Package com.tnsai.llm.errors

Class BedrockProviderErrorMapper

java.lang.Object
com.tnsai.llm.errors.BedrockProviderErrorMapper
All Implemented Interfaces:
ProviderErrorMapper
public class BedrockProviderErrorMapper extends Object implements ProviderErrorMapper
ProviderErrorMapper for AWS Bedrock — sixth pilot mapper.

Bedrock's error envelope uses the standard AWS shape, which comes in two variations across services / endpoints: 


 // modern shape (most Bedrock APIs)
 { "__type": "ResourceNotFoundException",
   "message": "Could not resolve the foundation model from the provided model identifier." }

 // legacy XML-derived shape (some older paths)
 { "Code": "ThrottlingException", "Message": "Rate exceeded" }

The mapper tries both — __type → Code → fallback. Whichever wins, the AWS error code (e.g. ThrottlingException) is the durable signal; AWS pads it onto multiple HTTP statuses.

Different from the other 5 mappers: not invoked by executeRequest

BedrockClient uses the AWS SDK's BedrockRuntimeClient.invokeModel, not raw HTTP through AbstractLLMClient.executeRequest. Mapper invocation is therefore done at the BedrockClient.chat catch site by extracting status / body / headers from AwsServiceException and calling map(int, Map, String, String, Throwable) directly. The mapper itself doesn't know or care about the SDK — it sees the same shape as it would over HTTP.

Status routing

  • ResourceNotFoundException (404) → MODEL_NOT_FOUND
  • AccessDeniedException (403) / UnrecognizedClientException (401 — invalid AWS creds) → AUTHENTICATION_FAILED
  • ThrottlingException / TooManyRequestsException (429) / ServiceQuotaExceededException → MODEL_OVERLOADED
  • ValidationException (400) → INVALID_REQUEST, demoted to CONTEXT_TOO_LONG when message hints at token-count overflow
  • ModelTimeoutException (408) → SERVER_ERROR (retryable)
  • ServiceUnavailableException (503) → MODEL_OVERLOADED (transient backend pressure)
  • InternalServerException / ModelErrorException / ModelStreamErrorException (500) → SERVER_ERROR
  • HTTP fallback when no AWS code recognised: 5xx → SERVER_ERROR; 401/403 → AUTHENTICATION_FAILED; 429 → MODEL_OVERLOADED; 4xx → INVALID_REQUEST

Captured headers: x-amzn-requestid, x-amzn-errortype, retry-after (rare on Bedrock — AWS prefers exponential backoff via SDK retry config).

Since:
0.5.3 (issue #87 follow-up — sixth pilot mapper)

  • Constructor Details

    • BedrockProviderErrorMapper

      public BedrockProviderErrorMapper()

  • Method Details

    • providerId

      public String providerId()
      Specified by:
      providerId in interface ProviderErrorMapper
      Returns:
      short, lowercase provider id (e.g. "openai", "anthropic"). Used to route a raw response to the right mapper.

    • map

      public com.tnsai.exceptions.LLMException map(int httpStatus, Map<String,String> responseHeaders, String responseBody, String model, Throwable cause)
      Description copied from interface: ProviderErrorMapper
      Translate a raw HTTP error response into a typed LLMException.
      Specified by:
      map in interface ProviderErrorMapper
      Parameters:
      httpStatus - HTTP status code from the response
      responseHeaders - selected response headers (lower-cased keys); empty map if none captured
      responseBody - raw response body; mapper parses provider-specific envelope (e.g. JSON error.code)
      model - model identifier from the request, for the resulting LLMException.getModel()
      cause - underlying transport throwable, may be null when the failure is purely a non-2xx response
      Returns:
      typed LLMException with ProviderDetails attached

    • mapErrorType

      public Optional<com.tnsai.exceptions.LLMException.ErrorType> mapErrorType(String providerCode)
      Description copied from interface: ProviderErrorMapper
      Map a provider-native error code to a framework LLMException.ErrorType. Implementations should err on the side of LLMException.ErrorType.UNKNOWN for codes they don't recognize so that a new provider error doesn't fail the whole mapping path.
      Specified by:
      mapErrorType in interface ProviderErrorMapper
      Parameters:
      providerCode - native code (e.g. "context_length_exceeded")
      Returns:
      matching LLMException.ErrorType, or Optional.empty() if the mapper has no opinion