nucliadb-utils 6.9.1.post5229__py3-none-any.whl → 6.10.0.post5732__py3-none-any.whl

nucliadb_utils/storages/gcs.py

@@ -23,10 +23,11 @@ import asyncio
 import base64
 import json
 import socket
+from collections.abc import AsyncGenerator, AsyncIterator
 from concurrent.futures import ThreadPoolExecutor
 from copy import deepcopy
 from datetime import datetime
-from typing import Any, AsyncGenerator, AsyncIterator, Dict, List, Optional, cast
+from typing import Any, cast
 from urllib.parse import quote_plus
 
 import aiohttp
@@ -153,7 +154,7 @@ class GCSStorageField(StorageField):
                 assert data["resource"]["name"] == destination_uri
 
     @storage_ops_observer.wrap({"type": "iter_data"})
-    async def iter_data(self, range: Optional[Range] = None) -> AsyncGenerator[bytes, None]:
+    async def iter_data(self, range: Range | None = None) -> AsyncGenerator[bytes]:
         attempt = 1
         while True:
             try:
@@ -170,13 +171,13 @@ class GCSStorageField(StorageField):
                     raise
                 wait_time = 2 ** (attempt - 1)
                 logger.warning(
-                    f"Error downloading from GCP. Retrying ({attempt} of {MAX_TRIES}) after {wait_time} seconds. Error: {ex}"  # noqa
+                    f"Error downloading from GCP. Retrying ({attempt} of {MAX_TRIES}) after {wait_time} seconds. Error: {ex}"
                 )
                 await asyncio.sleep(wait_time)
                 attempt += 1
 
     @storage_ops_observer.wrap({"type": "inner_iter_data"})
-    async def _inner_iter_data(self, range: Optional[Range] = None):
+    async def _inner_iter_data(self, range: Range | None = None):
         """
         Iterate through object data.
         """
@@ -322,7 +323,7 @@ class GCSStorageField(StorageField):
         async with self.storage.session.put(
             self.field.resumable_uri, headers=headers, data=data
         ) as call:
-            text = await call.text()  # noqa
+            text = await call.text()
             if call.status not in [200, 201, 308]:
                 if call.status == 410:
                     raise ResumableUploadGone(text)
@@ -377,7 +378,7 @@ class GCSStorageField(StorageField):
         max_tries=MAX_TRIES,
     )
     @storage_ops_observer.wrap({"type": "exists"})
-    async def exists(self) -> Optional[ObjectMetadata]:
+    async def exists(self) -> ObjectMetadata | None:
         """
         Existence can be checked either with a CloudFile data in the field attribute
         or own StorageField key and bucket. Field takes precendece
@@ -425,23 +426,23 @@ class GCSStorageField(StorageField):
 
 class GCSStorage(Storage):
     field_klass = GCSStorageField
-    _session: Optional[aiohttp.ClientSession] = None
+    _session: aiohttp.ClientSession | None = None
     _credentials = None
     _json_credentials = None
     chunk_size = CHUNK_SIZE
 
     def __init__(
         self,
-        account_credentials: Optional[str] = None,
-        bucket: Optional[str] = None,
-        location: Optional[str] = None,
-        project: Optional[str] = None,
-        executor: Optional[ThreadPoolExecutor] = None,
-        deadletter_bucket: Optional[str] = None,
-        indexing_bucket: Optional[str] = None,
-        labels: Optional[Dict[str, str]] = None,
+        account_credentials: str | None = None,
+        bucket: str | None = None,
+        location: str | None = None,
+        project: str | None = None,
+        executor: ThreadPoolExecutor | None = None,
+        deadletter_bucket: str | None = None,
+        indexing_bucket: str | None = None,
+        labels: dict[str, str] | None = None,
         url: str = "https://www.googleapis.com",
-        scopes: Optional[List[str]] = None,
+        scopes: list[str] | None = None,
         anonymous: bool = False,
     ):
         if anonymous:
@@ -533,7 +534,7 @@ class GCSStorage(Storage):
     @storage_ops_observer.wrap({"type": "delete"})
     async def delete_upload(self, uri: str, bucket_name: str):
         if uri:
-            url = "{}/{}/o/{}".format(self.object_base_url, bucket_name, quote_plus(uri))
+            url = f"{self.object_base_url}/{bucket_name}/o/{quote_plus(uri)}"
             headers = await self.get_access_headers()
             async with self.session.delete(url, headers=headers) as resp:
                 if resp.status in (200, 204, 404):
@@ -569,7 +570,7 @@ class GCSStorage(Storage):
         max_tries=MAX_TRIES,
     )
     @storage_ops_observer.wrap({"type": "create_bucket"})
-    async def create_bucket(self, bucket_name: str, kbid: Optional[str] = None):
+    async def create_bucket(self, bucket_name: str, kbid: str | None = None):
         if await self.check_exists(bucket_name=bucket_name):
             return
 
