On 23 May 2011, at 13:23, Hugo Duncan wrote: >> Out of interest, how do swank-clj and swank-clojure (with merged in cdt >> stuff) compare? > > I would have to let others comment on the user experience, not having used > the cdt support in swank-clojure. From reading the code, I see two > differences. > > i) swank-clj uses SLDB vs the use of GUD in swank-clojure, > > ii) swank-clj starts two jvms, one for the debugger and one for the debuggee. > If I understand correctly, CDT uses a single jvm process. > > In swank-clj at the moment, a swank server runs in both jvms, but I am > working towards being able to run using just JPDA/JDI to talk to the debugee. > At that point it will be possible to attach the debugger to any jvm process > running with debugging enabled. >
Thanks for the explanation. I'm relatively new to Emacs - and certainly new to any debugging support. I hadn't heard of the Slime debugger or the "Grand Unified Debugger" before. Given my current understanding of the Emacs ecosystem it doesn't surprise me that there is more than one supported approach to debugging :-) What I do think is interesting is that there are two different approaches to Clojure debugging within Emacs that are relatively young in terms of implementation maturity. Competing systems can be good for cross-fertilisation and motivation - however they can also be confusing to the newcomer when faced with a choice. Currently I'm that newcomer and I have been using swank-clojure only because it was the only way I knew. George's work on the CDT has been very exciting and I was thrilled to see it get merged in with swank-clojure. However, I now see swank-clj which also seems remarkably exciting - truly cool stuff. It appears both approaches are incompatible - is this true? If so, I'm wondering what the merits of SLDB over GUD and visa versa are. SLDB sound great cos it has slime in the name and, after all, I'm using slime. However, GUD sounds amazing cos it has the words grand and unified in the title. My initial thoughts are focussed and bespoke vs generalised and abstract. Which personally makes me think SLDB. Clearly choosing purely by name is utter madness though. So if anyone has any more information that can help people decide which approach to invest their time learning that would be much appreciated. Sam --- http://sam.aaron.name -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
