Add class methods wrapper

docs/api/list.md

@@ -3,6 +3,3 @@
 ::: obstore.list
 ::: obstore.list_with_delimiter
 ::: obstore.list_with_delimiter_async
-::: obstore.ListResult
-::: obstore.ListStream
-::: obstore.ListChunkType

docs/api/store/aws.md

@@ -1,6 +1,9 @@
 # AWS S3
 
 ::: obstore.store.S3Store
+    options:
+        inherited_members: true
+        show_bases: false
 ::: obstore.store.S3Config
     options:
         show_if_no_docstring: true

docs/api/store/azure.md

@@ -1,6 +1,9 @@
 # Microsoft Azure
 
 ::: obstore.store.AzureStore
+    options:
+        inherited_members: true
+        show_bases: false
 ::: obstore.store.AzureConfig
     options:
         show_if_no_docstring: true

docs/api/store/gcs.md

@@ -1,6 +1,9 @@
 # Google Cloud Storage
 
 ::: obstore.store.GCSStore
+    options:
+        inherited_members: true
+        show_bases: false
 ::: obstore.store.GCSConfig
     options:
         show_if_no_docstring: true

docs/api/store/http.md

@@ -1,3 +1,6 @@
 # HTTP
 
 ::: obstore.store.HTTPStore
+    options:
+        inherited_members: true
+        show_bases: false

docs/api/store/local.md

@@ -1,3 +1,6 @@
 # Local
 
 ::: obstore.store.LocalStore
+    options:
+        inherited_members: true
+        show_bases: false

docs/api/store/memory.md

@@ -1,3 +1,6 @@
 # Memory
 
 ::: obstore.store.MemoryStore
+    options:
+        inherited_members: true
+        show_bases: false

