unstructured-ingest 0.5.0__py3-none-any.whl → 0.5.2__py3-none-any.whl

unstructured_ingest/embed/vertexai.py

@@ -9,6 +9,7 @@ from pydantic import Field, Secret, ValidationError
 from pydantic.functional_validators import BeforeValidator
 
 from unstructured_ingest.embed.interfaces import (
+    EMBEDDINGS_KEY,
     AsyncBaseEmbeddingEncoder,
     BaseEmbeddingEncoder,
     EmbeddingConfig,
@@ -75,9 +76,12 @@ class VertexAIEmbeddingEncoder(BaseEmbeddingEncoder):
         return self._embed_documents(elements=[query])[0]
 
     def embed_documents(self, elements: list[dict]) -> list[dict]:
-        embeddings = self._embed_documents([e.get("text", "") for e in elements])
-        elements_with_embeddings = self._add_embeddings_to_elements(elements, embeddings)
-        return elements_with_embeddings
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        embeddings = self._embed_documents([e["text"] for e in elements_with_text])
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
 
     @requires_dependencies(
         ["vertexai"],
@@ -110,9 +114,12 @@ class AsyncVertexAIEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
         return embedding[0]
 
     async def embed_documents(self, elements: list[dict]) -> list[dict]:
-        embeddings = await self._embed_documents([e.get("text", "") for e in elements])
-        elements_with_embeddings = self._add_embeddings_to_elements(elements, embeddings)
-        return elements_with_embeddings
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        embeddings = await self._embed_documents([e["text"] for e in elements_with_text])
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
 
     @requires_dependencies(
         ["vertexai"],

unstructured_ingest/embed/voyageai.py

@@ -4,6 +4,7 @@ from typing import TYPE_CHECKING, Optional
 from pydantic import Field, SecretStr
 
 from unstructured_ingest.embed.interfaces import (
+    EMBEDDINGS_KEY,
     AsyncBaseEmbeddingEncoder,
     BaseEmbeddingEncoder,
     EmbeddingConfig,
@@ -107,8 +108,12 @@ class VoyageAIEmbeddingEncoder(BaseEmbeddingEncoder):
         return embeddings
 
     def embed_documents(self, elements: list[dict]) -> list[dict]:
-        embeddings = self._embed_documents([e.get("text", "") for e in elements])
-        return self._add_embeddings_to_elements(elements, embeddings)
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        embeddings = self._embed_documents([e["text"] for e in elements_with_text])
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
 
     def embed_query(self, query: str) -> list[float]:
         return self._embed_documents(elements=[query])[0]
@@ -135,8 +140,12 @@ class AsyncVoyageAIEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
         return embeddings
 
     async def embed_documents(self, elements: list[dict]) -> list[dict]:
-        embeddings = await self._embed_documents([e.get("text", "") for e in elements])
-        return self._add_embeddings_to_elements(elements, embeddings)
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        embeddings = await self._embed_documents([e["text"] for e in elements_with_text])
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
 
     async def embed_query(self, query: str) -> list[float]:
         embedding = await self._embed_documents(elements=[query])

unstructured_ingest/v2/processes/connectors/duckdb/base.py

@@ -81,6 +81,8 @@ class BaseDuckDBUploadStager(UploadStager):
         **kwargs: Any,
     ) -> Path:
         elements_contents = get_data(path=elements_filepath)
+        output_filename_suffix = Path(elements_filepath).suffix
+        output_filename = f"{Path(output_filename).stem}{output_filename_suffix}"
         output_path = self.get_output_path(output_filename=output_filename, output_dir=output_dir)
 
         output = [

unstructured_ingest/v2/processes/connectors/duckdb/motherduck.py

@@ -61,7 +61,7 @@ class MotherDuckConnectionConfig(ConnectionConfig):
                 "custom_user_agent": f"unstructured-io-ingest/{unstructured_io_ingest_version}"
             },
         ) as conn:
-            conn.sql(f"USE {self.database}")
+            conn.sql(f'USE "{self.database}"')
             yield conn
 
