fix: correct UUIDType partition representation for BucketTransform

.github/ISSUE_TEMPLATE/iceberg_bug_report.yml

@@ -28,7 +28,8 @@ body:
       description: What Apache Iceberg version are you using?
       multiple: false
       options:
-        - "0.9.0 (latest release)"
+        - "0.9.1 (latest release)"
+        - "0.9.0"
         - "0.8.1"
         - "0.8.0"
         - "0.7.1"

mkdocs/docs/configuration.md

@@ -95,6 +95,7 @@ Iceberg works with the concept of a FileIO which is a pluggable module for readi
 - **hdfs**: `PyArrowFileIO`
 - **abfs**, **abfss**: `FsspecFileIO`
 - **oss**: `PyArrowFileIO`
+- **hf**: `FsspecFileIO`
 
 You can also set the FileIO explicitly:
 
@@ -193,6 +194,17 @@ PyIceberg uses [S3FileSystem](https://arrow.apache.org/docs/python/generated/pya
 
 <!-- markdown-link-check-enable-->
 
+### Hugging Face
+
+<!-- markdown-link-check-disable -->
+
+| Key         | Example                  | Description                                               |
+| ----------- | ------------------------ | --------------------------------------------------------- |
+| hf.endpoint | <https://huggingface.co> | Configure the endpoint for Hugging Face                   |
+| hf.token    | hf_xxx                   | The Hugging Face token to access HF Datasets repositories |
+
+<!-- markdown-link-check-enable-->
+
 ### PyArrow
 
 <!-- markdown-link-check-disable -->

pyiceberg/avro/writer.py

@@ -32,6 +32,7 @@
     List,
     Optional,
     Tuple,
+    Union,
 )
 from uuid import UUID
 
@@ -121,8 +122,23 @@ def write(self, encoder: BinaryEncoder, val: Any) -> None:
 
 @dataclass(frozen=True)
 class UUIDWriter(Writer):
-    def write(self, encoder: BinaryEncoder, val: UUID) -> None:
-        encoder.write(val.bytes)
+    def write(self, encoder: BinaryEncoder, val: Union[UUID, str, bytes]) -> None:
+        if isinstance(val, UUID):
+            encoder.write(val.bytes)
+        elif isinstance(val, bytes):
+            encoder.write(val)
+        elif isinstance(val, str):
+            print(f"UUID string: {val=}")
+            if val.startswith("b'") and val.endswith("'"):
+                # Handle string representation of bytes
+                # Convert the escaped string to actual bytes
+                byte_string = val[2:-1].encode("utf-8").decode("unicode_escape").encode("latin1")
+                encoder.write(UUID(bytes=byte_string).bytes)
+            else:
+                # Regular UUID string
+                encoder.write(UUID(val).bytes)
+        else:
+            raise TypeError(f"Expected UUID, bytes, or string, got {type(val)}")
 
 
 @dataclass(frozen=True)