docs/cookbook.md

@@ -39,6 +39,19 @@ store = ... # store of your choice
 # Get a stream of Arrow RecordBatches of metadata
 list_stream = obs.list(store, prefix="data", return_arrow=True)
 for record_batch in list_stream:
+    # Perform zero-copy conversion to your arrow-backed library of choice
+    #
+    # To pyarrow:
+    # pyarrow.record_batch(record_batch)
+    #
+    # To polars:
+    # polars.DataFrame(record_batch)
+    #
+    # To pandas (with Arrow-backed data-types):
+    # pyarrow.record_batch(record_batch).to_pandas(types_mapper=pd.ArrowDtype)
+    #
+    # To arro3:
+    # arro3.core.RecordBatch(record_batch)
     print(record_batch.num_rows)
 ```
 

obstore/python/obstore/__init__.py

@@ -1,9 +1,50 @@
 from typing import TYPE_CHECKING
 
-from ._obstore import *
-from ._obstore import ___version
+from . import store
+from ._obstore import (
+    Bytes,
+    ___version,
+    copy,
+    copy_async,
+    delete,
+    delete_async,
+    get,
+    get_async,
+    get_range,
+    get_range_async,
+    get_ranges,
+    get_ranges_async,
+    head,
+    head_async,
+    list,  # noqa: A004
+    list_with_delimiter,
+    list_with_delimiter_async,
+    open_reader,
+    open_reader_async,
+    open_writer,
+    open_writer_async,
+    put,
+    put_async,
+    rename,
+    rename_async,
+    sign,
+    sign_async,
+)
 
 if TYPE_CHECKING:
-    from . import exceptions, store
-
+    from . import _store, exceptions
+    from ._obstore import (
+        HTTP_METHOD,
+        AsyncReadableFile,
+        AsyncWritableFile,
+        Bytes,
+        BytesStream,
+        GetResult,
+        ListChunkType,
+        ListResult,
+        ListStream,
+        ReadableFile,
+        SignCapableStore,
+        WritableFile,
+    )
 __version__: str = ___version()

obstore/python/obstore/_buffered.pyi

@@ -7,7 +7,7 @@ from obspec._attributes import Attributes
 
 from ._bytes import Bytes
 from ._list import ObjectMeta
-from .store import ObjectStore
+from ._store import ObjectStore
 
 if sys.version_info >= (3, 12):
     from collections.abc import Buffer

obstore/python/obstore/_copy.pyi

@@ -1,4 +1,4 @@
-from .store import ObjectStore
+from ._store import ObjectStore
 
 def copy(store: ObjectStore, from_: str, to: str, *, overwrite: bool = True) -> None:
     """Copy an object from one path to another in the same object store.

obstore/python/obstore/_delete.pyi

@@ -1,6 +1,6 @@
 from collections.abc import Sequence
 
-from .store import ObjectStore
+from ._store import ObjectStore
 
 def delete(store: ObjectStore, paths: str | Sequence[str]) -> None:
     """Delete the object at the specified location(s).

obstore/python/obstore/_get.pyi

@@ -6,7 +6,7 @@ from obspec._get import GetOptions
 
 from ._bytes import Bytes
 from ._list import ObjectMeta
-from .store import ObjectStore
+from ._store import ObjectStore
 
 class GetResult:
     """Result for a get request.

obstore/python/obstore/_list.pyi

@@ -1,77 +1,12 @@
-# ruff: noqa: UP006
-# ruff: noqa: UP035
-# Use `list` instead of `List` for type annotation
-# `typing.List` is deprecated, use `list` instead
-# ruff: noqa: A001
-# Variable `list` is shadowing a Python builtinRuff
+# ruff: noqa: A001, UP006, UP035
 
-from typing import Generic, List, Literal, Self, TypedDict, TypeVar, overload
+from typing import List, Literal, overload
 
 from arro3.core import RecordBatch, Table
+from obspec._list import ListResult, ListStream
 from obspec._meta import ObjectMeta
 
-from .store import ObjectStore
-
-ListChunkType = TypeVar("ListChunkType", List[ObjectMeta], RecordBatch, Table)  # noqa: PYI001
-"""The data structure used for holding list results.
-
-By default, listing APIs return a `list` of [`ObjectMeta`][obspec.ObjectMeta]. However
-for improved performance when listing large buckets, you can pass `return_arrow=True`.
-Then an Arrow `RecordBatch` will be returned instead.
-
-This implements [`obspec.ListChunkType_co`][], but is redefined here to specialize the
-exact instance of the Arrow return type, given that in the obstore implementation, an
-[`arro3.core.RecordBatch`][] or [`arro3.core.Table`][] will always be returned.
-"""
-
-class ListResult(TypedDict, Generic[ListChunkType]):
-    """Result of a list call.
-
-    Includes objects, prefixes (directories) and a token for the next set of results.
-    Individual result sets may be limited to 1,000 objects based on the underlying
-    object storage's limitations.
-
-    This implements [`obspec.ListResult`][].
-    """
-
-    common_prefixes: List[str]
-    """Prefixes that are common (like directories)"""
-
-    objects: ListChunkType
-    """Object metadata for the listing"""
-
-class ListStream(Generic[ListChunkType]):
-    """A stream of [ObjectMeta][obspec.ObjectMeta] that can be polled in a sync or
-    async fashion.
-
-    This implements [`obspec.ListStream`][].
-    """  # noqa: D205
-
-    def __aiter__(self) -> Self:
-        """Return `Self` as an async iterator."""
-
-    def __iter__(self) -> Self:
-        """Return `Self` as an async iterator."""
-
-    async def collect_async(self) -> ListChunkType:
-        """Collect all remaining ObjectMeta objects in the stream.
-
-        This ignores the `chunk_size` parameter from the `list` call and collects all
-        remaining data into a single chunk.
-        """
-
-    def collect(self) -> ListChunkType:
-        """Collect all remaining ObjectMeta objects in the stream.
-
-        This ignores the `chunk_size` parameter from the `list` call and collects all
-        remaining data into a single chunk.
-        """
-
-    async def __anext__(self) -> ListChunkType:
-        """Return the next chunk of ObjectMeta in the stream."""
-
-    def __next__(self) -> ListChunkType:
