def test_file_exists(self):
"""Ensure an error is raised when a file exists"""
with TemporaryDirectory() as tmp:
fp = os.path.join(tmp, "asdf.txt")
f2 = open(fp,'w+')
with self.assertRaises(Exception) as context:
with atomic_write(fp) as f:
f.write("exist")
def test_atomic_failure(self):
"""Ensure that file does not exist after failure during write"""
with TemporaryDirectory() as tmp:
fp = os.path.join(tmp, "asdf.txt")
with self.assertRaises(FakeFileFailure):
with atomic_write(fp, "w") as f:
tmpfile = f.name
assert os.path.exists(tmpfile)
raise FakeFileFailure()
assert not os.path.exists(tmpfile)
assert not os.path.exists(fp)
def test_atomic_write(self):
"""Ensure file exists after being written successfully"""
with TemporaryDirectory() as tmp:
fp = os.path.join(tmp, "asdf.txt")
with atomic_write(fp, "w") as f:
assert not os.path.exists(fp)
tmpfile = f.name
f.write("asdf")
assert not os.path.exists(tmpfile)
assert os.path.exists(fp)
with open(fp) as f:
self.assertEqual(f.read(), "asdf")
for user in ["gorlins", "frankiekienlam"]:
print("Id for {}: {}".format(user, get_user_id(user)))
data_source = "data/hashed.xlsx"
# TODO: read in, save as new parquet file, read back just id column, print
# Parquet Question
from pset_1.io import atomic_write
import pandas as pd
import pyarrow as pr
# transform excel file into .parquet file with the same content
df = pd.read_excel(data_source, index_col=0)
df.to_parquet('hashed.parquet', 'pyarrow')
# Print out the list of hashed_id using atomic_write
hash_id_list = list(
pr.parquet.read_table('hashed.parquet', columns=['hashed_id']))
hash_id_list_str = str(hash_id_list)
with atomic_write("hashed_id_list.txt") as f:
f.write(hash_id_list_str)
with open("hashed_id_list.txt") as f:
lines = f.read()
print(lines)