@@ -671,9 +672,9 @@ class GCSStorage(Storage):
         return deleted, conflict
 
     async def iterate_objects(
-        self, bucket: str, prefix: str, start: Optional[str] = None
-    ) -> AsyncGenerator[ObjectInfo, None]:
-        url = "{}/{}/o".format(self.object_base_url, bucket)
+        self, bucket: str, prefix: str, start: str | None = None
+    ) -> AsyncGenerator[ObjectInfo]:
+        url = f"{self.object_base_url}/{bucket}/o"
         headers = await self.get_access_headers()
         params = {"prefix": prefix}
         if start:

nucliadb_utils/storages/local.py

@@ -23,8 +23,8 @@ import glob
 import json
 import os
 import shutil
+from collections.abc import AsyncGenerator, AsyncIterator
 from datetime import datetime
-from typing import AsyncGenerator, AsyncIterator, Optional
 
 import aiofiles
 
@@ -38,7 +38,7 @@ class LocalStorageField(StorageField):
     storage: LocalStorage
     _handler = None
 
-    def metadata_key(self, uri: Optional[str] = None):
+    def metadata_key(self, uri: str | None = None):
         if uri is None and self.field is not None:
             return f"{self.field.uri}.metadata"
         elif uri is None and self.key is not None:
@@ -73,7 +73,7 @@ class LocalStorageField(StorageField):
         destination_path = f"{destination_bucket_path}/{destination_uri}"
         shutil.copy(origin_path, destination_path)
 
-    async def iter_data(self, range: Optional[Range] = None) -> AsyncGenerator[bytes, None]:
+    async def iter_data(self, range: Range | None = None) -> AsyncGenerator[bytes]:
         range = range or Range()
         key = self.field.uri if self.field else self.key
         if self.field is None:
@@ -191,7 +191,7 @@ class LocalStorageField(StorageField):
         self.field.ClearField("offset")
         self.field.ClearField("upload_uri")
 
-    async def exists(self) -> Optional[ObjectMetadata]:
+    async def exists(self) -> ObjectMetadata | None:
         file_path = self.storage.get_file_path(self.bucket, self.key)
         metadata_path = self.metadata_key(file_path)
         if os.path.exists(metadata_path):
@@ -218,7 +218,7 @@ class LocalStorage(Storage):
     field_klass = LocalStorageField
     chunk_size = CHUNK_SIZE
 
-    def __init__(self, local_testing_files: str, indexing_bucket: Optional[str] = None):
+    def __init__(self, local_testing_files: str, indexing_bucket: str | None = None):
         self.local_testing_files = local_testing_files.rstrip("/")
         self.bucket_format = "ndb_{kbid}"
         self.source = CloudFile.LOCAL
@@ -281,8 +281,8 @@ class LocalStorage(Storage):
         return deleted
 
     async def iterate_objects(
-        self, bucket: str, prefix: str, start: Optional[str] = None
-    ) -> AsyncGenerator[ObjectInfo, None]:
+        self, bucket: str, prefix: str, start: str | None = None
+    ) -> AsyncGenerator[ObjectInfo]:
         bucket_path = self.get_bucket_path(bucket)
         pathname = f"{self.get_file_path(bucket, prefix)}**/*"
         for key in sorted(glob.glob(pathname, recursive=True)):
@@ -296,7 +296,7 @@ class LocalStorage(Storage):
                 continue
             yield ObjectInfo(name=name)
 