     @contextmanager
@@ -102,11 +102,12 @@ class MotherDuckUploader(Uploader):
 
     def upload_dataframe(self, df: pd.DataFrame) -> None:
         logger.debug(f"uploading {len(df)} entries to {self.connection_config.database} ")
+        database = self.connection_config.database
+        db_schema = self.connection_config.db_schema
+        table = self.connection_config.table
 
         with self.connection_config.get_client() as conn:
-            conn.query(
-                f"INSERT INTO {self.connection_config.db_schema}.{self.connection_config.table} BY NAME SELECT * FROM df"  # noqa: E501
-            )
+            conn.query(f'INSERT INTO "{database}"."{db_schema}"."{table}" BY NAME SELECT * FROM df')
 
     def run_data(self, data: list[dict], file_data: FileData, **kwargs: Any) -> None:
         df = pd.DataFrame(data=data)

unstructured_ingest/v2/processes/connectors/google_drive.py

@@ -132,12 +132,141 @@ class GoogleDriveIndexer(Indexer):
         ]
     )
 
+    @staticmethod
+    def verify_drive_api_enabled(client) -> None:
+        from googleapiclient.errors import HttpError
+
+        """
+        Makes a lightweight API call to verify that the Drive API is enabled.
+        If the API is not enabled, an HttpError should be raised.
+        """
+        try:
+            # A very minimal call: list 1 file from the drive.
+            client.list(spaces="drive", pageSize=1, fields="files(id)").execute()
+        except HttpError as e:
+            error_content = e.content.decode() if hasattr(e, "content") else ""
+            lower_error = error_content.lower()
+            if "drive api" in lower_error and (
+                "not enabled" in lower_error or "not been used" in lower_error
+            ):
+                raise SourceConnectionError(
+                    "Google Drive API is not enabled for your project. \
+                    Please enable it in the Google Cloud Console."
+                )
+            else:
+                raise SourceConnectionError("Google drive API unreachable for an unknown reason!")
+
+    @staticmethod
+    def count_files_recursively(files_client, folder_id: str, extensions: list[str] = None) -> int:
+        """
+        Count non-folder files recursively under the given folder.
+        If `extensions` is provided, only count files
+        whose `fileExtension` matches one of the values.
+        """
+        count = 0
+        stack = [folder_id]
+        while stack:
+            current_folder = stack.pop()
+            # Always list all items under the current folder.
+            query = f"'{current_folder}' in parents"
+            page_token = None
+            while True:
+                response = files_client.list(
+                    spaces="drive",
+                    q=query,
+                    fields="nextPageToken, files(id, mimeType, fileExtension)",
+                    pageToken=page_token,
+                    pageSize=1000,
+                ).execute()
+                for item in response.get("files", []):
+                    if item.get("mimeType") == "application/vnd.google-apps.folder":
+                        # Always traverse sub-folders regardless of extension filter.
+                        stack.append(item["id"])
+                    else:
+                        if extensions:
+                            # Use a case-insensitive comparison for the file extension.
+                            file_ext = (item.get("fileExtension") or "").lower()
+                            valid_exts = [e.lower() for e in extensions]
+                            if file_ext in valid_exts:
+                                count += 1
+                        else:
+                            count += 1
+                page_token = response.get("nextPageToken")
+                if not page_token:
+                    break
+        return count
+
     def precheck(self) -> None:
+        """
+        Enhanced precheck that verifies not only connectivity
+        but also that the provided drive_id is valid and accessible.
+        """
         try:
-            self.connection_config.get_client()
+            with self.connection_config.get_client() as client:
+                # First, verify that the Drive API is enabled.
+                self.verify_drive_api_enabled(client)
+
+                # Try to retrieve metadata for the drive id.
+                # This will catch errors such as an invalid drive id or insufficient permissions.
+                root_info = self.get_root_info(
+                    files_client=client, object_id=self.connection_config.drive_id
+                )
+                logger.info(
+                    f"Successfully retrieved drive root info: "
+                    f"{root_info.get('name', 'Unnamed')} (ID: {root_info.get('id')})"
+                )
+
+            # If the target is a folder, perform file count check.
+            if self.is_dir(root_info):
+                if self.index_config.recursive:
+                    file_count = self.count_files_recursively(
+                        client,
+                        self.connection_config.drive_id,
+                        extensions=self.index_config.extensions,
+                    )
+                    if file_count == 0:
+                        logger.warning(
+                            "Empty folder: no files found recursively in the folder. \
+                             Please verify that the folder contains files and \
+                             that the service account has proper permissions."
+                        )
+                        # raise SourceConnectionError(
+                        #     "Empty folder: no files found recursively in the folder. "
+                        #     "Please verify that the folder contains files and \
+                        #     that the service account has proper permissions."
+                        # )
+                    else:
+                        logger.info(f"Found {file_count} files recursively in the folder.")
+                else:
+                    # Non-recursive: check for at least one immediate non-folder child.
+                    response = client.list(
+                        spaces="drive",
+                        fields="files(id)",
+                        pageSize=1,
+                        q=f"'{self.connection_config.drive_id}' in parents",
+                    ).execute()
+                    if not response.get("files"):
+                        logger.warning(
+                            "Empty folder: no files found at the folder's root level. "
+                            "Please verify that the folder contains files and \
+                            that the service account has proper permissions."
+                        )
+                        # raise SourceConnectionError(
+                        #     "Empty folder: no files found at the folder's root level. "
+                        #     "Please verify that the folder contains files and \
+                        #     that the service account has proper permissions."
+                        # )
+                    else:
+                        logger.info("Found files at the folder's root level.")
+            else:
+                # If the target is a file, precheck passes.
+                logger.info("Drive ID corresponds to a file. Precheck passed.")
+
         except Exception as e:
-            logger.error(f"failed to validate connection: {e}", exc_info=True)
-            raise SourceConnectionError(f"failed to validate connection: {e}")
+            logger.error(
+                "Failed to validate Google Drive connection during precheck", exc_info=True
+            )
+            raise SourceConnectionError(f"Precheck failed: {e}")
 
     @staticmethod
     def is_dir(record: dict) -> bool:
@@ -310,20 +439,22 @@ class GoogleDriveDownloader(Downloader):
         from googleapiclient.http import MediaIoBaseDownload
 
         logger.debug(f"fetching file: {file_data.source_identifiers.fullpath}")
-        mime_type = file_data.additional_metadata["mimeType"]
         record_id = file_data.identifier
+        mime_type = file_data.additional_metadata["mimeType"]
+        if not mime_type:
+            raise TypeError(
+                f"File not supported. Name: {file_data.source_identifiers.filename} "
+                f"ID: {record_id} "
+                f"MimeType: {mime_type}"
+            )
         with self.connection_config.get_client() as client:
-            if mime_type.startswith("application/vnd.google-apps"):
+            if (
+                mime_type.startswith("application/vnd.google-apps")
+                and mime_type in GOOGLE_DRIVE_EXPORT_TYPES
+            ):
                 export_mime = GOOGLE_DRIVE_EXPORT_TYPES.get(
-                    self.meta.get("mimeType"),  # type: ignore
+                    mime_type,  # type: ignore
                 )
-                if not export_mime:
-                    raise TypeError(
-                        f"File not supported. Name: {file_data.source_identifiers.filename} "
-                        f"ID: {record_id} "
-                        f"MimeType: {mime_type}"
-                    )
-
                 request = client.export_media(
                     fileId=record_id,
                     mimeType=export_mime,

unstructured_ingest/v2/processes/connectors/pinecone.py

@@ -81,6 +81,7 @@ ALLOWED_FIELDS = (
     "link_urls",
     "link_texts",
     "text_as_html",
+    "entities",
 )
 
 

unstructured_ingest/v2/processes/connectors/sql/snowflake.py

@@ -1,6 +1,7 @@
+import json
 from contextlib import contextmanager
 from dataclasses import dataclass, field
-from typing import TYPE_CHECKING, Generator, Optional
+from typing import TYPE_CHECKING, Any, Generator, Optional
 
 import numpy as np
 import pandas as pd
@@ -15,6 +16,7 @@ from unstructured_ingest.v2.processes.connector_registry import (
     SourceRegistryEntry,
 )
 from unstructured_ingest.v2.processes.connectors.sql.sql import (
+    _DATE_COLUMNS,
     SQLAccessConfig,
     SqlBatchFileData,
     SQLConnectionConfig,
@@ -26,6 +28,7 @@ from unstructured_ingest.v2.processes.connectors.sql.sql import (
     SQLUploaderConfig,
     SQLUploadStager,
     SQLUploadStagerConfig,
+    parse_date_string,
 )
 
 if TYPE_CHECKING:
@@ -34,6 +37,17 @@ if TYPE_CHECKING:
 
 CONNECTOR_TYPE = "snowflake"
 
+_ARRAY_COLUMNS = (
+    "embeddings",
+    "languages",
+    "link_urls",
+    "link_texts",
+    "sent_from",
+    "sent_to",
+    "emphasized_text_contents",
+    "emphasized_text_tags",
+)
+
 
 class SnowflakeAccessConfig(SQLAccessConfig):
     password: Optional[str] = Field(default=None, description="DB password")
@@ -160,6 +174,42 @@ class SnowflakeUploader(SQLUploader):
     connector_type: str = CONNECTOR_TYPE
     values_delimiter: str = "?"
 
+    def prepare_data(
+        self, columns: list[str], data: tuple[tuple[Any, ...], ...]
+    ) -> list[tuple[Any, ...]]:
+        output = []
+        for row in data:
+            parsed = []
+            for column_name, value in zip(columns, row):
+                if column_name in _DATE_COLUMNS:
+                    if value is None or pd.isna(value):  # pandas is nan
+                        parsed.append(None)
+                    else:
+                        parsed.append(parse_date_string(value))
+                elif column_name in _ARRAY_COLUMNS:
+                    if not isinstance(value, list) and (
+                        value is None or pd.isna(value)
+                    ):  # pandas is nan
+                        parsed.append(None)
+                    else:
+                        parsed.append(json.dumps(value))
+                else:
+                    parsed.append(value)
+            output.append(tuple(parsed))
+        return output
+
+    def _parse_values(self, columns: list[str]) -> str:
+        return ",".join(
+            [
+                (
+                    f"PARSE_JSON({self.values_delimiter})"
+                    if col in _ARRAY_COLUMNS
+                    else self.values_delimiter
+                )
+                for col in columns
+            ]
+        )
+
     def upload_dataframe(self, df: pd.DataFrame, file_data: FileData) -> None:
         if self.can_delete():
             self.delete_by_record_id(file_data=file_data)
@@ -173,10 +223,10 @@ class SnowflakeUploader(SQLUploader):
         self._fit_to_schema(df=df)
 
         columns = list(df.columns)
-        stmt = "INSERT INTO {table_name} ({columns}) VALUES({values})".format(
+        stmt = "INSERT INTO {table_name} ({columns}) SELECT {values}".format(
             table_name=self.upload_config.table_name,
             columns=",".join(columns),
-            values=",".join([self.values_delimiter for _ in columns]),
+            values=self._parse_values(columns),
         )
         logger.info(
             f"writing a total of {len(df)} elements via"

unstructured_ingest/v2/processes/connectors/sql/sql.py

@@ -38,48 +38,6 @@ from unstructured_ingest.v2.interfaces import (
 from unstructured_ingest.v2.logger import logger
 from unstructured_ingest.v2.utils import get_enhanced_element_id
 
-_COLUMNS = (
-    "id",
-    "element_id",
-    "text",
-    "embeddings",
-    "type",
-    "system",
-    "layout_width",
-    "layout_height",
-    "points",
-    "url",
-    "version",
-    "date_created",
-    "date_modified",
-    "date_processed",
-    "permissions_data",
-    "record_locator",
-    "category_depth",
-    "parent_id",
-    "attached_filename",
-    "filetype",
-    "last_modified",
-    "file_directory",
-    "filename",
-    "languages",
-    "page_number",
-    "links",
-    "page_name",
-    "link_urls",
-    "link_texts",
-    "sent_from",
-    "sent_to",
-    "subject",
-    "section",
-    "header_footer_type",
-    "emphasized_text_contents",
-    "emphasized_text_tags",
-    "text_as_html",
-    "regex_metadata",
-    "detection_class_prob",
-)
-
 _DATE_COLUMNS = ("date_created", "date_modified", "date_processed", "last_modified")
 
 
@@ -270,10 +228,8 @@ class SQLUploadStager(UploadStager):
 
         data["id"] = get_enhanced_element_id(element_dict=data, file_data=file_data)
 
-        # remove extraneous, not supported columns
-        element = {k: v for k, v in data.items() if k in _COLUMNS}
-        element[RECORD_ID_LABEL] = file_data.identifier
-        return element
+        data[RECORD_ID_LABEL] = file_data.identifier
+        return data
 
     def conform_dataframe(self, df: pd.DataFrame) -> pd.DataFrame:
         for column in filter(lambda x: x in df.columns, _DATE_COLUMNS):
@@ -375,7 +331,7 @@ class SQLUploader(Uploader):
         missing_columns = schema_fields - columns
 
         if columns_to_drop:
-            logger.warning(
+            logger.info(
                 "Following columns will be dropped to match the table's schema: "
                 f"{', '.join(columns_to_drop)}"
             )