-        """Return the next chunk of ObjectMeta in the stream."""
+from ._store import ObjectStore
 
 @overload
 def list(
@@ -163,7 +98,7 @@ def list(
     !!! note
         There is no async version of this method, because `list` is not async under the
         hood, rather it only instantiates a stream, which can be polled in synchronous
-        or asynchronous fashion. See [`ListStream`][obstore.ListStream].
+        or asynchronous fashion. See [`ListStream`][obspec.ListStream].
 
     Args:
         store: The ObjectStore instance to use.
@@ -174,8 +109,8 @@ def list(
         chunk_size: The number of items to collect per chunk in the returned
             (async) iterator. All chunks except for the last one will have this many
             items. This is ignored in the
-            [`collect`][obstore.ListStream.collect] and
-            [`collect_async`][obstore.ListStream.collect_async] methods of
+            [`collect`][obspec.ListStream.collect] and
+            [`collect_async`][obspec.ListStream.collect_async] methods of
             `ListStream`.
         return_arrow: If `True`, return each batch of list items as an Arrow
             `RecordBatch`, not as a list of Python `dict`s. Arrow removes serialization

obstore/python/obstore/_obstore.pyi

@@ -1,3 +1,4 @@
+from . import _store as _store
 from ._buffered import AsyncReadableFile as AsyncReadableFile
 from ._buffered import AsyncWritableFile as AsyncWritableFile
 from ._buffered import ReadableFile as ReadableFile
@@ -21,16 +22,14 @@ from ._get import get_ranges as get_ranges
 from ._get import get_ranges_async as get_ranges_async
 from ._head import head as head
 from ._head import head_async as head_async
-from ._list import ListChunkType as ListChunkType
-from ._list import ListResult as ListResult
-from ._list import ListStream as ListStream
 from ._list import list as list  # noqa: A004
 from ._list import list_with_delimiter as list_with_delimiter
 from ._list import list_with_delimiter_async as list_with_delimiter_async
 from ._put import put as put
 from ._put import put_async as put_async
 from ._rename import rename as rename
 from ._rename import rename_async as rename_async
+from ._scheme import parse_scheme as parse_scheme
 from ._sign import HTTP_METHOD as HTTP_METHOD
 from ._sign import SignCapableStore as SignCapableStore
 from ._sign import sign as sign

obstore/python/obstore/_rename.pyi

@@ -1,4 +1,4 @@
-from .store import ObjectStore
+from ._store import ObjectStore
 
 def rename(store: ObjectStore, from_: str, to: str, *, overwrite: bool = True) -> None:
     """Move an object from one path to another in the same object store.

obstore/python/obstore/_scheme.pyi

@@ -0,0 +1,5 @@
+from typing import Literal
+
+def parse_scheme(
+    url: str,
+) -> Literal["s3", "gcs", "http", "local", "memory", "azure"]: ...

obstore/python/obstore/store/__init__.pyi → obstore/python/obstore/_store/__init__.pyi

@@ -1,7 +1,7 @@
 # TODO: move to reusable types package
 from collections.abc import Callable
 from pathlib import Path
-from typing import Any, TypeAlias, Unpack, overload
+from typing import Any, Self, TypeAlias, Unpack, overload
 
 from ._aws import S3Config as S3Config
 from ._aws import S3Credential as S3Credential
@@ -154,7 +154,7 @@ class LocalStore:
         *,
         automatic_cleanup: bool = False,
         mkdir: bool = False,
-    ) -> LocalStore:
+    ) -> Self:
         """Construct a new LocalStore from a `file://` URL.
 
         **Examples:**

obstore/python/obstore/store/_aws.pyi → obstore/python/obstore/_store/_aws.pyi

@@ -1,6 +1,15 @@
 from collections.abc import Coroutine
 from datetime import datetime
-from typing import Any, Literal, NotRequired, Protocol, TypeAlias, TypedDict, Unpack
+from typing import (
+    Any,
+    Literal,
+    NotRequired,
+    Protocol,
+    Self,
+    TypeAlias,
+    TypedDict,
+    Unpack,
+)
 
 from ._client import ClientConfig
 from ._retry import RetryConfig
@@ -501,7 +510,7 @@ class S3Store:
         retry_config: RetryConfig | None = None,
         credential_provider: S3CredentialProvider | None = None,
         **kwargs: Unpack[S3Config],
-    ) -> S3Store:
+    ) -> Self:
         """Parse available connection info from a well-known storage URL.
 
         The supported url schemes are:

obstore/python/obstore/store/_azure.pyi → obstore/python/obstore/_store/_azure.pyi

@@ -1,6 +1,6 @@
 from collections.abc import Coroutine
 from datetime import datetime
-from typing import Any, Protocol, TypeAlias, TypedDict, Unpack
+from typing import Any, Protocol, Self, TypeAlias, TypedDict, Unpack
 
 from ._client import ClientConfig
 from ._retry import RetryConfig
@@ -353,7 +353,7 @@ class AzureStore:
         retry_config: RetryConfig | None = None,
         credential_provider: AzureCredentialProvider | None = None,
         **kwargs: Unpack[AzureConfig],
-    ) -> AzureStore:
+    ) -> Self:
         """Construct a new AzureStore with values populated from a well-known storage URL.
 
         The supported url schemes are:

obstore/python/obstore/store/_gcs.pyi → obstore/python/obstore/_store/_gcs.pyi

@@ -1,6 +1,6 @@
 from collections.abc import Coroutine
 from datetime import datetime
-from typing import Any, Protocol, TypedDict, Unpack
+from typing import Any, Protocol, Self, TypedDict, Unpack
 
 from ._client import ClientConfig
 from ._retry import RetryConfig
@@ -164,7 +164,7 @@ class GCSStore:
         retry_config: RetryConfig | None = None,
         credential_provider: GCSCredentialProvider | None = None,
         **kwargs: Unpack[GCSConfig],
-    ) -> GCSStore:
+    ) -> Self:
         """Construct a new GCSStore with values populated from a well-known storage URL.
 
         The supported url schemes are: