`AWS_S3_ALLOW_UNSAFE_RENAME="true"` in `maybe_set_aws_credentials` incorrectly passed to pyarrow

[rzig]

Hi! In maybe_set_aws_credentials, the AWS_S3_ALLOW_UNSAFE_RENAME is set to true. This causes a the following call, with no credentials provided, to fail:

import dask_deltatable as ddt
ddf = ddt.read_deltalake("s3://my/deltatable", storage_options={})
ddf.head(n=10)

This call currently fails with the following error (full traceback at end):

File [/opt/tljh/user/lib/python3.12/site-packages/s3fs/core.py:526](https://nb.cxproduction.cc/opt/tljh/user/lib/python3.12/site-packages/s3fs/core.py#line=525), in S3FileSystem.set_session(self, refresh, kwargs)
    524 conf = AioConfig(**config_kwargs)
    525 if self.session is None:
--> 526     self.session = aiobotocore.session.AioSession(**self.kwargs)
    528 for parameters in (config_kwargs, self.kwargs, init_kwargs, client_kwargs):
    529     for option in ("region_name", "endpoint_url"):

TypeError: AioSession.__init__() got an unexpected keyword argument 'AWS_S3_ALLOW_UNSAFE_RENAME'

The read_deltalake call should succeed, since I'm executing it on an EC2 instance that has an IAM role configured with permissions to my bucket. I've confirmed I can read this same table with polars:

import polars as pl
df = pl.read_delta("s3://my/deltatable")
df.head()

I am able to work around this temporarily by using path s3a://my/deltatable since maybe_set_aws_credentials explicitly checks for the s3:// prefix (is this also a bug? it should probably check for s3a and s3n too?), but I'm pretty sure this is a bug related to how these credentials eventually make their way to pyarrow. I've included my package versions and full traceback below. If this is indeed a bug, I'm happy to try to write a fix.

Traceback

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[21], line 1
----> 1 ddf.head(n=10)
File /opt/tljh/user/lib/python3.12/site-packages/dask_expr/_collection.py:706, in FrameBase.head(self, n, npartitions, compute)

704 out = new_collection(expr.Head(self, n=n, npartitions=npartitions))

705 if compute:

--> 706     out = out.compute()

707 return out
File /opt/tljh/user/lib/python3.12/site-packages/dask_expr/_collection.py:480, in FrameBase.compute(self, fuse, concatenate, **kwargs)

478     out = out.repartition(npartitions=1)

479 out = out.optimize(fuse=fuse)

--> 480 return DaskMethodsMixin.compute(out, **kwargs)
File /opt/tljh/user/lib/python3.12/site-packages/dask/base.py:372, in DaskMethodsMixin.compute(self, **kwargs)

348 def compute(self, **kwargs):

349     """Compute this dask collection

350

351     This turns a lazy Dask collection into its in-memory equivalent.

(...)

370     dask.compute

371     """

--> 372     (result,) = compute(self, traverse=False, **kwargs)

373     return result
File /opt/tljh/user/lib/python3.12/site-packages/dask/base.py:660, in compute(traverse, optimize_graph, scheduler, get, *args, **kwargs)

657     postcomputes.append(x.dask_postcompute())

659 with shorten_traceback():

--> 660     results = schedule(dsk, keys, **kwargs)

662 return repack([f(r, *a) for r, (f, a) in zip(results, postcomputes)])
File /opt/tljh/user/lib/python3.12/site-packages/dask_deltatable/core.py:76, in _read_delta_partition(filename, schema, fs, columns, filter, pyarrow_to_pandas, **_kwargs)

70 pyarrow_to_pandas["types_mapper"] = _get_type_mapper(

71     pyarrow_to_pandas.get("types_mapper")

72 )

73 pyarrow_to_pandas["ignore_metadata"] = pyarrow_to_pandas.get(

74     "ignore_metadata", False

75 )

---> 76 table = pa_ds.dataset(

77     source=filename,

78     schema=schema,

79     filesystem=fs,

80     format="parquet",

81     partitioning="hive",

82 ).to_table(filter=filter_expression, columns=columns)

83 return table.to_pandas(**pyarrow_to_pandas)
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/dataset.py:794, in dataset(source, schema, format, filesystem, partitioning, partition_base_dir, exclude_invalid_files, ignore_prefixes)

783 kwargs = dict(

784     schema=schema,

785     filesystem=filesystem,

(...)

