def test_blob_df():
name = gen_name()
# upload the blob
with deletable(dp.Blob.upload_df(df=df, name=name)) as b1:
# are fields added?
check_name(b1, name)
# was the import as expected?
assert b1.num_rows == 4
assert b1.num_columns == 2
assert b1.cells == 8
# download
df1 = b1.download_df()
# is the exported df as expected?
assert df1.size == b1.cells
assert df1.shape == (b1.num_rows, b1.num_columns)
check_df_equal(df1, df)
# test obj lookup using name
b2 = dp.Blob.get(b1.name)
assert b2.name == b1.name
# test obj lookup using id
b3 = dp.Blob.by_id(b1.id)
assert b3.name == b2.name
def test_blob_csv_export(tmp_path: Path):
# upload a df file
b1 = dp.Blob.upload_df(df, name=gen_name())
with deletable(b1):
# export back to a csv and compare
fn1 = tmp_path / "exported.csv"
b1.download_file(fn1)
df1 = pd.read_csv(fn1)
check_df_equal(df, df1)
# export back to excel and compare
fn2 = tmp_path / "exported.xlsx"
b1.download_file(fn2)
df2 = pd.read_excel(fn2, engine="openpyxl")
check_df_equal(df, df2)
def test_blob_csv(tmp_path: Path):
# upload a csv file
# TODO - use df_processor?
fn = tmp_path / "initial.csv"
df.to_csv(fn, index=False)
b1 = dp.Blob.upload_file(fn, name=gen_name())
with deletable(b1):
# export back to a csv and compare
fn1 = tmp_path / "exported.csv"
b1.download_file(fn1)
df1 = pd.read_csv(fn1)
check_df_equal(df, df1)
# export back to excel and compare
fn2 = tmp_path / "exported.xlsx"
b1.download_file(fn2)
df2 = pd.read_excel(fn2, engine="openpyxl")
check_df_equal(df, df2)