You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
210 lines
6.6 KiB
210 lines
6.6 KiB
from io import BytesIO
|
|
import os
|
|
import pathlib
|
|
import tarfile
|
|
import zipfile
|
|
|
|
import numpy as np
|
|
import pytest
|
|
|
|
from pandas import (
|
|
DataFrame,
|
|
date_range,
|
|
read_csv,
|
|
read_excel,
|
|
read_json,
|
|
read_parquet,
|
|
)
|
|
import pandas._testing as tm
|
|
from pandas.util import _test_decorators as td
|
|
|
|
|
|
@pytest.fixture
|
|
def gcs_buffer():
|
|
"""Emulate GCS using a binary buffer."""
|
|
pytest.importorskip("gcsfs")
|
|
fsspec = pytest.importorskip("fsspec")
|
|
|
|
gcs_buffer = BytesIO()
|
|
gcs_buffer.close = lambda: True
|
|
|
|
class MockGCSFileSystem(fsspec.AbstractFileSystem):
|
|
@staticmethod
|
|
def open(*args, **kwargs):
|
|
gcs_buffer.seek(0)
|
|
return gcs_buffer
|
|
|
|
def ls(self, path, **kwargs):
|
|
# needed for pyarrow
|
|
return [{"name": path, "type": "file"}]
|
|
|
|
# Overwrites the default implementation from gcsfs to our mock class
|
|
fsspec.register_implementation("gs", MockGCSFileSystem, clobber=True)
|
|
|
|
return gcs_buffer
|
|
|
|
|
|
# Patches pyarrow; other processes should not pick up change
|
|
@pytest.mark.single_cpu
|
|
@pytest.mark.parametrize("format", ["csv", "json", "parquet", "excel", "markdown"])
|
|
def test_to_read_gcs(gcs_buffer, format, monkeypatch, capsys):
|
|
"""
|
|
Test that many to/read functions support GCS.
|
|
|
|
GH 33987
|
|
"""
|
|
|
|
df1 = DataFrame(
|
|
{
|
|
"int": [1, 3],
|
|
"float": [2.0, np.nan],
|
|
"str": ["t", "s"],
|
|
"dt": date_range("2018-06-18", periods=2),
|
|
}
|
|
)
|
|
|
|
path = f"gs://test/test.{format}"
|
|
|
|
if format == "csv":
|
|
df1.to_csv(path, index=True)
|
|
df2 = read_csv(path, parse_dates=["dt"], index_col=0)
|
|
elif format == "excel":
|
|
path = "gs://test/test.xlsx"
|
|
df1.to_excel(path)
|
|
df2 = read_excel(path, parse_dates=["dt"], index_col=0)
|
|
elif format == "json":
|
|
df1.to_json(path)
|
|
df2 = read_json(path, convert_dates=["dt"])
|
|
elif format == "parquet":
|
|
pytest.importorskip("pyarrow")
|
|
pa_fs = pytest.importorskip("pyarrow.fs")
|
|
|
|
class MockFileSystem(pa_fs.FileSystem):
|
|
@staticmethod
|
|
def from_uri(path):
|
|
print("Using pyarrow filesystem")
|
|
to_local = pathlib.Path(path.replace("gs://", "")).absolute().as_uri()
|
|
return pa_fs.LocalFileSystem(to_local)
|
|
|
|
with monkeypatch.context() as m:
|
|
m.setattr(pa_fs, "FileSystem", MockFileSystem)
|
|
df1.to_parquet(path)
|
|
df2 = read_parquet(path)
|
|
captured = capsys.readouterr()
|
|
assert captured.out == "Using pyarrow filesystem\nUsing pyarrow filesystem\n"
|
|
elif format == "markdown":
|
|
pytest.importorskip("tabulate")
|
|
df1.to_markdown(path)
|
|
df2 = df1
|
|
|
|
tm.assert_frame_equal(df1, df2)
|
|
|
|
|
|
def assert_equal_zip_safe(result: bytes, expected: bytes, compression: str):
|
|
"""
|
|
For zip compression, only compare the CRC-32 checksum of the file contents
|
|
to avoid checking the time-dependent last-modified timestamp which
|
|
in some CI builds is off-by-one
|
|
|
|
See https://en.wikipedia.org/wiki/ZIP_(file_format)#File_headers
|
|
"""
|
|
if compression == "zip":
|
|
# Only compare the CRC checksum of the file contents
|
|
with zipfile.ZipFile(BytesIO(result)) as exp, zipfile.ZipFile(
|
|
BytesIO(expected)
|
|
) as res:
|
|
for res_info, exp_info in zip(res.infolist(), exp.infolist()):
|
|
assert res_info.CRC == exp_info.CRC
|
|
elif compression == "tar":
|
|
with tarfile.open(fileobj=BytesIO(result)) as tar_exp, tarfile.open(
|
|
fileobj=BytesIO(expected)
|
|
) as tar_res:
|
|
for tar_res_info, tar_exp_info in zip(
|
|
tar_res.getmembers(), tar_exp.getmembers()
|
|
):
|
|
actual_file = tar_res.extractfile(tar_res_info)
|
|
expected_file = tar_exp.extractfile(tar_exp_info)
|
|
assert (actual_file is None) == (expected_file is None)
|
|
if actual_file is not None and expected_file is not None:
|
|
assert actual_file.read() == expected_file.read()
|
|
else:
|
|
assert result == expected
|
|
|
|
|
|
@pytest.mark.parametrize("encoding", ["utf-8", "cp1251"])
|
|
def test_to_csv_compression_encoding_gcs(
|
|
gcs_buffer, compression_only, encoding, compression_to_extension
|
|
):
|
|
"""
|
|
Compression and encoding should with GCS.
|
|
|
|
GH 35677 (to_csv, compression), GH 26124 (to_csv, encoding), and
|
|
GH 32392 (read_csv, encoding)
|
|
"""
|
|
df = tm.makeDataFrame()
|
|
|
|
# reference of compressed and encoded file
|
|
compression = {"method": compression_only}
|
|
if compression_only == "gzip":
|
|
compression["mtime"] = 1 # be reproducible
|
|
buffer = BytesIO()
|
|
df.to_csv(buffer, compression=compression, encoding=encoding, mode="wb")
|
|
|
|
# write compressed file with explicit compression
|
|
path_gcs = "gs://test/test.csv"
|
|
df.to_csv(path_gcs, compression=compression, encoding=encoding)
|
|
res = gcs_buffer.getvalue()
|
|
expected = buffer.getvalue()
|
|
assert_equal_zip_safe(res, expected, compression_only)
|
|
|
|
read_df = read_csv(
|
|
path_gcs, index_col=0, compression=compression_only, encoding=encoding
|
|
)
|
|
tm.assert_frame_equal(df, read_df)
|
|
|
|
# write compressed file with implicit compression
|
|
file_ext = compression_to_extension[compression_only]
|
|
compression["method"] = "infer"
|
|
path_gcs += f".{file_ext}"
|
|
df.to_csv(path_gcs, compression=compression, encoding=encoding)
|
|
|
|
res = gcs_buffer.getvalue()
|
|
expected = buffer.getvalue()
|
|
assert_equal_zip_safe(res, expected, compression_only)
|
|
|
|
read_df = read_csv(path_gcs, index_col=0, compression="infer", encoding=encoding)
|
|
tm.assert_frame_equal(df, read_df)
|
|
|
|
|
|
def test_to_parquet_gcs_new_file(monkeypatch, tmpdir):
|
|
"""Regression test for writing to a not-yet-existent GCS Parquet file."""
|
|
pytest.importorskip("fastparquet")
|
|
pytest.importorskip("gcsfs")
|
|
|
|
from fsspec import AbstractFileSystem
|
|
|
|
df1 = DataFrame(
|
|
{
|
|
"int": [1, 3],
|
|
"float": [2.0, np.nan],
|
|
"str": ["t", "s"],
|
|
"dt": date_range("2018-06-18", periods=2),
|
|
}
|
|
)
|
|
|
|
class MockGCSFileSystem(AbstractFileSystem):
|
|
def open(self, path, mode="r", *args):
|
|
if "w" not in mode:
|
|
raise FileNotFoundError
|
|
return open(os.path.join(tmpdir, "test.parquet"), mode, encoding="utf-8")
|
|
|
|
monkeypatch.setattr("gcsfs.GCSFileSystem", MockGCSFileSystem)
|
|
df1.to_parquet(
|
|
"gs://test/test.csv", index=True, engine="fastparquet", compression=None
|
|
)
|
|
|
|
|
|
@td.skip_if_installed("gcsfs")
|
|
def test_gcs_not_present_exception():
|
|
with tm.external_error_raised(ImportError):
|
|
read_csv("gs://test/test.csv")
|
|
|