unstructured-ingest 0.5.2__py3-none-any.whl → 0.5.4__py3-none-any.whl

test/integration/connectors/test_sharepoint.py

@@ -19,24 +19,31 @@ from unstructured_ingest.v2.processes.connectors.sharepoint import (
 )
 
 
+def sharepoint_config():
+    class SharepointTestConfig:
+        def __init__(self):
+            self.client_id = os.environ["SHAREPOINT_CLIENT_ID"]
+            self.client_cred = os.environ["SHAREPOINT_CRED"]
+            self.user_pname = os.environ["MS_USER_PNAME"]
+            self.tenant = os.environ["MS_TENANT_ID"]
+
+    return SharepointTestConfig()
+
+
 @pytest.mark.asyncio
 @pytest.mark.tags(CONNECTOR_TYPE, SOURCE_TAG, BLOB_STORAGE_TAG)
 @requires_env("SHAREPOINT_CLIENT_ID", "SHAREPOINT_CRED", "MS_TENANT_ID", "MS_USER_PNAME")
 async def test_sharepoint_source(temp_dir):
-    # Retrieve environment variables
     site = "https://unstructuredio.sharepoint.com/sites/utic-platform-test-source"
-    client_id = os.environ["SHAREPOINT_CLIENT_ID"]
-    client_cred = os.environ["SHAREPOINT_CRED"]
-    user_pname = os.environ["MS_USER_PNAME"]
-    tenant = os.environ["MS_TENANT_ID"]
+    config = sharepoint_config()
 
