Fix missing sensitive fields for embedders (#3263)

A few embedder types are missing sensitive field annotations.

Author: vangheem

Diff for: CHANGELOG.md

@@ -10,6 +10,8 @@
 
 * **`partition()` now forwards `strategy` arg to `partition_docx()`, `partition_ppt()`, and `partition_pptx()`.** A `strategy` argument passed to `partition()` (or the default value "auto" assigned by `partition()`) is now forwarded to `partition_docx()`, `partition_ppt()`, and `partition_pptx()` when those filetypes are detected.
 
+* **Fix missing sensitive field markers** for embedders
+
 ## 0.14.7
 
 ### Enhancements

Diff for: unstructured/embed/octoai.py

@@ -7,6 +7,7 @@
     Element,
 )
 from unstructured.embed.interfaces import BaseEmbeddingEncoder, EmbeddingConfig
+from unstructured.ingest.enhanced_dataclass import enhanced_field
 from unstructured.ingest.error import EmbeddingEncoderConnectionError
 from unstructured.utils import requires_dependencies
 
@@ -18,7 +19,7 @@
 
 @dataclass
 class OctoAiEmbeddingConfig(EmbeddingConfig):
-    api_key: str
+    api_key: str = enhanced_field(sensitive=True)
     model_name: str = "thenlper/gte-large"
 
 

Diff for: unstructured/embed/openai.py

@@ -7,6 +7,7 @@
     Element,
 )
 from unstructured.embed.interfaces import BaseEmbeddingEncoder, EmbeddingConfig
+from unstructured.ingest.enhanced_dataclass import enhanced_field
 from unstructured.ingest.error import EmbeddingEncoderConnectionError
 from unstructured.utils import requires_dependencies
 
@@ -16,7 +17,7 @@
 
 @dataclass
 class OpenAIEmbeddingConfig(EmbeddingConfig):
-    api_key: str
+    api_key: str = enhanced_field(sensitive=True)
     model_name: str = "text-embedding-ada-002"
 
 

Diff for: unstructured/embed/vertexai.py

@@ -10,6 +10,7 @@
     Element,
 )
 from unstructured.embed.interfaces import BaseEmbeddingEncoder, EmbeddingConfig
+from unstructured.ingest.enhanced_dataclass import enhanced_field
 from unstructured.ingest.error import EmbeddingEncoderConnectionError
 from unstructured.utils import FileHandler, requires_dependencies
 
@@ -19,7 +20,7 @@
 
 @dataclass
 class VertexAIEmbeddingConfig(EmbeddingConfig):
-    api_key: str
+    api_key: str = enhanced_field(sensitive=True)
     model_name: Optional[str] = "textembedding-gecko@001"
 
 

Diff for: unstructured/embed/voyageai.py

@@ -5,6 +5,7 @@
 
 from unstructured.documents.elements import Element
 from unstructured.embed.interfaces import BaseEmbeddingEncoder, EmbeddingConfig
+from unstructured.ingest.enhanced_dataclass import enhanced_field
 from unstructured.ingest.error import EmbeddingEncoderConnectionError
 from unstructured.utils import requires_dependencies
 
@@ -14,7 +15,7 @@
 
 @dataclass
 class VoyageAIEmbeddingConfig(EmbeddingConfig):
-    api_key: str
+    api_key: str = enhanced_field(sensitive=True)
     model_name: str
     batch_size: Optional[int] = None
     truncation: Optional[bool] = None