unstructured_ingest/v2/processes/connectors/sql/vastdb.py

@@ -19,7 +19,6 @@ from unstructured_ingest.v2.processes.connector_registry import (
     SourceRegistryEntry,
 )
 from unstructured_ingest.v2.processes.connectors.sql.sql import (
-    _COLUMNS,
     SQLAccessConfig,
     SqlBatchFileData,
     SQLConnectionConfig,
@@ -149,13 +148,11 @@ class VastdbUploadStagerConfig(SQLUploadStagerConfig):
         default=None,
         description="Map of column names to rename, ex: {'old_name': 'new_name'}",
     )
-    additional_columns: Optional[list[str]] = Field(
-        default_factory=list, description="Additional columns to include in the upload"
-    )
 
 
+@dataclass
 class VastdbUploadStager(SQLUploadStager):
-    upload_stager_config: VastdbUploadStagerConfig
+    upload_stager_config: VastdbUploadStagerConfig = field(default_factory=VastdbUploadStagerConfig)
 
     def conform_dict(self, element_dict: dict, file_data: FileData) -> dict:
         data = element_dict.copy()
@@ -168,13 +165,8 @@ class VastdbUploadStager(SQLUploadStager):
         data.update(coordinates)
 
         data["id"] = get_enhanced_element_id(element_dict=data, file_data=file_data)
-
-        # remove extraneous, not supported columns
-        # but also allow for additional columns
-        approved_columns = set(_COLUMNS).union(self.upload_stager_config.additional_columns)
-        element = {k: v for k, v in data.items() if k in approved_columns}
-        element[RECORD_ID_LABEL] = file_data.identifier
-        return element
+        data[RECORD_ID_LABEL] = file_data.identifier
+        return data
 
     def conform_dataframe(self, df: pd.DataFrame) -> pd.DataFrame:
         df = super().conform_dataframe(df=df)