     # Create connection and indexer configurations
-    access_config = SharepointAccessConfig(client_cred=client_cred)
+    access_config = SharepointAccessConfig(client_cred=config.client_cred)
     connection_config = SharepointConnectionConfig(
-        client_id=client_id,
+        client_id=config.client_id,
         site=site,
-        tenant=tenant,
-        user_pname=user_pname,
+        tenant=config.tenant,
+        user_pname=config.user_pname,
         access_config=access_config,
     )
     index_config = SharepointIndexerConfig(recursive=True)
@@ -58,7 +65,151 @@ async def test_sharepoint_source(temp_dir):
         indexer=indexer,
         downloader=downloader,
         configs=SourceValidationConfigs(
-            test_id="sharepoint",
+            test_id="sharepoint1",
+            expected_num_files=4,
+            validate_downloaded_files=True,
+            exclude_fields_extend=[
+                "metadata.date_created",
+                "metadata.date_modified",
+                "additional_metadata.LastModified",
+                "additional_metadata.@microsoft.graph.downloadUrl",
+            ],
+        ),
+    )
+
+
+@pytest.mark.asyncio
+@pytest.mark.tags(CONNECTOR_TYPE, SOURCE_TAG, BLOB_STORAGE_TAG)
+@requires_env("SHAREPOINT_CLIENT_ID", "SHAREPOINT_CRED", "MS_TENANT_ID", "MS_USER_PNAME")
+async def test_sharepoint_source_with_path(temp_dir):
+    site = "https://unstructuredio.sharepoint.com/sites/utic-platform-test-source"
+    config = sharepoint_config()
+
+    # Create connection and indexer configurations
+    access_config = SharepointAccessConfig(client_cred=config.client_cred)
+    connection_config = SharepointConnectionConfig(
+        client_id=config.client_id,
+        site=site,
+        tenant=config.tenant,
+        user_pname=config.user_pname,
+        access_config=access_config,
+    )
+    index_config = SharepointIndexerConfig(recursive=True, path="Folder1")
+
+    download_config = SharepointDownloaderConfig(download_dir=temp_dir)
+
+    # Instantiate indexer and downloader
+    indexer = SharepointIndexer(
+        connection_config=connection_config,
+        index_config=index_config,
+    )
+    downloader = SharepointDownloader(
+        connection_config=connection_config,
+        download_config=download_config,
+    )
+
+    # Run the source connector validation
+    await source_connector_validation(
+        indexer=indexer,
+        downloader=downloader,
+        configs=SourceValidationConfigs(
+            test_id="sharepoint2",
+            expected_num_files=2,
+            validate_downloaded_files=True,
+            exclude_fields_extend=[
+                "metadata.date_created",
+                "metadata.date_modified",
+                "additional_metadata.LastModified",
+                "additional_metadata.@microsoft.graph.downloadUrl",
+            ],
+        ),
+    )
+
+
+@pytest.mark.asyncio
+@pytest.mark.tags(CONNECTOR_TYPE, SOURCE_TAG, BLOB_STORAGE_TAG)
+@requires_env("SHAREPOINT_CLIENT_ID", "SHAREPOINT_CRED", "MS_TENANT_ID", "MS_USER_PNAME")
+async def test_sharepoint_root_with_path(temp_dir):
+    site = "https://unstructuredio.sharepoint.com/"
+    config = sharepoint_config()
+
+    # Create connection and indexer configurations
+    access_config = SharepointAccessConfig(client_cred=config.client_cred)
+    connection_config = SharepointConnectionConfig(
+        client_id=config.client_id,
+        site=site,
+        tenant=config.tenant,
+        user_pname=config.user_pname,
+        access_config=access_config,
+    )
+    index_config = SharepointIndexerConfig(recursive=True, path="e2e-test-folder")
+
+    download_config = SharepointDownloaderConfig(download_dir=temp_dir)
+
+    # Instantiate indexer and downloader
+    indexer = SharepointIndexer(
+        connection_config=connection_config,
+        index_config=index_config,
+    )
+    downloader = SharepointDownloader(
+        connection_config=connection_config,
+        download_config=download_config,
+    )
+
+    # Run the source connector validation
+    await source_connector_validation(
+        indexer=indexer,
+        downloader=downloader,
+        configs=SourceValidationConfigs(
+            test_id="sharepoint3",
+            expected_num_files=1,
+            validate_downloaded_files=True,
+            exclude_fields_extend=[
+                "metadata.date_created",
+                "metadata.date_modified",
+                "additional_metadata.LastModified",
+                "additional_metadata.@microsoft.graph.downloadUrl",
+            ],
+        ),
+    )
+
+
+@pytest.mark.asyncio
+@pytest.mark.tags(CONNECTOR_TYPE, SOURCE_TAG, BLOB_STORAGE_TAG)
+@requires_env("SHAREPOINT_CLIENT_ID", "SHAREPOINT_CRED", "MS_TENANT_ID", "MS_USER_PNAME")
+async def test_sharepoint_shared_documents(temp_dir):
+    site = "https://unstructuredio.sharepoint.com/sites/utic-platform-test-source"
+    config = sharepoint_config()
+
+    # Create connection and indexer configurations
+    access_config = SharepointAccessConfig(client_cred=config.client_cred)
+    connection_config = SharepointConnectionConfig(
+        client_id=config.client_id,
+        site=site,
+        tenant=config.tenant,
+        user_pname=config.user_pname,
+        access_config=access_config,
+    )
+    index_config = SharepointIndexerConfig(recursive=True, path="Shared Documents")
+
+    download_config = SharepointDownloaderConfig(download_dir=temp_dir)
+
+    # Instantiate indexer and downloader
+    indexer = SharepointIndexer(
+        connection_config=connection_config,
+        index_config=index_config,
+    )
+    downloader = SharepointDownloader(
+        connection_config=connection_config,
+        download_config=download_config,
+    )
+
+    # Run the source connector validation
+    await source_connector_validation(
+        indexer=indexer,
+        downloader=downloader,
+        configs=SourceValidationConfigs(
+            test_id="sharepoint4",
             expected_num_files=4,
             validate_downloaded_files=True,
             exclude_fields_extend=[

test/unit/v2/embedders/test_bedrock.py

@@ -15,7 +15,7 @@ def generate_embedder_config_params() -> dict:
         "region_name": fake.city(),
     }
     if random.random() < 0.5:
-        params["embed_model_name"] = fake.word()
+        params["embedder_model_name"] = fake.word()
     return params
 
