ci: fix OCR lint and patch coverage

codecov.yaml

@@ -3,6 +3,9 @@ codecov:
   notify:
     wait_for_ci: false # post as soon as expected uploads arrive, don't wait on CI
 
+ignore:
+  - "litellm-rust/**"
+
 # Uploads are flagged per workflow/shard (GHA) or "circleci". carryforward makes
 # a re-upload of a flag replace its prior session instead of accumulating a
 # conflicting one, and lets a commit reuse a flag from its parent when that flag

litellm/llms/azure_ai/ocr/document_intelligence/transformation.py

@@ -11,7 +11,7 @@ The operation location must be polled until the analysis completes.
 import asyncio
 import re
 import time
-from typing import Any, Dict, Optional
+from typing import Any, Dict
 from urllib.parse import quote
 
 import httpx
@@ -56,7 +56,7 @@ class AzureDocumentIntelligenceOCRConfig(BaseOCRConfig):
     def __init__(self) -> None:
         super().__init__()
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         return AZURE_DOCUMENT_INTELLIGENCE_API_KEY_ENV_VAR
 
     def get_supported_ocr_params(self, model: str) -> list:
@@ -149,9 +149,9 @@ class AzureDocumentIntelligenceOCRConfig(BaseOCRConfig):
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -187,10 +187,10 @@ class AzureDocumentIntelligenceOCRConfig(BaseOCRConfig):
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/llms/azure_ai/ocr/transformation.py

@@ -2,7 +2,7 @@
 Azure AI OCR transformation implementation.
 """
 
-from typing import Dict, Optional
+from typing import Dict
 
 from litellm._logging import verbose_logger
 from litellm.litellm_core_utils.prompt_templates.image_handling import (
@@ -32,16 +32,16 @@ class AzureAIOCRConfig(MistralOCRConfig):
     def __init__(self) -> None:
         super().__init__()
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         return AZURE_AI_OCR_API_KEY_ENV_VAR
 
     def validate_environment(
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -77,10 +77,10 @@ class AzureAIOCRConfig(MistralOCRConfig):
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/llms/base_llm/ocr/transformation.py

@@ -2,7 +2,7 @@
 Base OCR transformation configuration.
 """
 
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Union
 
 import httpx
 from pydantic import PrivateAttr
@@ -25,16 +25,16 @@ DocumentType = Dict[str, str]
 class OCRPageDimensions(LiteLLMPydanticObjectBase):
     """Page dimensions from OCR response."""
 
-    dpi: Optional[int] = None
-    height: Optional[int] = None
-    width: Optional[int] = None
+    dpi: int | None = None
+    height: int | None = None
+    width: int | None = None
 
 
 class OCRPageImage(LiteLLMPydanticObjectBase):
     """Image extracted from OCR page."""
 
-    image_base64: Optional[str] = None
-    bbox: Optional[Dict[str, Any]] = None
+    image_base64: str | None = None
+    bbox: Dict[str, Any] | None = None
 
     model_config = {"extra": "allow"}
 
@@ -44,8 +44,8 @@ class OCRPage(LiteLLMPydanticObjectBase):
 
     index: int
     markdown: str
-    images: Optional[List[OCRPageImage]] = None
-    dimensions: Optional[OCRPageDimensions] = None
+    images: List[OCRPageImage] | None = None
+    dimensions: OCRPageDimensions | None = None
 
     model_config = {"extra": "allow"}
 
@@ -53,9 +53,9 @@ class OCRPage(LiteLLMPydanticObjectBase):
 class OCRUsageInfo(LiteLLMPydanticObjectBase):
     """Usage information from OCR response."""
 
-    pages_processed: Optional[int] = None
-    credits: Optional[float] = None
-    doc_size_bytes: Optional[int] = None
+    pages_processed: int | None = None
+    credits: float | None = None
+    doc_size_bytes: int | None = None
 
     model_config = {"extra": "allow"}
 
@@ -68,8 +68,8 @@ class OCRResponse(LiteLLMPydanticObjectBase):
 
     pages: List[OCRPage]
     model: str
-    document_annotation: Optional[Any] = None
-    usage_info: Optional[OCRUsageInfo] = None
+    document_annotation: Any | None = None
+    usage_info: OCRUsageInfo | None = None
     object: str = "ocr"
 
     model_config = {"extra": "allow"}
@@ -81,8 +81,8 @@ class OCRResponse(LiteLLMPydanticObjectBase):
 class OCRRequestData(LiteLLMPydanticObjectBase):
     """OCR request data structure."""
 
