def test_put_file(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
lfs = LocalFileSystem()
fs.mkdir("putdir")
# Check that put on an empty file works
with open("sample.txt", "wb") as f:
f.write(b"")
fs.put("sample.txt", "putdir/sample.txt")
fs.get("putdir/sample.txt", "sample2.txt")
with open("sample.txt", "rb") as f:
f1 = f.read()
with open("sample2.txt", "rb") as f:
f2 = f.read()
assert f1 == f2
lfs.rm("sample.txt")
lfs.rm("sample2.txt")
# Check that put on a file with data works
with open("sample3.txt", "wb") as f:
f.write(b"01234567890")
fs.put("sample3.txt", "putdir/sample3.txt")
fs.get("putdir/sample3.txt", "sample4.txt")
with open("sample3.txt", "rb") as f:
f3 = f.read()
with open("sample4.txt", "rb") as f:
f4 = f.read()
assert f3 == f4
fs.rm("putdir", recursive=True)
def test_mkdir_rm_recursive(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("test_mkdir_rm_recursive")
assert "test_mkdir_rm_recursive/" in fs.ls("")
with fs.open("test_mkdir_rm_recursive/file.txt", "wb") as f:
f.write(b"0123456789")
with fs.open("test_mkdir_rm_recursive/dir/file.txt", "wb") as f:
f.write(b"ABCD")
with fs.open("test_mkdir_rm_recursive/dir/file2.txt", "wb") as f:
f.write(b"abcdef")
assert fs.find("test_mkdir_rm_recursive") == [
"test_mkdir_rm_recursive/dir/file.txt",
"test_mkdir_rm_recursive/dir/file2.txt",
"test_mkdir_rm_recursive/file.txt",
]
fs.rm("test_mkdir_rm_recursive", recursive=True)
assert "test_mkdir_rm_recursive/" not in fs.ls("")
assert fs.find("test_mkdir_rm_recursive") == []
def test_mkdir(storage):
fs = AzureBlobFileSystem(
account_name=storage.account_name, connection_string=CONN_STR,
)
# Verify mkdir will create a new container when create_parents is True
fs.mkdir("new-container", create_parents=True)
assert "new-container" in fs.ls(".")
fs.rm("new-container")
# Verify a new container will not be created when create_parents
# is False
with pytest.raises(PermissionError):
fs.mkdir("new-container", create_parents=False)
# Test creating subdirectory when container does not exist
# Since mkdir is a no-op, if create_parents=True, it will create
# the top level container, but will NOT create nested directories
fs.mkdir("new-container/dir", create_parents=True)
assert "new-container/dir" not in fs.ls("new-container")
assert "new-container" in fs.ls(".")
fs.rm("new-container", recursive=True)
# Test that creating a directory when already exists passes
fs.mkdir("data")
assert "data" in fs.ls(".")
# Test raising error when container does not exist
with pytest.raises(PermissionError):
fs.mkdir("new-container/dir", create_parents=False)
def test_metadata_write(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("test_metadata_write")
data = b"0123456789"
metadata = {"meta": "data"}
# standard blob type
with fs.open("test_metadata_write/file.txt", "wb", metadata=metadata) as f:
f.write(data)
info = fs.info("test_metadata_write/file.txt")
assert info["metadata"] == metadata
metadata_changed_on_write = {"meta": "datum"}
with fs.open("test_metadata_write/file.txt",
"wb",
metadata=metadata_changed_on_write) as f:
f.write(data)
info = fs.info("test_metadata_write/file.txt")
assert info["metadata"] == metadata_changed_on_write
# append blob type
new_metadata = {"data": "meta"}
with fs.open("test_metadata_write/append-file.txt",
"ab",
metadata=metadata) as f:
f.write(data)
# try change metadata on block appending
with fs.open("test_metadata_write/append-file.txt",
"ab",
metadata=new_metadata) as f:
f.write(data)
info = fs.info("test_metadata_write/append-file.txt")
# azure blob client doesn't seem to support metadata mutation when appending blocks
# lets be sure this behavior doesn't change as this would imply
# a potential breaking change
assert info["metadata"] == metadata
# getxattr / setxattr
assert fs.getxattr("test_metadata_write/file.txt", "meta") == "datum"
fs.setxattrs("test_metadata_write/file.txt", metadata="data2")
assert fs.getxattr("test_metadata_write/file.txt", "metadata") == "data2"
assert fs.info("test_metadata_write/file.txt")["metadata"] == {
"metadata": "data2"
}
# empty file and nested directory
with fs.open("test_metadata_write/a/b/c/nested-file.txt",
"wb",
metadata=metadata) as f:
f.write(b"")
assert fs.getxattr("test_metadata_write/a/b/c/nested-file.txt",
"meta") == "data"
fs.setxattrs("test_metadata_write/a/b/c/nested-file.txt", metadata="data2")
assert fs.info(
"test_metadata_write/a/b/c/nested-file.txt")["metadata"] == {
"metadata": "data2"
}
fs.rmdir("test_metadata_write")
def test_cat(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("catdir")
data = b"0123456789"
with fs.open("catdir/catfile.txt", "wb") as f:
f.write(data)
assert fs.cat("catdir/catfile.txt") == data
fs.rm("catdir/catfile.txt")
def test_mkdir_rmdir(storage):
fs = AzureBlobFileSystem(
account_name=storage.account_name,
connection_string=CONN_STR,
)
fs.mkdir("new-container")
assert "new-container/" in fs.ls("")
assert fs.ls("new-container") == []
with fs.open(path="new-container/file.txt", mode="wb") as f:
f.write(b"0123456789")
with fs.open("new-container/dir/file.txt", "wb") as f:
f.write(b"0123456789")
with fs.open("new-container/dir/file2.txt", "wb") as f:
f.write(b"0123456789")
# Check to verify you can skip making a directory if the container
# already exists, but still create a file in that directory
fs.mkdir("new-container/dir/file.txt", exist_ok=False)
assert "new-container/" in fs.ls("")
fs.mkdir("new-container/file2.txt", exist_ok=False)
assert "new-container/file2.txt" in fs.ls("new-container")
# Test to verify that the file contains expected contents
with fs.open("new-container/file2.txt", "rb") as f:
outfile = f.read()
assert outfile == b""
# Check that trying to overwrite an existing nested file in append mode works as expected
fs.mkdir("new-container/dir/file2.txt", exist_ok=False)
assert "new-container/dir/file2.txt" in fs.ls("new-container/dir")
# Also verify you can make a nested directory structure
fs.mkdir("new-container/dir2/file.txt", exist_ok=False)
with fs.open("new-container/dir2/file.txt", "wb") as f:
f.write(b"0123456789")
assert "new-container/dir2/file.txt" in fs.ls("new-container/dir2")
fs.rm("new-container/dir2", recursive=True)
fs.rm("new-container/dir", recursive=True)
assert fs.ls("new-container") == [
"new-container/file.txt",
"new-container/file2.txt",
]
fs.rm("new-container/file.txt")
fs.rm("new-container/file2.txt")
fs.rmdir("new-container")
assert "new-container/" not in fs.ls("")
def test_cp_file(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("homedir")
fs.mkdir("homedir/enddir")
fs.touch("homedir/startdir/test_file.txt")
fs.cp_file("homedir/startdir/test_file.txt",
"homedir/enddir/test_file.txt")
files = fs.ls("homedir/enddir")
assert "homedir/enddir/test_file.txt" in files
fs.rm("homedir", recursive=True)
def test_mkdir_rmdir(storage):
fs = AzureBlobFileSystem(
account_name=storage.account_name,
connection_string=CONN_STR,
)
fs.mkdir("new-container")
assert "new-container/" in fs.ls("")
assert fs.ls("new-container") == []
with fs.open(path="new-container/file.txt", mode="wb") as f:
f.write(b"0123456789")
