def test_can_run_load_documents(tmpdir, dataset, length):
datapath = tmpdir.join("dataset.txt")
datapath.write(dataset)
docs = load_documents(datapath)
assert length == len(
docs
), f"Wrong loaded length with file {datapath}, expected {length}, got {len(docs)}"
def test_inverted_index_can_dump_load_index(tmpdir):
datapath = tmpdir.join("dataset.txt")
datapath.write(DATASET_SMALL_STR)
documents = load_documents(datapath)
inverted_index = build_inverted_index(documents)
tmp_fio = tmpdir.join('test.dump')
inverted_index.dump(StructStoragePolicy(), tmp_fio)
loaded_index = inverted_index.load(StructStoragePolicy(), tmp_fio)
assert inverted_index._index == loaded_index._index, "File was not created"
def test_load_documents_do_correct(tiny_dataset_fio):
documents = load_documents(tiny_dataset_fio)
etalon_document = {
123: "Show must go on!",
321: "Still loving you",
547: "The Adventures of Rain Dance Maggie",
645: "The House of Rising Son",
789: "A A B"
}
assert documents == etalon_document, "Dataset loaded incorrectly"
def test_callback_query_list(tiny_dataset_fio, tmpdir, capsys, query, answer):
documents = load_documents(tiny_dataset_fio)
inverted_index = build_inverted_index(documents)
tmp_fio = tmpdir.join('test.dump')
inverted_index.dump(StructStoragePolicy(), tmp_fio)
arguments = Namespace(query=query, index_path=tmp_fio)
callback_query(arguments)
captured = capsys.readouterr()
if captured.out == "\n":
query_ans = set()
else:
query_ans = set(int(var) for var in captured.out.rstrip().split(","))
assert query_ans == answer, "Wrong answer"
def test_callback_query_utf8(tiny_dataset_fio, tmpdir, capsys, query, answer):
documents = load_documents(tiny_dataset_fio)
inverted_index = build_inverted_index(documents)
tmp_fio = tmpdir.join('test.dump')
inverted_index.dump(StructStoragePolicy(), tmp_fio)
tmp_file = tmpdir.join('test.utf8')
with open(tmp_file, "w", encoding="utf8") as file:
file.write(" ".join(query) + "\n")
arguments = Namespace(query=[],
query_file=open(tmp_file, "r", encoding="utf8"),
index_path=tmp_fio)
callback_query(arguments)
captured = capsys.readouterr()
if captured.out == "\n":
query_ans = set()
else:
query_ans = set(int(var) for var in captured.out.rstrip().split(","))
assert query_ans == answer, "Wrong answer"
def test_can_run_build_index(tmpdir, dataset):
datapath = tmpdir.join("dataset.txt")
datapath.write(dataset)
documents = load_documents(datapath)
build_inverted_index(documents)
def test_error_load_documents_wrong_filepath():
with pytest.raises(FileNotFoundError):
load_documents("impresed")
def test_inverted_index_can_dump_index(tiny_dataset_fio, tmpdir):
documents = load_documents(tiny_dataset_fio)
inverted_index = build_inverted_index(documents)
tmp_fio = tmpdir.join('test.dump')
inverted_index.dump(StructStoragePolicy(), tmp_fio)
assert os.path.isfile(tmp_fio), "File was not created"
def test_query_return_right_answer(tiny_dataset_fio, query, answer):
documents = load_documents(tiny_dataset_fio)
inverted_index = build_inverted_index(documents)
assert sorted(answer) == sorted(inverted_index.query(query))
def test_can_build_index_do_correct(tiny_dataset_fio):
documents = load_documents(tiny_dataset_fio)
inverted_index = build_inverted_index(documents)
assert inverted_index._index == ETALON_TINY_INDEX, "Wrong build for inverted index"