-    data: Optional[Union[Dict, bytes]] = None
-    files: Optional[Dict[str, Any]] = None
+    data: Union[Dict, bytes] | None = None
+    files: Dict[str, Any] | None = None
 
 
 class BaseOCRConfig:
@@ -101,7 +101,7 @@ class BaseOCRConfig:
         """
         return []
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         """
         Return the provider-specific API key environment variable name, if any.
         """
@@ -120,9 +120,9 @@ class BaseOCRConfig:
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -133,10 +133,10 @@ class BaseOCRConfig:
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/llms/mistral/ocr/transformation.py

@@ -2,7 +2,7 @@
 Mistral OCR transformation implementation.
 """
 
-from typing import Any, Dict, Optional
+from typing import Any, Dict
 
 import httpx
 
@@ -61,7 +61,7 @@ class MistralOCRConfig(BaseOCRConfig):
             "id",
         ]
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         return MISTRAL_OCR_API_KEY_ENV_VAR
 
     def map_ocr_params(
@@ -90,9 +90,9 @@ class MistralOCRConfig(BaseOCRConfig):
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -118,10 +118,10 @@ class MistralOCRConfig(BaseOCRConfig):
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/llms/vertex_ai/ocr/deepseek_transformation.py

@@ -3,7 +3,7 @@ Vertex AI DeepSeek OCR transformation implementation.
 """
 
 import json
-from typing import TYPE_CHECKING, Any, Dict, Optional
+from typing import TYPE_CHECKING, Any, Dict
 
 import httpx
 
@@ -38,16 +38,16 @@ class VertexAIDeepSeekOCRConfig(BaseOCRConfig):
         super().__init__()
         self.vertex_base = VertexBase()
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         return VERTEX_AI_DEEPSEEK_OCR_API_KEY_ENV_VAR
 
     def validate_environment(
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -89,10 +89,10 @@ class VertexAIDeepSeekOCRConfig(BaseOCRConfig):
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/llms/vertex_ai/ocr/transformation.py

@@ -2,7 +2,7 @@
 Vertex AI Mistral OCR transformation implementation.
 """
 
-from typing import Dict, Optional
+from typing import Dict
 
 from litellm._logging import verbose_logger
 from litellm.litellm_core_utils.prompt_templates.image_handling import (
@@ -34,16 +34,16 @@ class VertexAIOCRConfig(MistralOCRConfig):
         super().__init__()
         self.vertex_base = VertexBase()
 
-    def get_api_key_env_var(self) -> Optional[str]:
+    def get_api_key_env_var(self) -> str | None:
         return VERTEX_AI_OCR_API_KEY_ENV_VAR
 
     def validate_environment(
         self,
         headers: Dict,
         model: str,
-        api_key: Optional[str] = None,
-        api_base: Optional[str] = None,
-        litellm_params: Optional[dict] = None,
+        api_key: str | None = None,
+        api_base: str | None = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> Dict:
         """
@@ -85,10 +85,10 @@ class VertexAIOCRConfig(MistralOCRConfig):
 
     def get_complete_url(
         self,
-        api_base: Optional[str],
+        api_base: str | None,
         model: str,
         optional_params: dict,
-        litellm_params: Optional[dict] = None,
+        litellm_params: dict | None = None,
         **kwargs,
     ) -> str:
         """

litellm/ocr/main.py

@@ -9,7 +9,7 @@ import os
 import re
 from dataclasses import dataclass
 from io import IOBase
-from typing import Any, Callable, Coroutine, Optional, Union, cast
+from typing import Any, Callable, Coroutine, Union, cast
 
 import httpx
 
@@ -38,10 +38,10 @@ base_llm_http_handler = BaseLLMHTTPHandler()
 class _PreparedOCRRequest:
     model: str
     document: dict[str, Any]
-    api_key: Optional[str]
-    api_base: Optional[str]
+    api_key: str | None
+    api_base: str | None
     custom_llm_provider: str
-    extra_headers: Optional[dict[str, object]]
+    extra_headers: dict[str, object] | None
     provider_config: BaseOCRConfig
     optional_params: dict[str, object]
     litellm_params: dict[str, object]
@@ -51,8 +51,8 @@ class _PreparedOCRRequest:
 