with fs.open("new-container/dir/file.txt", "wb") as f:
f.write(b"0123456789")
with fs.open("new-container/dir/file2.txt", "wb") as f:
f.write(b"0123456789")
# Check to verify you can skip making a directory if the container
# already exists, but still create a file in that directory
fs.mkdir("new-container/dir/file.txt", exists_ok=True)
assert "new-container/" in fs.ls("")
fs.mkdir("new-container/file2.txt", exists_ok=True)
with fs.open("new-container/file2.txt", "wb") as f:
f.write(b"0123456789")
assert "new-container/file2.txt" in fs.ls("new-container")
fs.mkdir("new-container/dir/file2.txt", exists_ok=True)
with fs.open("new-container/dir/file2.txt", "wb") as f:
f.write(b"0123456789")
assert "new-container/dir/file2.txt" in fs.ls("new-container/dir")
# Also verify you can make a nested directory structure
fs.mkdir("new-container/dir2/file.txt", exists_ok=True)
with fs.open("new-container/dir2/file.txt", "wb") as f:
f.write(b"0123456789")
assert "new-container/dir2/file.txt" in fs.ls("new-container/dir2")
fs.rm("new-container/dir2", recursive=True)
fs.rm("new-container/dir", recursive=True)
assert fs.ls("new-container") == [
"new-container/file.txt",
"new-container/file2.txt",
]
fs.rm("new-container/file.txt")
fs.rm("new-container/file2.txt")
fs.rmdir("new-container")
assert "new-container/" not in fs.ls("")
def test_append_operation(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("append-container")
# Check that appending to an existing file works as expected
with fs.open("append-container/append_file.txt", "ab") as f:
f.write(b"0123456789")
with fs.open("append-container/append_file.txt", "ab") as f:
f.write(b"0123456789")
with fs.open("new-container/dir/file2.txt", "rb") as f:
outfile = f.read()
assert outfile == b"01234567890123456789"
fs.rm("append-container", recursive=True)
def test_url(storage):
fs = AzureBlobFileSystem(
account_name=storage.account_name, connection_string=CONN_STR, account_key=KEY
)
fs.mkdir("catdir")
data = b"0123456789"
with fs.open("catdir/catfile.txt", "wb") as f:
f.write(data)
import requests
r = requests.get(fs.url("catdir/catfile.txt"))
assert r.status_code == 200
assert r.content == data
fs.rm("catdir/catfile.txt")
def fs(self):
from adlfs import AzureBlobFileSystem
from azure.core.exceptions import AzureError
try:
file_system = AzureBlobFileSystem(**self.login_info)
if self.bucket not in [
container.rstrip("/") for container in file_system.ls("/")
]:
file_system.mkdir(self.bucket)
except (ValueError, AzureError) as e:
raise AzureAuthError(
f"Authentication to Azure Blob Storage via {self.login_method}"
" failed.\nLearn more about configuration settings at"
f" {format_link('https://man.dvc.org/remote/modify')}") from e
return file_system
def test_dask_parquet(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("test")
STORAGE_OPTIONS = {
"account_name": "devstoreaccount1",
"connection_string": CONN_STR,
}
df = pd.DataFrame({
"col1": [1, 2, 3, 4],
"col2": [2, 4, 6, 8],
"index_key": [1, 1, 2, 2],
"partition_key": [1, 1, 2, 2],
})
dask_dataframe = dd.from_pandas(df, npartitions=1)
for protocol in ["abfs", "az"]:
dask_dataframe.to_parquet(
"{}://[email protected]/test_group.parquet".format(
protocol),
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
)
fs = AzureBlobFileSystem(**STORAGE_OPTIONS)
assert fs.ls("test/test_group.parquet") == [
"test/test_group.parquet/_common_metadata",
"test/test_group.parquet/_metadata",
"test/test_group.parquet/part.0.parquet",
]
fs.rm("test/test_group.parquet")
df_test = dd.read_parquet(
"abfs://test/test_group.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
assert_frame_equal(df, df_test)
def test_deep_paths(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("test_deep")
assert "test_deep/" in fs.ls("")
with fs.open("test_deep/a/b/c/file.txt", "wb") as f:
f.write(b"0123456789")
assert fs.ls("test_deep") == ["test_deep/a/"]
assert fs.ls("test_deep/") == ["test_deep/a/"]
assert fs.ls("test_deep/a") == ["test_deep/a/b/"]
assert fs.ls("test_deep/a/") == ["test_deep/a/b/"]
assert fs.find("test_deep") == ["test_deep/a/b/c/file.txt"]
assert fs.find("test_deep/") == ["test_deep/a/b/c/file.txt"]
assert fs.find("test_deep/a") == ["test_deep/a/b/c/file.txt"]
assert fs.find("test_deep/a/") == ["test_deep/a/b/c/file.txt"]
fs.rm("test_deep", recursive=True)
assert "test_deep/" not in fs.ls("")
assert fs.find("test_deep") == []
def test_mkdir(storage):
fs = AzureBlobFileSystem(
account_name=storage.account_name,
connection_string=CONN_STR,
)
# Verify mkdir will create a new container when create_parents is True
fs.mkdir("new-container", create_parents=True)
assert "new-container/" in fs.ls(".")
