Add support for AWS SigV4-authenticated OTLP Exporter in Python SDK (Implementation Ready)

What are you trying to achieve?

Enable first-class support for AWS SigV4-authenticated OTLP HTTP exports from the OpenTelemetry Python SDK — particularly to support containerized AWS Lambda apps sending data to AWS X-Ray.

Currently, AWS provides a Lambda Layer for OTLP export, but it is:
• Incompatible with containerized Lambda deployments
• Tightly scoped to environments with preconfigured agents
• Not usable from within standard OTLP Python

What did you expect to see?

A pluggable, officially supported way to send OTLP spans from Python to AWS X-Ray using AWS SigV4 signing (e.g., via an auth extension or built-in configuration). Ideally, this would:
• Respect the existing OTLP exporter structure
• Use botocore for credential resolution and SigV4 signing
• Support compression (gzip/deflate), retries, and timeout config
• Be compatible with OpenTelemetry SDK conventions (e.g., SpanExporter, BatchSpanProcessor)

Additional context

As a temporary solution, I built a custom SigV4OTLPSpanExporter from scratch, fully compatible with:
• xray.{region}.amazonaws.com
• All OTLP span export semantics
• AWS SDK credential chains
• CI pipelines (Docker + LocalStack integration tests)
• Observability best practices

GitHub Repo: https://github.com/Stupidoodle/opentelemetry-exporter-otlp-proto-http-sigv4

I’m happy to contribute this upstream (as core or contrib) or collaborate with others working on auth extension designs. Just let me know what path would be most useful.

Start by reviewing the linked custom SigV4OTLPSpanExporter repository and the OpenTelemetry Python OTLP exporter structure. Check its Docker and LocalStack integration tests, then determine whether the contribution belongs in core or contrib and how it fits SpanExporter and BatchSpanProcessor conventions. Done means the chosen path supports AWS credential resolution, SigV4-authenticated OTLP spans, and the listed export configuration.