 @dataclass
 class _PreparedRustOCRCall:
-    api_key: Optional[str]
-    api_base: Optional[str]
+    api_key: str | None
+    api_base: str | None
     headers: dict[str, object]
     optional_params: dict[str, object]
 
@@ -66,8 +66,8 @@ _RUST_OCR_PROVIDERS = {
 
 
 def _timeout_to_seconds(
-    timeout: Optional[Union[float, httpx.Timeout]],
-) -> Optional[float]:
+    timeout: Union[float, httpx.Timeout] | None,
+) -> float | None:
     """Convert the Python OCR timeout to a single seconds value for the Rust bridge.
 
     The Rust HTTP client takes one duration; ``httpx.Timeout`` carries separate
@@ -84,15 +84,15 @@ def _timeout_to_seconds(
 def _prepare_ocr_request(
     model: str,
     document: dict[str, Any],
-    api_key: Optional[str],
-    api_base: Optional[str],
-    timeout: Optional[Union[float, httpx.Timeout]],
-    custom_llm_provider: Optional[str],
-    extra_headers: Optional[dict[str, Any]],
+    api_key: str | None,
+    api_base: str | None,
+    timeout: Union[float, httpx.Timeout] | None,
+    custom_llm_provider: str | None,
+    extra_headers: dict[str, Any] | None,
     kwargs: dict[str, Any],
 ) -> _PreparedOCRRequest:
     litellm_logging_obj = cast(LiteLLMLoggingObj, kwargs.pop("litellm_logging_obj"))
-    litellm_call_id = cast(Optional[str], kwargs.get("litellm_call_id", None))
+    litellm_call_id = cast(str | None, kwargs.get("litellm_call_id", None))
 
     if not isinstance(document, dict):
         raise ValueError(
@@ -173,7 +173,7 @@ def _prepare_ocr_request(
         api_key=api_key,
         api_base=api_base,
         custom_llm_provider=custom_llm_provider,
-        extra_headers=cast(Optional[dict[str, object]], extra_headers),
+        extra_headers=cast(dict[str, object] | None, extra_headers),
         provider_config=ocr_provider_config,
         optional_params=cast(dict[str, object], optional_params),
         litellm_params=dict(litellm_params),
@@ -188,7 +188,7 @@ def _rust_ocr_supported(prepared_request: _PreparedOCRRequest) -> bool:
 
 def _rust_bridge_optional_params(
     prepared_request: _PreparedOCRRequest,
-    resolve_secret: Callable[[str], Optional[str]],
+    resolve_secret: Callable[[str], str | None],
 ) -> dict[str, object]:
     optional_params = dict(prepared_request.optional_params)
     if prepared_request.custom_llm_provider == "vertex_ai":
@@ -214,8 +214,8 @@ def _rust_bridge_optional_params(
 
 def _rust_bridge_api_base(
     prepared_request: _PreparedOCRRequest,
-    resolve_secret: Callable[[str], Optional[str]],
-) -> Optional[str]:
+    resolve_secret: Callable[[str], str | None],
+) -> str | None:
     if prepared_request.api_base is not None:
         return prepared_request.api_base
     if prepared_request.custom_llm_provider == "azure_ai/doc-intelligence":
@@ -232,7 +232,7 @@ def _rust_bridge_api_base(
 
 def _prepare_rust_ocr_call(
     prepared_request: _PreparedOCRRequest,
-    resolve_api_key: Callable[[str], Optional[str]],
+    resolve_api_key: Callable[[str], str | None],
 ) -> _PreparedRustOCRCall:
     provider_config = prepared_request.provider_config
     api_key_env_var = provider_config.get_api_key_env_var()
@@ -280,7 +280,7 @@ def _prepare_rust_ocr_call(
 def _run_rust_ocr(
     rust_ocr: RustOcr,
     prepared_request: _PreparedOCRRequest,
-    resolve_api_key: Callable[[str], Optional[str]],
+    resolve_api_key: Callable[[str], str | None],
 ) -> OCRResponse:
     """Run the Mistral OCR call through the Rust bridge and wrap the result.
 
