Jon Bracy

Would love this! I was looking to replace JSON in our stack with MessagePack but can't because we stream JSON out to clients. Because the element count is unknown (1...

@markusschaber Correct the format doesn't block the the receiver; I don't think you need to compare the count with the number from the header, if it's not equal MessagePack should...

This solution solves both of those and still allow people to call save in `after_save` and `after_validation` callbacks. 😱. In the [past PR](#46438) I've tried it on a connection variable...

Leaving this [gist](https://gist.github.com/malomalo/269bda0439b1402e432acdb1fcabd910) here for anyone who has the same issue or think they might. You can just add this to an `ext` folder in your Rails app and require...

@PikachuEXE Can you run the reports yourself and see if you see similar results? The dirk test make rabl/oj and jbuilder/oj look realy slow 🧐; I know it was faster...

Interesting, I wonder why there is a big difference now? I think the `dirk` test is a caching test no? Need to do some investigating....

Added to the test suite over at https://github.com/joshmn/json-benchmark/pull/27 as well; hopefully it gets pulled in.

Just tried some debugging to see what was going on. Nothing has helped. The last think I can __maybe__ think of is `json` gem performance regressions or patches for security;...

I can have some coworkers run it and see if they get the same results I'll merge. re: graphs I was playing around with smaller output for the README

@benubois I tried a similar approach a while back from what I recall there were numerous cases where the template didn't contain enough information about where the cache was being...