-    async def download(self, bucket: str, key: str, range: Optional[Range] = None):
+    async def download(self, bucket: str, key: str, range: Range | None = None):
         key_path = self.get_file_path(bucket, key)
         if not os.path.exists(key_path):
             return

nucliadb_utils/storages/nuclia.py

@@ -17,7 +17,6 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program. If not, see <http://www.gnu.org/licenses/>.
 #
-from typing import Optional
 
 import aiohttp
 
@@ -32,7 +31,7 @@ class NucliaStorage:
         self,
         nuclia_public_url: str,
         nuclia_zone: str,
-        service_account: Optional[str] = None,
+        service_account: str | None = None,
     ):
         self.service_account = service_account
         self.nuclia_public_url = nuclia_public_url.format(zone=nuclia_zone)

nucliadb_utils/storages/object_store.py

@@ -19,7 +19,7 @@
 #
 
 import abc
-from typing import AsyncGenerator, AsyncIterator, Optional, Union
+from collections.abc import AsyncGenerator, AsyncIterator
 
 from nucliadb_utils.storages.utils import ObjectInfo, ObjectMetadata, Range
 
@@ -37,7 +37,7 @@ class ObjectStore(abc.ABC, metaclass=abc.ABCMeta):
     async def finalize(self) -> None: ...
 
     @abc.abstractmethod
-    async def bucket_create(self, bucket: str, labels: Optional[dict[str, str]] = None) -> bool:
+    async def bucket_create(self, bucket: str, labels: dict[str, str] | None = None) -> bool:
         """
         Create a new bucket in the object storage. Labels the bucket with the given labels if provided.
         Returns True if the bucket was created, False if it already existed.
@@ -93,7 +93,7 @@ class ObjectStore(abc.ABC, metaclass=abc.ABCMeta):
         self,
         bucket: str,
         key: str,
-        data: Union[bytes, AsyncGenerator[bytes, None]],
+        data: bytes | AsyncGenerator[bytes, None],
         metadata: ObjectMetadata,
     ) -> None: ...
 
@@ -114,14 +114,14 @@ class ObjectStore(abc.ABC, metaclass=abc.ABCMeta):
 
     @abc.abstractmethod
     async def download_stream(
-        self, bucket: str, key: str, range: Optional[Range] = None
+        self, bucket: str, key: str, range: Range | None = None
     ) -> AsyncGenerator[bytes, None]:
         raise NotImplementedError()
         yield b""
 
     @abc.abstractmethod
     async def iterate(
-        self, bucket: str, prefix: str, start: Optional[str] = None
+        self, bucket: str, prefix: str, start: str | None = None
     ) -> AsyncGenerator[ObjectInfo, None]:
         raise NotImplementedError()
         yield ObjectInfo(name="")
@@ -132,7 +132,7 @@ class ObjectStore(abc.ABC, metaclass=abc.ABCMeta):
     @abc.abstractmethod
     async def upload_multipart_start(
         self, bucket: str, key: str, metadata: ObjectMetadata
-    ) -> Optional[str]:
+    ) -> str | None:
         """
         Start a multipart upload. May return the url for the resumable upload.
         """

nucliadb_utils/storages/s3.py

@@ -20,9 +20,9 @@
 from __future__ import annotations
 
 import base64
+from collections.abc import AsyncGenerator, AsyncIterator
 from contextlib import AsyncExitStack
 from datetime import datetime
-from typing import AsyncGenerator, AsyncIterator, Optional
 
 import aiobotocore  # type: ignore
 import aiohttp
@@ -55,12 +55,12 @@ RETRIABLE_EXCEPTIONS = (
 POLICY_DELETE = {
     "Rules": [
         {
-            "Expiration": {"Days": 1},
+            "Expiration": {"Days": 7},
             "ID": "FullDelete",
             "Filter": {"Prefix": ""},
             "Status": "Enabled",
-            "NoncurrentVersionExpiration": {"NoncurrentDays": 1},
-            "AbortIncompleteMultipartUpload": {"DaysAfterInitiation": 1},
+            "NoncurrentVersionExpiration": {"NoncurrentDays": 7},
+            "AbortIncompleteMultipartUpload": {"DaysAfterInitiation": 7},
         },
         {
             "Expiration": {"ExpiredObjectDeleteMarker": True},
@@ -86,7 +86,7 @@ class S3StorageField(StorageField):
         self,
         uri,
         bucket,
-        range: Optional[Range] = None,
+        range: Range | None = None,
     ):
         range = range or Range()
         if range.any():
@@ -103,7 +103,7 @@ class S3StorageField(StorageField):
                 raise
 
     @s3_ops_observer.wrap({"type": "iter_data"})
-    async def iter_data(self, range: Optional[Range] = None) -> AsyncGenerator[bytes, None]:
+    async def iter_data(self, range: Range | None = None) -> AsyncGenerator[bytes]:
         # Suports field and key based iter
         uri = self.field.uri if self.field else self.key
         if self.field is None:
@@ -179,7 +179,7 @@ class S3StorageField(StorageField):
             Bucket=bucket_name,
             Key=upload_id,
             Metadata={
-                "FILENAME": cf.filename,
+                "base64_filename": base64.b64encode(cf.filename.encode()).decode(),
                 "SIZE": str(cf.size),
                 "CONTENT_TYPE": cf.content_type,
             },
@@ -277,7 +277,7 @@ class S3StorageField(StorageField):
         )
 
