86ad99b1':
assert annotation['start'] == 0
assert annotation['end'] == 3
# second body sentence with adjusted indices
elif annotation['md5sum'] == '25faaf0960a68ae741125ca436b330ee':
assert annotation['start'] == 29
assert annotation['end'] == 31
def test_single_document_processing(self):
j = Jeremia()
print('submitting document...')
document_annotated = j.submit_document(self.DOCS[1])
self.assertTrue(document_annotated != "")
'''
add support for calling Jeremia tests as part of a test suite
'''
suite = unittest.TestSuite()
suite.addTest(unittest.makeSuite(JeremiaTest, 'test'))
return suite
if __name__ == '__main__':
if len(argv) > 1:
txt = argv[1]
docs = {
'id': '192292',
'body': txt,
'title': '',
'format': 'text/html',
'header': {
'test': 'testvalue'
}
}
j = Jeremia()
docs['body_annotation'] = [{
'start': 0,
'end': 3,
'key': 'test annotation'
}]
l = j.submit_document(docs)
print(l)
else:
unittest.main()
def test_single_document_processing(self):
j = Jeremia()
print('submitting document...')
document_annotated = j.submit_document(self.DOCS[1])
self.assertTrue(document_annotated != "")
# single document
result = j.submit_document(DOCS[0])
assert REFERENCE_SINGLE == result
def test_suite():
'''
add support for calling Jeremia tests as part of a test suite
'''
suite = unittest.TestSuite()
suite.addTest(unittest.makeSuite(JeremiaTest, 'test'))
return suite
if __name__ == '__main__':
if len(argv) > 1:
txt = argv[1]
docs = {'id': '192292',
'body': txt,
'title': '',
'format': 'text/html',
'header': {'test': 'testvalue'}}
j = Jeremia()
docs['body_annotation'] = [
{'start': 0, 'end': 3, 'key': 'test annotation'}]
l = j.submit_document(docs)
print(l)
else:
unittest.main()