def test_delete(conn, tmp_path):
a = tmp_path / 'a'
b = a / 'b'
c = a / 'c'
d = a / 'd'
for p in [a, b, c, d]:
p.mkdir()
scan(tmp_path, conn)
q = sa.select([sa.func.count()]).select_from(model.paths)
r = conn.execute(q).scalar()
assert r == count_files(tmp_path)
# Remove a file and scan it
b.rmdir()
scan(tmp_path, conn)
q = sa.select([sa.func.count()]).select_from(model.paths)
r = conn.execute(q).scalar()
assert r == count_files(tmp_path)
# Remove a file and don't scan it - doesn't update DB
d.rmdir()
scan(c, conn)
q = sa.select([sa.func.count()]).select_from(model.paths)
r = conn.execute(q).scalar()
assert r != count_files(tmp_path)
def test_exclude(conn, sample_data, sample_db):
# Find all files except those under 'a/c'
q = find(conn, path=sample_data, exclude='c')
results = [r.path for r in conn.execute(q)]
assert 'a/c' not in results
assert 'a/c/d/e' not in results
assert len(results) == count_files(sample_data) - \
count_files(sample_data / 'a' / 'c')
def test_scan_twice(conn, sample_db, sample_data):
# Scanning the same directory twice should not change data
q = sa.select([model.paths.c.id])
r = conn.execute(q)
assert len(list(r)) == count_files(sample_data)
scan(sample_data, conn)
q = sa.select([model.paths.c.id])
r = conn.execute(q)
assert len(list(r)) == count_files(sample_data)
def test_paths_count(conn, sample_db, sample_data):
# There should be a paths and paths_fullpath row for each file
q = sa.select([sa.func.count()]).select_from(model.paths)
r = conn.execute(q).scalar()
assert r == count_files(sample_data)
q = sa.select([sa.func.count()]).select_from(model.paths_fullpath)
r = conn.execute(q).scalar()
assert r == count_files(sample_data)
def test_find_subtree(conn, sample_data, sample_db):
# Find all files under 'a/c'
q = find(path=sample_data / 'a' / 'c', connection=conn)
results = [r.path for r in conn.execute(q)]
assert str(sample_data / 'a/c/d/e') in results
assert str(sample_data / 'a/b/f') not in results
assert len(results) == count_files(sample_data / 'a' / 'c')
def test_find_all(conn, sample_db, sample_data):
# Find all files in the DB
q = find(path=None, connection=conn)
results = conn.execute(q)
paths = [r.path for r in results]
assert len(paths) == count_files(sample_data)
def test_non_utf(conn, tmp_path):
a = str(tmp_path / '\ud83d').encode('utf-8', 'surrogatepass')
os.mkdir(a)
scan(tmp_path, conn)
q = sa.select([sa.func.count()]).select_from(model.paths)
r = conn.execute(q).scalar()
assert r == count_files(tmp_path)
def test_scan_sample(conn, sample_db, sample_data):
# Check the scanned sample data
q = sa.select([model.paths.c.inode])
r = conn.execute(q)
# Note the root path is not included in the scan
assert len(list(r)) == count_files(sample_data)
q = sa.select([model.paths.c.last_seen])
for r in conn.execute(q):
assert r.last_seen is not None