     @s3_ops_observer.wrap({"type": "exists"})
-    async def exists(self) -> Optional[ObjectMetadata]:
+    async def exists(self) -> ObjectMetadata | None:
         """
         Existence can be checked either with a CloudFile data in the field attribute
         or own StorageField key and bucket. Field takes precendece
@@ -348,18 +348,18 @@ class S3Storage(Storage):
 
     def __init__(
         self,
-        aws_client_id: Optional[str] = None,
-        aws_client_secret: Optional[str] = None,
-        deadletter_bucket: Optional[str] = None,
-        indexing_bucket: Optional[str] = None,
-        endpoint_url: Optional[str] = None,
+        aws_client_id: str | None = None,
+        aws_client_secret: str | None = None,
+        deadletter_bucket: str | None = None,
+        indexing_bucket: str | None = None,
+        endpoint_url: str | None = None,
         verify_ssl: bool = True,
         use_ssl: bool = True,
-        region_name: Optional[str] = None,
-        kms_key_id: Optional[str] = None,
+        region_name: str | None = None,
+        kms_key_id: str | None = None,
         max_pool_connections: int = 30,
-        bucket: Optional[str] = None,
-        bucket_tags: Optional[dict[str, str]] = None,
+        bucket: str | None = None,
+        bucket_tags: dict[str, str] | None = None,
     ):
         self.source = CloudFile.S3
         self.deadletter_bucket = deadletter_bucket
@@ -394,7 +394,7 @@ class S3Storage(Storage):
             self._session = get_session()
         return self._session
 
-    async def initialize(self: "S3Storage") -> None:
+    async def initialize(self: S3Storage) -> None:
         session = AioSession()
         self._s3aioclient: AioBaseClient = await self._exit_stack.enter_async_context(
             session.create_client("s3", **self.opts)
@@ -425,8 +425,8 @@ class S3Storage(Storage):
             raise AttributeError("No valid uri")
 
     async def iterate_objects(
-        self, bucket: str, prefix: str = "/", start: Optional[str] = None
-    ) -> AsyncGenerator[ObjectInfo, None]:
+        self, bucket: str, prefix: str = "/", start: str | None = None
+    ) -> AsyncGenerator[ObjectInfo]:
         paginator = self._s3aioclient.get_paginator("list_objects")
         async for result in paginator.paginate(
             Bucket=bucket, Prefix=prefix, PaginationConfig={"StartingToken": start}
@@ -531,9 +531,9 @@ async def bucket_exists(client: AioSession, bucket_name: str) -> bool:
 async def create_bucket(
     client: AioSession,
     bucket_name: str,
-    bucket_tags: Optional[dict[str, str]] = None,
-    region_name: Optional[str] = None,
-    kms_key_id: Optional[str] = None,
+    bucket_tags: dict[str, str] | None = None,
+    region_name: str | None = None,
+    kms_key_id: str | None = None,
 ):
     bucket_creation_options = {}
     if region_name is not None:

nucliadb_utils/storages/settings.py

@@ -18,23 +18,22 @@
 # along with this program. If not, see <http://www.gnu.org/licenses/>.
 
 import os
-from typing import Dict, Optional
 
 from pydantic_settings import BaseSettings
 
 
 class Settings(BaseSettings):
-    gcs_deadletter_bucket: Optional[str] = None
-    gcs_indexing_bucket: Optional[str] = None
+    gcs_deadletter_bucket: str | None = None
+    gcs_indexing_bucket: str | None = None
 
     gcs_threads: int = 3
-    gcs_labels: Dict[str, str] = {}
+    gcs_labels: dict[str, str] = {}
 
-    s3_deadletter_bucket: Optional[str] = None
-    s3_indexing_bucket: Optional[str] = None
+    s3_deadletter_bucket: str | None = None
+    s3_indexing_bucket: str | None = None
 