{unstructured_ingest-0.5.0.dist-info → unstructured_ingest-0.5.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: unstructured-ingest
-Version: 0.5.0
+Version: 0.5.2
 Summary: A library that prepares raw documents for downstream ML tasks.
 Home-page: https://github.com/Unstructured-IO/unstructured-ingest
 Author: Unstructured Technologies
@@ -23,37 +23,37 @@ Requires-Python: >=3.9.0,<3.14
 Description-Content-Type: text/markdown
 License-File: LICENSE.md
 Requires-Dist: pandas
-Requires-Dist: dataclasses-json
 Requires-Dist: pydantic>=2.7
-Requires-Dist: click
-Requires-Dist: tqdm
+Requires-Dist: dataclasses-json
 Requires-Dist: python-dateutil
 Requires-Dist: opentelemetry-sdk
+Requires-Dist: click
+Requires-Dist: tqdm
 Provides-Extra: airtable
 Requires-Dist: pyairtable; extra == "airtable"
 Provides-Extra: astradb
 Requires-Dist: astrapy; extra == "astradb"
 Provides-Extra: azure
-Requires-Dist: adlfs; extra == "azure"
 Requires-Dist: fsspec; extra == "azure"
+Requires-Dist: adlfs; extra == "azure"
 Provides-Extra: azure-ai-search
 Requires-Dist: azure-search-documents; extra == "azure-ai-search"
 Provides-Extra: bedrock
-Requires-Dist: aioboto3; extra == "bedrock"
 Requires-Dist: boto3; extra == "bedrock"
+Requires-Dist: aioboto3; extra == "bedrock"
 Provides-Extra: biomed
 Requires-Dist: requests; extra == "biomed"
 Requires-Dist: bs4; extra == "biomed"
 Provides-Extra: box
-Requires-Dist: boxfs; extra == "box"
 Requires-Dist: fsspec; extra == "box"
+Requires-Dist: boxfs; extra == "box"
 Provides-Extra: chroma
 Requires-Dist: chromadb; extra == "chroma"
 Provides-Extra: clarifai
 Requires-Dist: clarifai; extra == "clarifai"
 Provides-Extra: confluence
-Requires-Dist: requests; extra == "confluence"
 Requires-Dist: atlassian-python-api; extra == "confluence"
+Requires-Dist: requests; extra == "confluence"
 Provides-Extra: couchbase
 Requires-Dist: couchbase; extra == "couchbase"
 Provides-Extra: csv
@@ -63,8 +63,8 @@ Requires-Dist: databricks-sql-connector; extra == "databricks-delta-tables"
 Provides-Extra: databricks-volumes
 Requires-Dist: databricks-sdk; extra == "databricks-volumes"
 Provides-Extra: delta-table
-Requires-Dist: deltalake; extra == "delta-table"
 Requires-Dist: boto3; extra == "delta-table"
+Requires-Dist: deltalake; extra == "delta-table"
 Provides-Extra: discord
 Requires-Dist: discord.py; extra == "discord"
 Provides-Extra: doc
@@ -83,8 +83,8 @@ Requires-Dist: sentence-transformers; extra == "embed-huggingface"
 Provides-Extra: embed-mixedbreadai
 Requires-Dist: mixedbread-ai; extra == "embed-mixedbreadai"
 Provides-Extra: embed-octoai
-Requires-Dist: tiktoken; extra == "embed-octoai"
 Requires-Dist: openai; extra == "embed-octoai"
+Requires-Dist: tiktoken; extra == "embed-octoai"
 Provides-Extra: embed-vertexai
 Requires-Dist: vertexai; extra == "embed-vertexai"
 Provides-Extra: embed-voyageai
@@ -92,9 +92,9 @@ Requires-Dist: voyageai; extra == "embed-voyageai"
 Provides-Extra: epub
 Requires-Dist: unstructured[epub]; extra == "epub"
 Provides-Extra: gcs
-Requires-Dist: gcsfs; extra == "gcs"
-Requires-Dist: bs4; extra == "gcs"
 Requires-Dist: fsspec; extra == "gcs"
+Requires-Dist: bs4; extra == "gcs"
+Requires-Dist: gcsfs; extra == "gcs"
 Provides-Extra: github
 Requires-Dist: pygithub>1.58.0; extra == "github"
 Requires-Dist: requests; extra == "github"
@@ -126,10 +126,10 @@ Requires-Dist: networkx; extra == "neo4j"
 Requires-Dist: cymple; extra == "neo4j"
 Requires-Dist: neo4j; extra == "neo4j"
 Provides-Extra: notion
-Requires-Dist: httpx; extra == "notion"
+Requires-Dist: htmlBuilder; extra == "notion"
 Requires-Dist: backoff; extra == "notion"
 Requires-Dist: notion-client; extra == "notion"
-Requires-Dist: htmlBuilder; extra == "notion"
+Requires-Dist: httpx; extra == "notion"
 Provides-Extra: odt
 Requires-Dist: unstructured[odt]; extra == "odt"
 Provides-Extra: onedrive
@@ -137,8 +137,8 @@ Requires-Dist: Office365-REST-Python-Client; extra == "onedrive"
 Requires-Dist: bs4; extra == "onedrive"
 Requires-Dist: msal; extra == "onedrive"
 Provides-Extra: openai
-Requires-Dist: tiktoken; extra == "openai"
 Requires-Dist: openai; extra == "openai"
+Requires-Dist: tiktoken; extra == "openai"
 Provides-Extra: opensearch
 Requires-Dist: opensearch-py; extra == "opensearch"
 Provides-Extra: org
@@ -174,8 +174,8 @@ Requires-Dist: fsspec; extra == "s3"
 Provides-Extra: salesforce
 Requires-Dist: simple-salesforce; extra == "salesforce"
 Provides-Extra: sftp
-Requires-Dist: paramiko; extra == "sftp"
 Requires-Dist: fsspec; extra == "sftp"
+Requires-Dist: paramiko; extra == "sftp"
 Provides-Extra: sharepoint
 Requires-Dist: Office365-REST-Python-Client; extra == "sharepoint"
 Requires-Dist: msal; extra == "sharepoint"
@@ -192,11 +192,11 @@ Provides-Extra: tsv
 Requires-Dist: unstructured[tsv]; extra == "tsv"
 Provides-Extra: vastdb
 Requires-Dist: vastdb; extra == "vastdb"
-Requires-Dist: ibis; extra == "vastdb"
 Requires-Dist: pyarrow; extra == "vastdb"
+Requires-Dist: ibis; extra == "vastdb"
 Provides-Extra: vectara
-Requires-Dist: httpx; extra == "vectara"
 Requires-Dist: requests; extra == "vectara"
+Requires-Dist: httpx; extra == "vectara"
 Requires-Dist: aiofiles; extra == "vectara"
 Provides-Extra: weaviate
 Requires-Dist: weaviate-client; extra == "weaviate"