def test__same_text_correlation(self):
""" Test that for same text correlation is 1"""
_log.info('-'*80)
# arrange
text1 = "love is rain as long story short"
text2 = text1
dump_file = getInputFile("swiki_knowledge_output.xml")
parsed_file = getOutputFile("swiki_knowledge_output.parsed.xml")
#wdb_file = getOutputFile("swiki_knowledge_output.wdb")
articles = ['Rain', 'Love', 'Tree']
# act
wn.make_dump(dump_file, articles, compress=False)
wn.parse_dump(dump_file, parsed_file)
db_wrapper = wn.build_database_wrapper(parsed_file, StopWordsStemmer([]))
#self.addCleanup(os.remove, self.tmp_dump_file)
comparer = SemanticComparer(db_wrapper)
correlation = comparer.compare(text1, text2)
_log.info(test_utils.get_texts_correlation_message(text1, text2, correlation))
self.assertAlmostEqual(correlation, 1.0, msg="for same text correlation should be 1")
def test__same_text_correlation(self):
""" Test that for same text correlation is 1"""
_log.info('-' * 80)
# arrange
text1 = "love is rain as long story short"
text2 = text1
dump_file = getInputFile("swiki_knowledge_output.xml")
parsed_file = getOutputFile("swiki_knowledge_output.parsed.xml")
#wdb_file = getOutputFile("swiki_knowledge_output.wdb")
articles = ['Rain', 'Love', 'Tree']
# act
wn.make_dump(dump_file, articles, compress=False)
wn.parse_dump(dump_file, parsed_file)
db_wrapper = wn.build_database_wrapper(parsed_file,
StopWordsStemmer([]))
#self.addCleanup(os.remove, self.tmp_dump_file)
comparer = SemanticComparer(db_wrapper)
correlation = comparer.compare(text1, text2)
_log.info(
test_utils.get_texts_correlation_message(text1, text2,
correlation))
self.assertAlmostEqual(correlation,
1.0,
msg="for same text correlation should be 1")
def test__many_articles(self):
wiki_dump_path = getInputFile("many_articles_dump.xml")
parsed_xml_path = getOutputFile("many_articles_dump.parsed.xml")
wiki_knowledge.parse_dump(wiki_dump_path, parsed_xml_path)
db_wrapper = wiki_knowledge.build_database_wrapper(parsed_xml_path, PorterStemmer())
c = db_wrapper.get_readable_centroid(ibm_licence_text)
print c
def test__many_articles(self):
wdb_path = getInputFile("many_articles_dump.wdb")
db_wrapper = wiki_knowledge.load_db_wrapper_from_wdb(wdb_path)
d = db_wrapper.get_readable_centroid(ibm_licence_text_full)
s = sorted(d.items(), key=lambda x: x[1], reverse=True)[:5]
print s
d = db_wrapper.get_readable_centroid(ibm_licence_text)
s = sorted(d.items(), key=lambda x: x[1], reverse=True)[:5]
print s
def test__many_articles(self):
wiki_dump_path = getInputFile("many_articles_dump.xml")
parsed_xml_path = getOutputFile("many_articles_dump.parsed.xml")
wiki_knowledge.parse_dump(wiki_dump_path, parsed_xml_path)
db_wrapper = wiki_knowledge.build_database_wrapper(
parsed_xml_path, PorterStemmer())
c = db_wrapper.get_readable_centroid(ibm_licence_text)
print c
def test__many_articles(self):
wdb_path = getInputFile("many_articles_dump.wdb")
db_wrapper = wiki_knowledge.load_db_wrapper_from_wdb(wdb_path)
d = db_wrapper.get_readable_centroid(ibm_licence_text_full)
s = sorted(d.items(), key=lambda x: x[1], reverse=True)[:5]
print s
d = db_wrapper.get_readable_centroid(ibm_licence_text)
s = sorted(d.items(), key=lambda x: x[1], reverse=True)[:5]
print s
def test__many_articles(self):
wdb_path = getInputFile("many_articles_dump.wdb")
db_wrapper = wiki_knowledge.load_db_wrapper_from_wdb(wdb_path)
d = db_wrapper.get_readable_centroid(ibm_licence_text_full)
top = get_top(d, 5)
self.assertIn("Computer", dict(top))
d = db_wrapper.get_readable_centroid(ibm_licence_text)
top = get_top(d, 5)
self.assertIn("Computer", dict(top))
def test__many_articles(self):
wdb_path = getInputFile("many_articles_dump.wdb")
db_wrapper = wiki_knowledge.load_db_wrapper_from_wdb(wdb_path)
