Skip to content

Added force virtual addressing configuration for S3, Alibaba OSS protocol to use PyArrowFileIO #1392

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Dec 9, 2024
Merged

Added force virtual addressing configuration for S3, Alibaba OSS protocol to use PyArrowFileIO #1392

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
5b24172
Added force virtual addressing configuration for S3. Also added oss a…
helmiazizm Dec 2, 2024
360e305
Rewrote force virtual addressing as written in PyArrow documentation
helmiazizm Dec 3, 2024
31291f4
Added the missing r2 key value in schema_to_file_io
helmiazizm Dec 3, 2024
6575552
Removed R2 protocol for now
helmiazizm Dec 3, 2024
282bd47
Linter fix
helmiazizm Dec 4, 2024
7315525
Updated documentation for OSS support
helmiazizm Dec 4, 2024
727666d
Another linter fix
helmiazizm Dec 4, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions mkdocs/docs/configuration.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,6 +88,7 @@ Iceberg works with the concept of a FileIO which is a pluggable module for readi
- **file**: `PyArrowFileIO`
- **hdfs**: `PyArrowFileIO`
- **abfs**, **abfss**: `FsspecFileIO`
- **oss**: `PyArrowFileIO`

You can also set the FileIO explicitly:

Expand Down Expand Up @@ -115,6 +116,7 @@ For the FileIO there are several configuration options available:
| s3.region | us-west-2 | Sets the region of the bucket |
| s3.proxy-uri | <http://my.proxy.com:8080> | Configure the proxy server to be used by the FileIO. |
| s3.connect-timeout | 60.0 | Configure socket connection timeout, in seconds. |
| s3.force-virtual-addressing | False | Whether to use virtual addressing of buckets. If true, then virtual addressing is always enabled. If false, then virtual addressing is only enabled if endpoint_override is empty. This can be used for non-AWS backends that only support virtual hosted-style access. |

<!-- markdown-link-check-enable-->

Expand Down Expand Up @@ -167,6 +169,22 @@ For the FileIO there are several configuration options available:

<!-- markdown-link-check-enable-->

### Alibaba Cloud Object Storage Service (OSS)

<!-- markdown-link-check-disable -->

PyIceberg uses [S3FileSystem](https://arrow.apache.org/docs/python/generated/pyarrow.fs.S3FileSystem.html) class to connect to OSS bucket as the service is [compatible with S3 SDK](https://www.alibabacloud.com/help/en/oss/developer-reference/use-amazon-s3-sdks-to-access-oss) as long as the endpoint is addressed with virtual hosted style.

| Key | Example | Description |
| -------------------- | ------------------- | ------------------------------------------------ |
| s3.endpoint | <https://s3.oss-your-bucket-region.aliyuncs.com/> | Configure an endpoint of the OSS service for the FileIO to access. Be sure to use S3 compatible endpoint as given in the example. |
| s3.access-key-id | admin | Configure the static access key id used to access the FileIO. |
| s3.secret-access-key | password | Configure the static secret access key used to access the FileIO. |
| s3.session-token | AQoDYXdzEJr... | Configure the static session token used to access the FileIO. |
| s3.force-virtual-addressing | True | Whether to use virtual addressing of buckets. This must be set to True as OSS can only be accessed with virtual hosted style address. |

<!-- markdown-link-check-enable-->

### PyArrow

<!-- markdown-link-check-disable -->
Expand Down
2 changes: 2 additions & 0 deletions pyiceberg/io/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,6 +74,7 @@
S3_SIGNER_ENDPOINT_DEFAULT = "v1/aws/s3/sign"
S3_ROLE_ARN = "s3.role-arn"
S3_ROLE_SESSION_NAME = "s3.role-session-name"
S3_FORCE_VIRTUAL_ADDRESSING = "s3.force-virtual-addressing"
HDFS_HOST = "hdfs.host"
HDFS_PORT = "hdfs.port"
HDFS_USER = "hdfs.user"
Expand Down Expand Up @@ -304,6 +305,7 @@ def delete(self, location: Union[str, InputFile, OutputFile]) -> None:
"s3": [ARROW_FILE_IO, FSSPEC_FILE_IO],
"s3a": [ARROW_FILE_IO, FSSPEC_FILE_IO],
"s3n": [ARROW_FILE_IO, FSSPEC_FILE_IO],
"oss": [ARROW_FILE_IO],
"gs": [ARROW_FILE_IO],
"file": [ARROW_FILE_IO, FSSPEC_FILE_IO],
"hdfs": [ARROW_FILE_IO],
Expand Down
6 changes: 5 additions & 1 deletion pyiceberg/io/pyarrow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,6 +102,7 @@
S3_ACCESS_KEY_ID,
S3_CONNECT_TIMEOUT,
S3_ENDPOINT,
S3_FORCE_VIRTUAL_ADDRESSING,
S3_PROXY_URI,
S3_REGION,
S3_ROLE_ARN,
Expand Down Expand Up @@ -350,7 +351,7 @@ def parse_location(location: str) -> Tuple[str, str, str]:
return uri.scheme, uri.netloc, f"{uri.netloc}{uri.path}"

def _initialize_fs(self, scheme: str, netloc: Optional[str] = None) -> FileSystem:
if scheme in {"s3", "s3a", "s3n"}:
if scheme in {"s3", "s3a", "s3n", "oss"}:
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lets also update the docs to mention oss
https://py.iceberg.apache.org/configuration/#fileio

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we also add oss to fsspec?

iceberg-python/pyiceberg/io/fsspec.py

Lines 245 to 255 in 68e17af

SCHEME_TO_FS = {
"": _file,
"file": _file,
"s3": _s3,
"s3a": _s3,
"s3n": _s3,
"abfs": _adls,
"abfss": _adls,
"gs": _gs,
"gcs": _gs,
}

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks like theres https://github.com/fsspec/ossfs

Copy link
Contributor Author

@helmiazizm helmiazizm Dec 4, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks like theres https://github.com/fsspec/ossfs

The project hasn't been maintained for almost a year now, and when I tried it on DuckDB it returns an unknown dependency error (?). I'd say it's probably safer to use s3fs than ossfs (not that I've tried it, though). I'll try making a new pull request for that later.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

makes sense, thanks for looking into it! lets keep the scope just pyarrow for now

from pyarrow.fs import S3FileSystem

client_kwargs: Dict[str, Any] = {
Expand All @@ -373,6 +374,9 @@ def _initialize_fs(self, scheme: str, netloc: Optional[str] = None) -> FileSyste
if session_name := get_first_property_value(self.properties, S3_ROLE_SESSION_NAME, AWS_ROLE_SESSION_NAME):
client_kwargs["session_name"] = session_name

if force_virtual_addressing := self.properties.get(S3_FORCE_VIRTUAL_ADDRESSING):
client_kwargs["force_virtual_addressing"] = property_as_bool(self.properties, force_virtual_addressing, False)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

force_virtual_addressing docs
https://arrow.apache.org/docs/python/generated/pyarrow.fs.S3FileSystem.html


return S3FileSystem(**client_kwargs)
elif scheme in ("hdfs", "viewfs"):
from pyarrow.fs import HadoopFileSystem
Expand Down
Loading