 

test/unit/v2/embedders/test_huggingface.py

@@ -16,7 +16,7 @@ fake = faker.Faker()
 def generate_embedder_config_params() -> dict:
     params = {}
     if random.random() < 0.5:
-        params["embed_model_name"] = fake.word() if random.random() < 0.5 else None
+        params["embedder_model_name"] = fake.word() if random.random() < 0.5 else None
         params["embedder_model_kwargs"] = (
             generate_random_dictionary(key_type=str, value_type=Any)
             if random.random() < 0.5

unstructured_ingest/__version__.py

@@ -1 +1 @@
-__version__ = "0.5.2"  # pragma: no cover
+__version__ = "0.5.4"  # pragma: no cover

unstructured_ingest/embed/azure_openai.py

@@ -44,7 +44,13 @@ class AzureOpenAIEmbeddingConfig(OpenAIEmbeddingConfig):
 class AzureOpenAIEmbeddingEncoder(OpenAIEmbeddingEncoder):
     config: AzureOpenAIEmbeddingConfig
 
+    def get_client(self) -> "AzureOpenAI":
+        return self.config.get_client()
+
 
 @dataclass
 class AsyncAzureOpenAIEmbeddingEncoder(AsyncOpenAIEmbeddingEncoder):
     config: AzureOpenAIEmbeddingConfig
+
+    def get_client(self) -> "AsyncAzureOpenAI":
+        return self.config.get_async_client()

unstructured_ingest/embed/bedrock.py

@@ -15,7 +15,13 @@ from unstructured_ingest.embed.interfaces import (
 )
 from unstructured_ingest.logger import logger
 from unstructured_ingest.utils.dep_check import requires_dependencies
-from unstructured_ingest.v2.errors import ProviderError, RateLimitError, UserAuthError, UserError
+from unstructured_ingest.v2.errors import (
+    ProviderError,
+    RateLimitError,
+    UserAuthError,
+    UserError,
+    is_internal_error,
+)
 
 if TYPE_CHECKING:
     from botocore.client import BaseClient
@@ -51,9 +57,11 @@ class BedrockEmbeddingConfig(EmbeddingConfig):
     aws_access_key_id: SecretStr
     aws_secret_access_key: SecretStr
     region_name: str = "us-west-2"
-    embed_model_name: str = Field(default="amazon.titan-embed-text-v1", alias="model_name")
+    embedder_model_name: str = Field(default="amazon.titan-embed-text-v1", alias="model_name")
 
     def wrap_error(self, e: Exception) -> Exception:
+        if is_internal_error(e=e):
+            return e
         from botocore.exceptions import ClientError
 
         if isinstance(e, ClientError):
@@ -122,7 +130,7 @@ class BedrockEmbeddingEncoder(BaseEmbeddingEncoder):
 
     def embed_query(self, query: str) -> list[float]:
         """Call out to Bedrock embedding endpoint."""
-        provider = self.config.embed_model_name.split(".")[0]
+        provider = self.config.embedder_model_name.split(".")[0]
         body = conform_query(query=query, provider=provider)
 
         bedrock_client = self.config.get_client()
@@ -130,7 +138,7 @@ class BedrockEmbeddingEncoder(BaseEmbeddingEncoder):
         try:
             response = bedrock_client.invoke_model(
                 body=json.dumps(body),
-                modelId=self.config.embed_model_name,
+                modelId=self.config.embedder_model_name,
                 accept="application/json",
                 contentType="application/json",
             )
@@ -148,6 +156,8 @@ class BedrockEmbeddingEncoder(BaseEmbeddingEncoder):
     def embed_documents(self, elements: list[dict]) -> list[dict]:
         elements = elements.copy()
         elements_with_text = [e for e in elements if e.get("text")]
+        if not elements_with_text:
+            return elements
         embeddings = [self.embed_query(query=e["text"]) for e in elements_with_text]
         for element, embedding in zip(elements_with_text, embeddings):
             element[EMBEDDINGS_KEY] = embedding
@@ -163,7 +173,7 @@ class AsyncBedrockEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
 
