Closed cgmartin closed 11 years ago
I seem to recall that was intentional, but it's been a long time. Does this fix it? Is it possible to add a test? I don't suppose I could entice you to do some archaeology to find why this was added?
Yes, this fixes it for me. I followed directions here to run my cloned branch of dieter with my compojure project. When I change back to checkout master from my branch I see it break. Again, clojure/ring is all new to me so it wouldn't surprise me if I've made some mistakes.
I'll be happy to attempt a unit test. That will be a good learning exercise. I can also try coming up with a small project to reproduce, using lein new compojure testproj
(if it would be helpful).
The little research I did back through the git commit log seemed to point to that merge commit. I used blame and looked at the lines within asset-pipeline
function.
Cheers!
Made an attempt at some unit tests around asset-pipeline
to test the mime types.
I'm not entirely certain of clojure tests best practices, so I'm very open to feedback. Thanks!
Looks great, thanks!
First off, thank you for sharing dieter! I am new to clojure, and for my first compojure project I have been trying to integrate dieter 0.4.0. I have a
site.css.dieter
file that always returns mime-typeapplication/octet-stream
in development mode. But, it works correctly in production mode, returningtext/css
.I noticed an extra
wrap-file-info
that may have been introduced in a bad merge commit? https://github.com/edgecase/dieter/commit/b018f86fb0ff09f2fc99b9c1949a6ff736695ef1Hope this helps!