@danielcompton excellent writeup. Removing the (incorrect!) Last-Modified header might be a good thing to do as well in figwheel

Yeah it’s a bit of a red herring really for CLJS

What’s the source on https://clojurians.slack.com/archives/CALJ3BFLP/p1535722393000100 ?

Is that an RFC?

Was wondering if the ring not modified middleware needed another PR

@danielcompton https://tools.ietf.org/html/rfc7232#section-3.3

I'm not sure if the CLJS compiler's decision is a good one, i.e. to set the mtime of js files written when compiled to match the mtime of the source cljs file, which is always out of date by that point (and constant on subsequent macro recompiles)

it would be absolutely fabulous if the compiler had stable random number generation for the temp variables. This way the compiled output would only change if there was actually a change to the source code.

the date copying is a weak proxy for that behavior

because of macros like run-tests