fs.rm("new-container")
# Verify a new container will not be created when create_parents
# is False
with pytest.raises(PermissionError):
fs.mkdir("new-container", create_parents=False)
# Test creating subdirectory when container does not exist
fs.mkdir("new-container/dir", create_parents=True)
assert "new-container/dir" in fs.ls("new-container")
fs.rm("new-container", recursive=True)
# Test raising error when container does not exist
with pytest.raises(PermissionError):
fs.mkdir("new-container/dir", create_parents=False)
def test_dask_parquet(storage):
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
fs.mkdir("test")
STORAGE_OPTIONS = {
"account_name": "devstoreaccount1",
"connection_string": CONN_STR,
}
df = pd.DataFrame({
"col1": [1, 2, 3, 4],
"col2": [2, 4, 6, 8],
"index_key": [1, 1, 2, 2],
"partition_key": [1, 1, 2, 2],
})
dask_dataframe = dd.from_pandas(df, npartitions=1)
for protocol in ["abfs", "az"]:
dask_dataframe.to_parquet(
"{}://test/test_group.parquet".format(protocol),
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
)
fs = AzureBlobFileSystem(**STORAGE_OPTIONS)
assert fs.ls("test/test_group.parquet") == [
"test/test_group.parquet/_common_metadata",
"test/test_group.parquet/_metadata",
"test/test_group.parquet/part.0.parquet",
]
fs.rm("test/test_group.parquet")
df_test = dd.read_parquet(
"abfs://test/test_group.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
assert_frame_equal(df, df_test)
A = np.random.randint(0, 100, size=(10000, 4))
df2 = pd.DataFrame(data=A, columns=list("ABCD"))
ddf2 = dd.from_pandas(df2, npartitions=4)
dd.to_parquet(
ddf2,
"abfs://test/test_group2.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
)
assert fs.ls("test/test_group2.parquet") == [
"test/test_group2.parquet/_common_metadata",
"test/test_group2.parquet/_metadata",
"test/test_group2.parquet/part.0.parquet",
"test/test_group2.parquet/part.1.parquet",
"test/test_group2.parquet/part.2.parquet",
"test/test_group2.parquet/part.3.parquet",
]
df2_test = dd.read_parquet(
"abfs://test/test_group2.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
assert_frame_equal(df2, df2_test)
a = np.full(shape=(10000, 1), fill_value=1)
b = np.full(shape=(10000, 1), fill_value=2)
c = np.full(shape=(10000, 1), fill_value=3)
d = np.full(shape=(10000, 1), fill_value=4)
B = np.concatenate((a, b, c, d), axis=1)
df3 = pd.DataFrame(data=B, columns=list("ABCD"))
ddf3 = dd.from_pandas(df3, npartitions=4)
dd.to_parquet(
ddf3,
"abfs://test/test_group3.parquet",
partition_on=["A", "B"],
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
)
assert fs.glob("test/test_group3.parquet/*") == [
"test/test_group3.parquet/A=1",
"test/test_group3.parquet/_common_metadata",
"test/test_group3.parquet/_metadata",
]
df3_test = dd.read_parquet(
"abfs://test/test_group3.parquet",
filters=[("A", "=", 1)],
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
df3_test = df3_test[["A", "B", "C", "D"]]
df3_test = df3_test[["A", "B", "C", "D"]].astype(int)
assert_frame_equal(df3, df3_test)
A = np.random.randint(0, 100, size=(10000, 4))
