Re: Odd bug, apparently in ring.middleware or clojure.core.memoize?

2014-11-14 Thread James Reeves
On 14 November 2014 13:56, Simon Brooke wrote: > Thank you very much for the suggestion. I believe that your diagnosis is > correct; unfortunately, your solution does not work. Furthermore, even if I > remove the dependency on clj-jgit and all references to it from my code, do > a 'lein clean', a

Re: Odd bug, apparently in ring.middleware or clojure.core.memoize?

2014-11-14 Thread Chris Sims
I had a problem very similar to yours using clj-jgit - I ended up simply excluding jgit from pulling in core.memoize as a dependency: [clj-jgit "0.8.0" :exclusions [org.clojure/core.memoize]] > On Nov 14, 2014, at 5:56 AM, Simon Brooke wrote: > > Thank you very much for the suggestion

Re: Odd bug, apparently in ring.middleware or clojure.core.memoize?

2014-11-14 Thread Simon Brooke
Thank you very much for the suggestion. I believe that your diagnosis is correct; unfortunately, your solution does not work. Furthermore, even if I remove the dependency on clj-jgit and all references to it from my code, do a 'lein clean', and rebuild, I still get the same error on startup (al

Re: Odd bug, apparently in ring.middleware or clojure.core.memoize?

2014-11-13 Thread James Reeves
The problem is that your dependency tree pulls in core.memoize 0.5.3, but other libraries have a dependency on 0.5.6. If you run "lein deps :tree" you can see this for yourself: [clj-jgit "0.8.1"] -> [org.clojure/core.memoize "0.5.3"] overrides [lib-noir "0.9.4"] -> [ring-middleware-f

Odd bug, apparently in ring.middleware or clojure.core.memoize?

2014-11-13 Thread Simon Brooke
Hi everyone Yesterday I proudly announced my new Wiki engine; today I'm investigating an odd bug which prevents it compiling on some Ubuntu machines, but not my laptop (on which I did the development), and I'm really puzzled by it. Top level outline: When one clones the repository from https: