On Sat, 23 Jul 2005 18:05:04 +0100, Fred Spiessens wrote (in message <[EMAIL PROTECTED]>):
> > Simon, > > thanks for the encouraging words ;-) > About the time to be spent: I think it will be certainly worth the > effort, if it allows us to attract new oz programmers, and more > importantly, if it eases the learning curve and improves the overall > experience of Oz programming. Being an old fart myself, I know from > experience that a good editor can considerably boost programmer (and > designer) productivity, so it's not just the nice look I appreciate in > Eclipse, but its extensibility. > This is a good motive I agree that the extensibility of eclipse is makes it a good platform (like emacs in its day) I have found it to be very robust and the UI is well worth the effort - minimal efforts are rewarded with excellent integration and reuse of all the existing components. However the excellent java ide tools represent a consdierable amount of effort on top of that framework. > One of the things I want in an Oz OPI is a minimum of context sensitive > support for syntax. Maybe the support for multiple paradigms made the > interfaces of the Oz base and system modules just harder to remember, or > maybe I'm really getting old and slug in assimilating new languages, but > when programming in Oz it's amazing how often one has to recheck List, > Record, Module, etc, to find out the exact syntax of a procedure, and > the order of its arguments. I made a little survey in our research group > at UCL, and it's not just me fighting with this. hmm that is true in most languges with extensive apis - the code sense and completion stuff in the jave ide is very compelling when using a new api; the hierarchy/outline views also. ... > > Since you had experience, ... any chance we can reuse lisp and/or mozart > code in an Eclipse plugin? > Any chance we could get you to contribute to such a project? I'll review the code I did for the prolog and make it available - I did an outline view - and consult/reconsult functions in the explorer/outline and text editor views - no nice syntax sensitive stuff and formatting though. Even the oultine view required a shallow parse of the source code - it's not too daunting a task - maybe an antlr oz source parser is feasbile here. You'd have to take a close look at the java editor for some hints on this. Also I did background tasks (for long running evals). It's a similar paradigm for oz - take a look at the emacs OPI code to get a view on the integration effort. Full communication with the oz compiler and oz debugger would be the first task (sine qua non). > > > cheers, > > Fred. Cheers, Simon _________________________________________________________________________________ mozart-users mailing list [email protected] http://www.mozart-oz.org/mailman/listinfo/mozart-users
