This should speed up compilation and reduce binary size (as aws-client is one of our slowest to compile and biggest dependencies).
Also refactors the flushing logic a bit to use our new retry! macro, and fixes a regression from 0.10 that could cause failed flushes if a batch was too big.
This should speed up compilation and reduce binary size (as aws-client is one of our slowest to compile and biggest dependencies).
Also refactors the flushing logic a bit to use our new
retry!macro, and fixes a regression from 0.10 that could cause failed flushes if a batch was too big.