     async def embed_query(self, query: str) -> list[float]:
         """Call out to Bedrock embedding endpoint."""
-        provider = self.config.embed_model_name.split(".")[0]
+        provider = self.config.embedder_model_name.split(".")[0]
         body = conform_query(query=query, provider=provider)
         try:
             async with self.config.get_async_client() as bedrock_client:
@@ -171,7 +181,7 @@ class AsyncBedrockEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
                 try:
                     response = await bedrock_client.invoke_model(
                         body=json.dumps(body),
-                        modelId=self.config.embed_model_name,
+                        modelId=self.config.embedder_model_name,
                         accept="application/json",
                         contentType="application/json",
                     )

unstructured_ingest/embed/huggingface.py

@@ -47,7 +47,7 @@ class HuggingFaceEmbeddingConfig(EmbeddingConfig):
 class HuggingFaceEmbeddingEncoder(BaseEmbeddingEncoder):
     config: HuggingFaceEmbeddingConfig
 
-    def embed_query(self, query: str) -> list[float]:
+    def _embed_query(self, query: str) -> list[float]:
         return self._embed_documents(texts=[query])[0]
 
     def _embed_documents(self, texts: list[str]) -> list[list[float]]:
@@ -58,6 +58,8 @@ class HuggingFaceEmbeddingEncoder(BaseEmbeddingEncoder):
     def embed_documents(self, elements: list[dict]) -> list[dict]:
         elements = elements.copy()
         elements_with_text = [e for e in elements if e.get("text")]
+        if not elements_with_text:
+            return elements
         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

unstructured_ingest/embed/interfaces.py

@@ -1,11 +1,12 @@
-import asyncio
-from abc import ABC, abstractmethod
+from abc import ABC
 from dataclasses import dataclass
-from typing import Optional
+from typing import Any, Optional
 
 import numpy as np
 from pydantic import BaseModel, Field
 
+from unstructured_ingest.utils.data_prep import batch_generator
+
 EMBEDDINGS_KEY = "embeddings"
 
 
@@ -50,21 +51,37 @@ class BaseEmbeddingEncoder(BaseEncoder, ABC):
         exemplary_embedding = self.get_exemplary_embedding()
         return np.isclose(np.linalg.norm(exemplary_embedding), 1.0)
 
-    @abstractmethod
-    def embed_documents(self, elements: list[dict]) -> list[dict]:
-        pass
+    def get_client(self):
+        raise NotImplementedError
 
-    @abstractmethod
-    def embed_query(self, query: str) -> list[float]:
-        pass
+    def embed_batch(self, client: Any, batch: list[str]) -> list[list[float]]:
+        raise NotImplementedError
 
-    def _embed_documents(self, elements: list[str]) -> list[list[float]]:
-        results = []
-        for text in elements:
-            response = self.embed_query(query=text)
-            results.append(response)
+    def embed_documents(self, elements: list[dict]) -> list[dict]:
+        client = self.get_client()
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        texts = [e["text"] for e in elements_with_text]
+        embeddings = []
+        try:
+            for batch in batch_generator(texts, batch_size=self.config.batch_size or len(texts)):
+                embeddings = self.embed_batch(client=client, batch=batch)
+                embeddings.extend(embeddings)
+        except Exception as e:
+            raise self.wrap_error(e=e)
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
+
+    def _embed_query(self, query: str) -> list[float]:
+        client = self.get_client()
+        return self.embed_batch(client=client, batch=[query])[0]
 
-        return results
+    def embed_query(self, query: str) -> list[float]:
+        try:
+            return self._embed_query(query=query)
+        except Exception as e:
+            raise self.wrap_error(e=e)
 
 
 @dataclass
@@ -88,14 +105,35 @@ class AsyncBaseEmbeddingEncoder(BaseEncoder, ABC):
         exemplary_embedding = await self.get_exemplary_embedding()
         return np.isclose(np.linalg.norm(exemplary_embedding), 1.0)
 
-    @abstractmethod
+    def get_client(self):
+        raise NotImplementedError
+
+    async def embed_batch(self, client: Any, batch: list[str]) -> list[list[float]]:
+        raise NotImplementedError
+
     async def embed_documents(self, elements: list[dict]) -> list[dict]:
-        pass
+        client = self.get_client()
+        elements = elements.copy()
+        elements_with_text = [e for e in elements if e.get("text")]
+        texts = [e["text"] for e in elements_with_text]
+        embeddings = []
+        try:
+            for batch in batch_generator(texts, batch_size=self.config.batch_size or len(texts)):
+                embeddings = await self.embed_batch(client=client, batch=batch)
+                embeddings.extend(embeddings)
+        except Exception as e:
+            raise self.wrap_error(e=e)
+        for element, embedding in zip(elements_with_text, embeddings):
+            element[EMBEDDINGS_KEY] = embedding
+        return elements
+
+    async def _embed_query(self, query: str) -> list[float]:
+        client = self.get_client()
+        embeddings = await self.embed_batch(client=client, batch=[query])
+        return embeddings[0]
 
-    @abstractmethod
     async def embed_query(self, query: str) -> list[float]:
-        pass
-
-    async def _embed_documents(self, elements: list[str]) -> list[list[float]]:
-        results = await asyncio.gather(*[self.embed_query(query=text) for text in elements])
-        return results
+        try:
+            return await self._embed_query(query=query)
+        except Exception as e:
+            raise self.wrap_error(e=e)

unstructured_ingest/embed/mixedbreadai.py

@@ -1,4 +1,3 @@
-import asyncio
 import os
 from dataclasses import dataclass
 from typing import TYPE_CHECKING
@@ -6,12 +5,10 @@ from typing import TYPE_CHECKING
 from pydantic import Field, SecretStr
 
 from unstructured_ingest.embed.interfaces import (
-    EMBEDDINGS_KEY,
     AsyncBaseEmbeddingEncoder,
     BaseEmbeddingEncoder,
     EmbeddingConfig,
 )
-from unstructured_ingest.utils.data_prep import batch_generator
 from unstructured_ingest.utils.dep_check import requires_dependencies
 
 USER_AGENT = "@mixedbread-ai/unstructured"
@@ -85,7 +82,7 @@ class MixedbreadAIEmbeddingEncoder(BaseEmbeddingEncoder):
 
     def get_exemplary_embedding(self) -> list[float]:
         """Get an exemplary embedding to determine dimensions and unit vector status."""
-        return self._embed(["Q"])[0]
+        return self.embed_query(query="Q")
 
     @requires_dependencies(
         ["mixedbread_ai"],
@@ -100,59 +97,19 @@ class MixedbreadAIEmbeddingEncoder(BaseEmbeddingEncoder):
             additional_headers={"User-Agent": USER_AGENT},
         )
 
