Rex Zhou
Rex Zhou
yes, just worked on the merge and repo transfer, and everything lives under openflagr org now.
will prepare a release very soon
https://github.com/openflagr/flagr/releases/tag/1.1.14 is live
Can you share what's the error message? I wonder if you miss any authentication or SSL config for the aws kafka connection.
This is a very interesting use case I didn't think of before, can you elaborate on how are you going to use 100+ variants? Flagr's model uses static `1000` buckets...
Yes, however segment has attributes like "rollout percent" and "distribution" that's usually not sharable across flags. I'm think you are likely referring to a concept called "Audience" (or constraints set,...
Yes, definitely doable to "split the input boxes into two". However, I also want to support the use case of generating the keys to be explicitly at random with unguessable...
The closest research I can find is https://michiel.buddingh.eu/distribution-of-hash-values#summary, which has a summary section says CRC32 is a good choice: > >  The value space of CRC32 is smaller than...
@victor-mariano-leite I added a test in the openflagr repo to verify my hypothesis https://github.com/openflagr/flagr/pull/35 I would also check the distribution of `entity_id` from the input data. Due to the deterministic...
Thanks, that generic callback endpoint is nice to have! I do think that if we have some control over the remote server, e.g. setting/passing Auth headers or configuring callback URL/headers,...