Since the fresh new official Guide sparked some new interest in using webpack with ClojureScript I thought I’d write up my experience in trying to combine webpack+CLJS and why shadow-cljs ultimately went a different path.
https://code.thheller.com/blog/shadow-cljs/2018/06/15/why-not-webpack.html
Although I am not using shadow-cljs yet, but after several attempts I have similar conclusion. Thanks for bringing this on the table for discussion!
Left us hanging a bit on the “How it works in shadow-cljs now” with the lack of examples—which were great up until that point 
Can you show the previous foo.js and bar.js examples and how shadow-cljs processes them today? And maybe clarify how this method solves the problems detailed in the post.
I’ll see about adding that although I’m not sure how interesting the actual code is. It is functionally identical to webpack. Create a function wrapper, invoke it at some point, expose exports to others.
I really appreciate the “perfectionist” approach in addressing these problems, @thheller. Webpack is definitely the 80% solution, but getting the last 20% in terms of performance, developer experience, and all of the various use cases is a lot of time and effort. Thank you so much for contributing that to the community!
I expanded the “How it works in shadow-cljs now” section a bit. Hopefully the examples capture all the important details now.
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.