@@ -311,7 +311,7 @@ def _run_rust_ocr(
 async def _run_rust_aocr(
     rust_aocr: RustAocr,
     prepared_request: _PreparedOCRRequest,
-    resolve_api_key: Callable[[str], Optional[str]],
+    resolve_api_key: Callable[[str], str | None],
 ) -> OCRResponse:
     prepared = _prepare_rust_ocr_call(
         prepared_request=prepared_request,
@@ -335,11 +335,11 @@ async def _run_rust_aocr(
 async def aocr(
     model: str,
     document: dict[str, Any],
-    api_key: Optional[str] = None,
-    api_base: Optional[str] = None,
-    timeout: Optional[Union[float, httpx.Timeout]] = None,
-    custom_llm_provider: Optional[str] = None,
-    extra_headers: Optional[dict[str, Any]] = None,
+    api_key: str | None = None,
+    api_base: str | None = None,
+    timeout: Union[float, httpx.Timeout] | None = None,
+    custom_llm_provider: str | None = None,
+    extra_headers: dict[str, Any] | None = None,
     **kwargs,
 ) -> OCRResponse:
     """
@@ -536,7 +536,7 @@ def convert_file_document_to_url_document(document: dict[str, Any]) -> dict[str,
 
     file_bytes: bytes
     mime_type: str = "application/octet-stream"
-    file_name: Optional[str] = None
+    file_name: str | None = None
 
     if isinstance(file_input, str):
         # Bare strings are rejected here. The OCR ``document`` accepts a
@@ -606,11 +606,11 @@ def convert_file_document_to_url_document(document: dict[str, Any]) -> dict[str,
 def ocr(
     model: str,
     document: dict[str, Any],
-    api_key: Optional[str] = None,
-    api_base: Optional[str] = None,
-    timeout: Optional[Union[float, httpx.Timeout]] = None,
-    custom_llm_provider: Optional[str] = None,
-    extra_headers: Optional[dict[str, Any]] = None,
+    api_key: str | None = None,
+    api_base: str | None = None,
+    timeout: Union[float, httpx.Timeout] | None = None,
+    custom_llm_provider: str | None = None,
+    extra_headers: dict[str, Any] | None = None,
     **kwargs,
 ) -> Union[OCRResponse, Coroutine[Any, Any, OCRResponse]]:
     """

tests/test_litellm/ocr/test_rust_bridge.py

@@ -732,3 +732,26 @@ def test_ocr_falls_back_to_python_when_bridge_unavailable(monkeypatch):
 
     assert captured.get("called") is True  # Python path was used
     assert isinstance(response, OCRResponse)
+
+
+def test_ocr_provider_configs_expose_api_key_env_vars():
+    from litellm.llms.azure_ai.ocr.document_intelligence.transformation import (
+        AzureDocumentIntelligenceOCRConfig,
+    )
+    from litellm.llms.azure_ai.ocr.transformation import AzureAIOCRConfig
+    from litellm.llms.base_llm.ocr.transformation import BaseOCRConfig
+    from litellm.llms.mistral.ocr.transformation import MistralOCRConfig
+    from litellm.llms.vertex_ai.ocr.deepseek_transformation import (
+        VertexAIDeepSeekOCRConfig,
+    )
+    from litellm.llms.vertex_ai.ocr.transformation import VertexAIOCRConfig
+
+    assert BaseOCRConfig().get_api_key_env_var() is None
+    assert MistralOCRConfig().get_api_key_env_var() == "MISTRAL_API_KEY"
+    assert AzureAIOCRConfig().get_api_key_env_var() == "AZURE_AI_API_KEY"
+    assert (
+        AzureDocumentIntelligenceOCRConfig().get_api_key_env_var()
+        == "AZURE_DOCUMENT_INTELLIGENCE_API_KEY"
+    )
+    assert VertexAIOCRConfig().get_api_key_env_var() == "VERTEX_AI_API_KEY"
+    assert VertexAIDeepSeekOCRConfig().get_api_key_env_var() == "VERTEX_AI_API_KEY"