def test_correct_amount_of_events_should_be_extracted():
"""
Test, if all events of the test-file marked_sentences.txt are extracted
"""
e = Event()
e.read_file("test/test_articles/marked_sentences.txt")
assert_equal(len(e.start_extraction()), 19)
def test_read_from_file_should_deliver_text():
"""
Test, if reading a file returns a string that is not empty.
"""
e = Event()
e.read_file("test/test_articles/01_wikipedia_syrian_civil_war.txt")
assert_not_equal(e.text, "")
def test_remove_references():
e = Event()
e.read_file("test/test_articles/01_wikipedia_syrian_civil_war.txt")
text = remove_references(e.text)
if re.search(str('\[\d{1,3}]'), text):
print("CHECK")
else:
print("NO")
def test_remove_references():
e = Event()
e.read_file("test/test_articles/01_wikipedia_syrian_civil_war.txt")
text = remove_references(e.text)
if re.search(str('\[\d{1,3}]'), text):
print("CHECK")
else:
print("NO")
def test_references_should_be_removed():
"""
Check, if references are removed from text.
Maybe to be refined later on, because, it is the same regex command that removes the references from text -
therefore it is most likely that this test is fine. hmm...
"""
e = Event()
e.read_file("test/test_articles/01_wikipedia_syrian_civil_war.txt")
# re searches for references in bracktes, like "[12]" and should not find any.
assert_false(re.search(str('\[\d{1,3}]'), remove_references(e.text)))
def extract_events(self):
"""
Each event in the goldmaster-file is analyzed and extracted by eventminer.
The results (extracted event, dates, etc.) will be compared with the initial data from the goldmaster-file.
"""
for event in self.event_text_raw:
e = Event()
e.read_text(event)
result_set = e.start_accuracy_extraction()
self.event_text_eventminer.append(result_set["event"].encode("utf-8"))
self.rule_nr_eventminer.append(result_set["rule_nr"])
self.location_eventminer.append(result_set["location"])
self.start_day_eventminer.append(result_set["start_day"])
self.start_month_eventminer.append(str(result_set["start_month"]))
self.start_year_eventminer.append(result_set["start_year"])
self.end_day_eventminer.append(str(result_set["end_day"]))
self.end_month_eventminer.append(result_set["end_month"])
self.end_year_eventminer.append(result_set["end_year"])
def extract_events(self):
"""
Each event in the goldmaster-file is analyzed and extracted by eventminer.
The results (extracted event, dates, etc.) will be compared with the initial data from the goldmaster-file.
"""
for event in self.event_text_raw:
e = Event()
e.read_text(event)
result_set = e.start_accuracy_extraction()
self.event_text_eventminer.append(
result_set["event"].encode("utf-8"))
self.rule_nr_eventminer.append(result_set["rule_nr"])
self.location_eventminer.append(result_set["location"])
self.start_day_eventminer.append(result_set["start_day"])
self.start_month_eventminer.append(str(result_set["start_month"]))
self.start_year_eventminer.append(result_set["start_year"])
self.end_day_eventminer.append(str(result_set["end_day"]))
self.end_month_eventminer.append(result_set["end_month"])
self.end_year_eventminer.append(result_set["end_year"])
def test_read_from_file():
e = Event()
e.read_file("test/test_articles/random_test.txt")
print(e.text)
def flask_start_extraction(wiki_text):
e = Event()
e.text = wiki_text
e.text = remove_references(e.text)
return e.start_extraction()
def test_start_extraction():
e = Event()
e.read_file("test/test_articles/random_test.txt")
e.text = remove_references(e.text)
e.start_extraction()
def test_start_extraction():
e = Event()
e.read_file("test/test_articles/random_test.txt")
e.text = remove_references(e.text)
e.start_extraction()