df4 = pd.DataFrame(data=A, columns=list("ABCD"))
ddf4 = dd.from_pandas(df4, npartitions=4)
dd.to_parquet(
ddf4,
"abfs://test/test_group4.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
flavor="spark",
write_statistics=False,
)
fs.rmdir("test/test_group4.parquet/_common_metadata", recursive=True)
fs.rmdir("test/test_group4.parquet/_metadata", recursive=True)
fs.rm("test/test_group4.parquet/_common_metadata")
fs.rm("test/test_group4.parquet/_metadata")
assert fs.ls("test/test_group4.parquet") == [
"test/test_group4.parquet/part.0.parquet",
"test/test_group4.parquet/part.1.parquet",
"test/test_group4.parquet/part.2.parquet",
"test/test_group4.parquet/part.3.parquet",
]
df4_test = dd.read_parquet(
"abfs://test/test_group4.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
assert_frame_equal(df4, df4_test)
A = np.random.randint(0, 100, size=(10000, 4))
df5 = pd.DataFrame(data=A, columns=list("ABCD"))
ddf5 = dd.from_pandas(df5, npartitions=4)
dd.to_parquet(
ddf5,
"abfs://test/test group5.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
)
assert fs.ls("test/test group5.parquet") == [
"test/test group5.parquet/_common_metadata",
"test/test group5.parquet/_metadata",
"test/test group5.parquet/part.0.parquet",
"test/test group5.parquet/part.1.parquet",
"test/test group5.parquet/part.2.parquet",
"test/test group5.parquet/part.3.parquet",
]
df5_test = dd.read_parquet(
"abfs://test/test group5.parquet",
storage_options=STORAGE_OPTIONS,
engine="pyarrow",
).compute()
assert_frame_equal(df5, df5_test)
def test_large_blob(storage):
import tempfile
import hashlib
import io
import shutil
from pathlib import Path
fs = AzureBlobFileSystem(account_name=storage.account_name,
connection_string=CONN_STR)
# create a 20MB byte array, ensure it's larger than blocksizes to force a
# chuncked upload
blob_size = 120_000_000
# blob_size = 2_684_354_560
assert blob_size > fs.blocksize
assert blob_size > AzureBlobFile.DEFAULT_BLOCK_SIZE
data = b"1" * blob_size
_hash = hashlib.md5(data)
expected = _hash.hexdigest()
# create container
fs.mkdir("chunk-container")
# upload the data using fs.open
path = "chunk-container/large-blob.bin"
with fs.open(path, "ab") as dst:
dst.write(data)
assert fs.exists(path)
assert fs.size(path) == blob_size
del data
# download with fs.open
bio = io.BytesIO()
with fs.open(path, "rb") as src:
shutil.copyfileobj(src, bio)
# read back the data and calculate md5
bio.seek(0)
data = bio.read()
_hash = hashlib.md5(data)
result = _hash.hexdigest()
assert expected == result
# do the same but using upload/download and a tempdir
path = path = "chunk-container/large_blob2.bin"
with tempfile.TemporaryDirectory() as td:
local_blob: Path = Path(td) / "large_blob2.bin"
with local_blob.open("wb") as fo:
fo.write(data)
assert local_blob.exists()
assert local_blob.stat().st_size == blob_size
fs.upload(str(local_blob), path)
assert fs.exists(path)
assert fs.size(path) == blob_size
# download now
local_blob.unlink()
fs.download(path, str(local_blob))
assert local_blob.exists()
assert local_blob.stat().st_size == blob_size