790     selector_ignore_prefixes=ignore_prefixes

791 )

793 if _is_path_like(source):

--> 794     return _filesystem_dataset(source, **kwargs)

795 elif isinstance(source, (tuple, list)):

796     if all(_is_path_like(elem) or isinstance(elem, FileInfo) for elem in source):
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/dataset.py:476, in _filesystem_dataset(source, schema, filesystem, partitioning, format, partition_base_dir, exclude_invalid_files, selector_ignore_prefixes)

474         fs, paths_or_selector = _ensure_multiple_sources(source, filesystem)

475 else:

--> 476     fs, paths_or_selector = _ensure_single_source(source, filesystem)

478 options = FileSystemFactoryOptions(

479     partitioning=partitioning,

480     partition_base_dir=partition_base_dir,

481     exclude_invalid_files=exclude_invalid_files,

482     selector_ignore_prefixes=selector_ignore_prefixes

483 )

484 factory = FileSystemDatasetFactory(fs, paths_or_selector, format, options)
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/dataset.py:432, in _ensure_single_source(path, filesystem)

429 path = filesystem.normalize_path(path)

431 # retrieve the file descriptor

--> 432 file_info = filesystem.get_file_info(path)

434 # depending on the path type either return with a recursive

435 # directory selector or as a list containing a single file

436 if file_info.type == FileType.Directory:
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/_fs.pyx:590, in pyarrow._fs.FileSystem.get_file_info()
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/error.pxi:155, in pyarrow.lib.pyarrow_internal_check_status()
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/error.pxi:89, in pyarrow.lib.check_status()
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/_fs.pyx:1498, in pyarrow._fs._cb_get_file_info()
File /opt/tljh/user/lib/python3.12/site-packages/pyarrow/fs.py:322, in FSSpecHandler.get_file_info(self, paths)

320 for path in paths:

321     try:

--> 322         info = self.fs.info(path)

323     except FileNotFoundError:

324         infos.append(FileInfo(path, FileType.NotFound))
File /opt/tljh/user/lib/python3.12/site-packages/fsspec/asyn.py:118, in sync_wrapper..wrapper(*args, **kwargs)

115 @functools.wraps(func)

116 def wrapper(*args, **kwargs):

117     self = obj or args[0]

--> 118     return sync(self.loop, func, *args, **kwargs)
File /opt/tljh/user/lib/python3.12/site-packages/fsspec/asyn.py:103, in sync(loop, func, timeout, *args, **kwargs)

101     raise FSTimeoutError from return_result

102 elif isinstance(return_result, BaseException):

--> 103     raise return_result

104 else:

105     return return_result
File /opt/tljh/user/lib/python3.12/site-packages/fsspec/asyn.py:56, in _runner(event, coro, result, timeout)

54     coro = asyncio.wait_for(coro, timeout=timeout)

55 try:

---> 56     result[0] = await coro

57 except Exception as ex:

58     result[0] = ex
File /opt/tljh/user/lib/python3.12/site-packages/s3fs/core.py:1426, in S3FileSystem._info(self, path, bucket, key, refresh, version_id)

1424 if key:

1425     try:

-> 1426         out = await self._call_s3(

1427             "head_object",

1428             self.kwargs,

1429             Bucket=bucket,

1430             Key=key,

1431             **version_id_kw(version_id),

1432             **self.req_kw,

1433         )

1434         return {

1435             "ETag": out.get("ETag", ""),

1436             "LastModified": out.get("LastModified", ""),

(...)

1442             "ContentType": out.get("ContentType"),

1443         }

1444     except FileNotFoundError:
File /opt/tljh/user/lib/python3.12/site-packages/s3fs/core.py:364, in S3FileSystem._call_s3(self, method, *akwarglist, **kwargs)

363 async def _call_s3(self, method, *akwarglist, **kwargs):

--> 364     await self.set_session()

365     s3 = await self.get_s3(kwargs.get("Bucket"))

366     method = getattr(s3, method)
File /opt/tljh/user/lib/python3.12/site-packages/s3fs/core.py:526, in S3FileSystem.set_session(self, refresh, kwargs)

524 conf = AioConfig(**config_kwargs)

525 if self.session is None:

--> 526     self.session = aiobotocore.session.AioSession(**self.kwargs)

528 for parameters in (config_kwargs, self.kwargs, init_kwargs, client_kwargs):

529     for option in ("region_name", "endpoint_url"):
TypeError: AioSession.init() got an unexpected keyword argument 'AWS_S3_ALLOW_UNSAFE_RENAME'

Package Versions

Deltalake: 0.23.2 Dask-deltatable: 0.3.3 Pyarrow: 18.1.0

[jacobtomlinson]

There is a comment that suggests the argument is intended for a delta specific API, so perhaps something has broken upstream.

dask-deltatable/dask_deltatable/utils.py

Line 59 in 3ff657b

# Capitalized is used in delta specific API and lowercase is for S3FileSystem

I've reproduced the issue locally and I've also downgraded pyarrow, deltalake and s3fs to the versions that were available at the time #78 was merged, but I can still produce the issue. So I don't think it's a regression in any of those packages.

It looks like polars may have gone through a similar regression in pola-rs/polars#19878, although it's not immediately clear if it's the same problem.