2016-10-18 8:42 GMT+02:00 Daniel Compton <[email protected]>:
> > I'd really think twice before requiring your users to insert matching > pairs of mutation in their code. It goes against the grain of clojure and > is error-prone. > > Why can’t on-jsload (is that figwheel's?) just run the update within a > (binding [*warn-on-overwrite* false] ...)? > > The (simplified) structure of the Figwheel reload code is > > (do > (before-jsload) > (figwheel-reloads-code) > (on-jsload)) > > I would prefer to use binding, but Figwheel doesn’t run it’s reload within > the scope of my before-jsload function, so binding and with-redefs wouldn’t > work here. I’m not in love with the matching mutations, but I’m not sure if > there’s a better option? A PR to Figwheel is always possible, but this is a > fairly niche requirement, so I’m hesitant to add more configuration to it. > Maybe you can get rid of load-time side-effects, such that a reload won't trigger warnings, but only the main function, called by on-jsload, would? Even though, this is out of scope for re-frame and figwheel alone, some more thoughts: Setting up the environment of your program doesn't seem like such a niche thing to me. Other use cases include reader tags, various (pretty-)print variables, ... The tricky part is finding an approach that works across: a) cljs classic (where the macro environment is clj) b) cljs self-hosted c) repls / dynamic loaders, like figwheel -- Note that posts from new members are moderated - please be patient with your first post. --- You received this message because you are subscribed to the Google Groups "ClojureScript" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/clojurescript.
