Closed brigadier closed 8 years ago
no i will revert the change. I had an idea to make the build faster and using less ram :) On Thu, 4 Aug 2016 at 16:22, Evgeny M. notifications@github.com wrote:
Is it really that necessary to have nif-based stringprep as a dep of idna and indirectly of hackney? It is used once for unicode normalization, maybe it would be better to leave the normalization for the caller?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/benoitc/erlang-idna/issues/14, or mute the thread https://github.com/notifications/unsubscribe-auth/AAA4ohZisgdHSbnTlCQ9UYfCarSvRqdoks5qcfWhgaJpZM4Jcv5M .
@brigadier b219f73b5fc4e8ec65cf669c9ace217766e03136 reverted it. I will publish a new version later tonight. Sorry for the inconvenience.
Is it really that necessary to have nif-based stringprep as a dep of idna and indirectly of hackney? It is used once for unicode normalization, maybe it would be better to leave the normalization for the caller?