d = db_wrapper.get_readable_centroid(ibm_licence_text_full)
top = get_top(d,5)
self.assertIn("Computer", dict(top))
d = db_wrapper.get_readable_centroid(ibm_licence_text)
top = get_top(d,5)
self.assertIn("Computer", dict(top))
def test_extract_pages(self):
'''regression check that extract_pages works well'''
# template.format('id', ' title', 'length'); template = "{:<12}{:<30}{:>12}"
expected =[
(243478, 'Ross Ice Shelf', 13734),
(18798090, 'Southern Cross Expedition', 39110),
(343246, 'Ice shelf', 8262)
]
test__parse_tools_xml = getInputFile(FilesList.test__parse_tools)
actual = [(wdoc.id, wdoc.title, len(wdoc.wiki_text))
for wdoc
in pt.iterate_wiki_pages(test__parse_tools_xml)] # extract_pages(test__parse_tools_xml)]
self.assertSequenceEqual(actual, expected, "Assertion failure: \nActual={}\nExpected={}".format(actual, expected))
def test_extract_pages(self):
'''regression check that extract_pages works well'''
# template.format('id', ' title', 'length'); template = "{:<12}{:<30}{:>12}"
expected = [(243478, 'Ross Ice Shelf', 13734),
(18798090, 'Southern Cross Expedition', 39110),
(343246, 'Ice shelf', 8262)]
test__parse_tools_xml = getInputFile(FilesList.test__parse_tools)
actual = [(wdoc.id, wdoc.title, len(wdoc.wiki_text))
for wdoc in pt.iterate_wiki_pages(test__parse_tools_xml)
] # extract_pages(test__parse_tools_xml)]
self.assertSequenceEqual(
actual, expected,
"Assertion failure: \nActual={}\nExpected={}".format(
actual, expected))
def test_number_of_concepts(self):
""" db builder reads parsed xml properly"""
_log.info('-'*80)
# arrange
dump_file = getInputFile("wikidump_Knowledge_Love_War.xml")
parsed_file = getOutputFile("wikidump_Knowledge_Love_War.parsed.xml")
# act
wn.parse_dump(dump_file, parsed_file)
db_wrapper = wn.build_database_wrapper(parsed_file, StopWordsStemmer([]))
titles_count =len(db_wrapper.title_index)
concepts_count =len(db_wrapper.concepts_index)
# assert
self.assertEqual(titles_count, 3, "number of tiltes should be 3, got {0}".format(titles_count))
self.assertEqual(concepts_count, 3, "number of tiltes should be 3, got {0}".format(concepts_count))
def test_number_of_concepts(self):
""" db builder reads parsed xml properly"""
_log.info('-' * 80)
# arrange
dump_file = getInputFile("wikidump_Knowledge_Love_War.xml")
parsed_file = getOutputFile("wikidump_Knowledge_Love_War.parsed.xml")
# act
wn.parse_dump(dump_file, parsed_file)
db_wrapper = wn.build_database_wrapper(parsed_file,
StopWordsStemmer([]))
titles_count = len(db_wrapper.title_index)
concepts_count = len(db_wrapper.concepts_index)
# assert
self.assertEqual(
titles_count, 3,
"number of tiltes should be 3, got {0}".format(titles_count))
self.assertEqual(
concepts_count, 3,
"number of tiltes should be 3, got {0}".format(concepts_count))
def test__parse_dump(self):
wiki_dump_path = io_tu.getInputFile(io_tu.FilesList.test__parse_tools)
wiki_parsed_dump_path = io_tu.getOutputFile(
io_tu.FilesList.test__parse_tools)
wn.parse_dump(wiki_dump_path, wiki_parsed_dump_path)
def test__many_articles_files(self):
wdb_path = getInputFile("many_articles_dump.wdb")
text_path = getInputFile("ibm_licence.txt")
d = wiki_knowledge.get_value_from_file(wdb_path, text_path)
top = get_top(d, 5)
self.assertIn("Computer", dict(top))
def test__parse_dump(self):
wiki_dump_path = io_tu.getInputFile(io_tu.FilesList.test__parse_tools)
wiki_parsed_dump_path = io_tu.getOutputFile(io_tu.FilesList.test__parse_tools)
wn.parse_dump(wiki_dump_path, wiki_parsed_dump_path)
def test__many_articles_files(self):
wdb_path = getInputFile("many_articles_dump.wdb")
text_path = getInputFile("ibm_licence.txt")
d = wiki_knowledge.get_value_from_file(wdb_path, text_path)
top = get_top(d,5)
self.assertIn("Computer", dict(top))