-    def _embed(self, texts: list[str]) -> list[list[float]]:
-        """
-        Embed a list of texts using the Mixedbread AI API.
-
-        Args:
-            texts (list[str]): List of texts to embed.
-
-        Returns:
-            list[list[float]]: List of embeddings.
-        """
-
-        responses = []
-        client = self.config.get_client()
-        for batch in batch_generator(texts, batch_size=self.config.batch_size or len(texts)):
-            response = client.embeddings(
-                model=self.config.embedder_model_name,
-                normalized=True,
-                encoding_format=ENCODING_FORMAT,
-                truncation_strategy=TRUNCATION_STRATEGY,
-                request_options=self.get_request_options(),
-                input=batch,
-            )
-            responses.append(response)
-        return [item.embedding for response in responses for item in response.data]
-
-    def embed_documents(self, elements: list[dict]) -> list[dict]:
-        """
-        Embed a list of document elements.
-
-        Args:
-            elements (list[Element]): List of document elements.
-
-        Returns:
-            list[Element]: Elements with embeddings.
-        """
-        elements = elements.copy()
-        elements_with_text = [e for e in elements if e.get("text")]
-        embeddings = self._embed([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]:
-        """
-        Embed a query string.
-
-        Args:
-            query (str): Query string to embed.
-
-        Returns:
-            list[float]: Embedding of the query.
-        """
-        return self._embed([query])[0]
+    def get_client(self) -> "MixedbreadAI":
+        return self.config.get_client()
+
+    def embed_batch(self, client: "MixedbreadAI", batch: list[str]) -> list[list[float]]:
+        response = client.embeddings(
+            model=self.config.embedder_model_name,
+            normalized=True,
+            encoding_format=ENCODING_FORMAT,
+            truncation_strategy=TRUNCATION_STRATEGY,
+            request_options=self.get_request_options(),
+            input=batch,
+        )
+        return [datum.embedding for datum in response.data]
 
 
 @dataclass
@@ -162,8 +119,7 @@ class AsyncMixedbreadAIEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
 
     async def get_exemplary_embedding(self) -> list[float]:
         """Get an exemplary embedding to determine dimensions and unit vector status."""
-        embedding = await self._embed(["Q"])
-        return embedding[0]
+        return await self.embed_query(query="Q")
 
     @requires_dependencies(
         ["mixedbread_ai"],
@@ -178,58 +134,16 @@ class AsyncMixedbreadAIEmbeddingEncoder(AsyncBaseEmbeddingEncoder):
             additional_headers={"User-Agent": USER_AGENT},
         )
 
-    async def _embed(self, texts: list[str]) -> list[list[float]]:
-        """
-        Embed a list of texts using the Mixedbread AI API.
-
-        Args:
-            texts (list[str]): List of texts to embed.
-
-        Returns:
-            list[list[float]]: List of embeddings.
-        """
-        client = self.config.get_async_client()
-        tasks = []
-        for batch in batch_generator(texts, batch_size=self.config.batch_size or len(texts)):
-            tasks.append(
-                client.embeddings(
-                    model=self.config.embedder_model_name,
-                    normalized=True,
-                    encoding_format=ENCODING_FORMAT,
-                    truncation_strategy=TRUNCATION_STRATEGY,
-                    request_options=self.get_request_options(),
-                    input=batch,
-                )
-            )
-        responses = await asyncio.gather(*tasks)
-        return [item.embedding for response in responses for item in response.data]
-
-    async def embed_documents(self, elements: list[dict]) -> list[dict]:
-        """
-        Embed a list of document elements.
-
-        Args:
-            elements (list[Element]): List of document elements.
-
-        Returns:
-            list[Element]: Elements with embeddings.
-        """
-        elements = elements.copy()
-        elements_with_text = [e for e in elements if e.get("text")]
-        embeddings = await self._embed([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]:
-        """
-        Embed a query string.
-
-        Args:
-            query (str): Query string to embed.
-
-        Returns:
-            list[float]: Embedding of the query.
-        """
-        embedding = await self._embed([query])
-        return embedding[0]
+    def get_client(self) -> "AsyncMixedbreadAI":
+        return self.config.get_async_client()
+
+    async def embed_batch(self, client: "AsyncMixedbreadAI", batch: list[str]) -> list[list[float]]:
+        response = await client.embeddings(
+            model=self.config.embedder_model_name,
+            normalized=True,
+            encoding_format=ENCODING_FORMAT,
+            truncation_strategy=TRUNCATION_STRATEGY,
+            request_options=self.get_request_options(),
+            input=batch,
+        )
+        return [datum.embedding for datum in response.data]