-    azure_deadletter_bucket: Optional[str] = None
-    azure_indexing_bucket: Optional[str] = None
+    azure_deadletter_bucket: str | None = None
+    azure_indexing_bucket: str | None = None
 
     local_testing_files: str = os.path.dirname(__file__)
 

nucliadb_utils/storages/storage.py

@@ -24,15 +24,10 @@ import asyncio
 import base64
 import hashlib
 import uuid
+from collections.abc import AsyncGenerator, AsyncIterator
 from io import BytesIO
 from typing import (
     Any,
-    AsyncGenerator,
-    AsyncIterator,
-    List,
-    Optional,
-    Type,
-    Union,
     cast,
 )
 
@@ -65,14 +60,14 @@ class StorageField(abc.ABC, metaclass=abc.ABCMeta):
     storage: Storage
     bucket: str
     key: str
-    field: Optional[CloudFile] = None
+    field: CloudFile | None = None
 
     def __init__(
         self,
         storage: Storage,
         bucket: str,
         fullkey: str,
-        field: Optional[CloudFile] = None,
+        field: CloudFile | None = None,
     ):
         self.storage = storage
         self.bucket = bucket
@@ -83,7 +78,7 @@ class StorageField(abc.ABC, metaclass=abc.ABCMeta):
     async def upload(self, iterator: AsyncIterator, origin: CloudFile) -> CloudFile: ...
 
     @abc.abstractmethod
-    async def iter_data(self, range: Optional[Range] = None) -> AsyncGenerator[bytes, None]:
+    async def iter_data(self, range: Range | None = None) -> AsyncGenerator[bytes]:
         raise NotImplementedError()
         yield b""
 
@@ -95,7 +90,7 @@ class StorageField(abc.ABC, metaclass=abc.ABCMeta):
         return deleted
 
     @abc.abstractmethod
-    async def exists(self) -> Optional[ObjectMetadata]: ...
+    async def exists(self) -> ObjectMetadata | None: ...
 
     @abc.abstractmethod
     async def copy(
@@ -130,10 +125,9 @@ class StorageField(abc.ABC, metaclass=abc.ABCMeta):
 
 class Storage(abc.ABC, metaclass=abc.ABCMeta):
     source: int
-    field_klass: Type
-    deadletter_bucket: Optional[str] = None
-    indexing_bucket: Optional[str] = None
-    cached_buckets: List[str] = []
+    field_klass: type
+    deadletter_bucket: str | None = None
+    indexing_bucket: str | None = None
     chunk_size = CHUNK_SIZE
 
     async def delete_resource(self, kbid: str, uuid: str):
@@ -166,7 +160,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         await self.upload_object(self.deadletter_bucket, key, message.SerializeToString())
 
     def get_indexing_storage_key(
-        self, *, kb: str, logical_shard: str, resource_uid: str, txid: Union[int, str]
+        self, *, kb: str, logical_shard: str, resource_uid: str, txid: int | str
     ):
         return INDEXING_KEY.format(kb=kb, shard=logical_shard, resource=resource_uid, txid=txid)
 
@@ -174,7 +168,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         self,
         message: BrainResource,
         txid: int,
-        partition: Optional[str],
+        partition: str | None,
         kb: str,
         logical_shard: str,
     ) -> str:
@@ -197,7 +191,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         self,
         message: BrainResource,
         reindex_id: str,
-        partition: Optional[str],
+        partition: str | None,
         kb: str,
         logical_shard: str,
     ) -> str:
@@ -328,7 +322,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         kbid: str,
         uuid: str,
         field: str,
-        old_field: Optional[CloudFile] = None,
+        old_field: CloudFile | None = None,
     ) -> StorageField:
         # Its a file field value
         bucket = self.get_bucket_name(kbid)
@@ -360,7 +354,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         payload: bytes,
         filename: str,
         content_type: str,
-        md5: Optional[str] = None,
+        md5: str | None = None,
     ):
         decoded_payload = base64.b64decode(payload)
         cf = CloudFile()
@@ -370,7 +364,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         cf.source = self.source  # type: ignore
 
         if md5 is None:
-            md5hash = hashlib.md5(decoded_payload).digest()
+            md5hash = hashlib.md5(decoded_payload, usedforsecurity=False).digest()
             cf.md5 = md5hash.decode()
         else:
             cf.md5 = md5
@@ -436,20 +430,19 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         self,
         bucket: str,
         key: str,
-        range: Optional[Range] = None,
-    ):
+        range: Range | None = None,
+    ) -> AsyncGenerator[bytes]:
         destination: StorageField = self.field_klass(storage=self, bucket=bucket, fullkey=key)
         try:
             async for data in destination.iter_data(range=range):
                 yield data
         except KeyError:
-            yield None
+            pass
 
     async def downloadbytes(self, bucket: str, key: str) -> BytesIO:
         result = BytesIO()
         async for data in self.download(bucket, key):
-            if data is not None:
-                result.write(data)
+            result.write(data)
 
         result.seek(0)
         return result
@@ -461,29 +454,24 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         result.seek(0)
         return result
 
-    async def downloadbytescf_iterator(
-        self, cf: CloudFile
-    ) -> AsyncGenerator[bytes, None]:  # pragma: no cover
+    async def downloadbytescf_iterator(self, cf: CloudFile) -> AsyncGenerator[bytes]:  # pragma: no cover
         # this is covered by other tests
         if cf.source == self.source:
             async for data in self.download(cf.bucket_name, cf.uri):
-                if data is not None:
-                    yield data
+                yield data
         elif cf.source == CloudFile.FLAPS:
             flaps_storage = await get_nuclia_storage()
             async for data in flaps_storage.download(cf):
-                if data is not None:
-                    yield data
+                yield data
         elif cf.source == CloudFile.LOCAL:
             local_storage = get_local_storage()
             async for data in local_storage.download(cf.bucket_name, cf.uri):
-                if data is not None:
-                    yield data
+                yield data
 
     async def upload_pb(self, sf: StorageField, payload: Any):
         await self.upload_object(sf.bucket, sf.key, payload.SerializeToString())
 
-    async def download_pb(self, sf: StorageField, PBKlass: Type):
+    async def download_pb(self, sf: StorageField, PBKlass: type):
         payload = await self.downloadbytes(sf.bucket, sf.key)
 
         if payload.getbuffer().nbytes == 0:
@@ -517,8 +505,8 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
 
     @abc.abstractmethod
     async def iterate_objects(
-        self, bucket: str, prefix: str, start: Optional[str] = None
-    ) -> AsyncGenerator[ObjectInfo, None]:
+        self, bucket: str, prefix: str, start: str | None = None
+    ) -> AsyncGenerator[ObjectInfo]:
         raise NotImplementedError()
         yield ObjectInfo(name="")
 
@@ -577,9 +565,7 @@ class Storage(abc.ABC, metaclass=abc.ABCMeta):
         ...
 
 
-async def iter_and_add_size(
-    stream: AsyncGenerator[bytes, None], cf: CloudFile
-) -> AsyncGenerator[bytes, None]:
+async def iter_and_add_size(stream: AsyncGenerator[bytes], cf: CloudFile) -> AsyncGenerator[bytes]:
     # This is needed because some storage types like GCS or S3 require
     # the size of the file at least at the request done for the last chunk.
     total_size = 0
@@ -590,9 +576,7 @@ async def iter_and_add_size(
         yield chunk
 
 
-async def iter_in_chunk_size(
-    iterator: AsyncGenerator[bytes, None], chunk_size: int
-) -> AsyncGenerator[bytes, None]:
+async def iter_in_chunk_size(iterator: AsyncGenerator[bytes], chunk_size: int) -> AsyncGenerator[bytes]:
     # This is needed to make sure bytes uploaded to the blob storage complies with a particular chunk size.
     buffer = b""
     async for chunk in iterator:
@@ -606,8 +590,8 @@ async def iter_in_chunk_size(
 
 
 async def iterate_storage_compatible(
-    iterator: AsyncGenerator[bytes, None], storage: Storage, cf: CloudFile
-) -> AsyncGenerator[bytes, None]:
+    iterator: AsyncGenerator[bytes], storage: Storage, cf: CloudFile
+) -> AsyncGenerator[bytes]:
     """
     Makes sure to add the size to the cloudfile and split the data in
     chunks that are compatible with the storage type of choice

nucliadb_utils/storages/utils.py

@@ -19,7 +19,6 @@
 #
 
 from dataclasses import dataclass
-from typing import Optional
 
 from pydantic import BaseModel
 
@@ -41,8 +40,8 @@ class Range:
     The start and end values are 0-based.
     """
 
-    start: Optional[int] = None
-    end: Optional[int] = None
+    start: int | None = None
+    end: int | None = None
 
     def any(self) -> bool:
         return self.start is not None or self.end is not None

nucliadb_utils/store.py

@@ -17,6 +17,6 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program. If not, see <http://www.gnu.org/licenses/>.
 
-from typing import Any, Dict
+from typing import Any
 
-MAIN: Dict[str, Any] = {}
+MAIN: dict[str, Any] = {}