From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 102b75,501ec19d1d81daee X-Google-Attributes: gid102b75,public X-Google-Thread: 103376,db9a11afb3da4240 X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2002-03-29 09:08:05 PST Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.online.be!ams.uu.net!dca.uu.net!news.chips.ibm.com!newsfeed.btv.ibm.com!news.btv.ibm.com!hack From: hack@watson.ibm.com (hack) Newsgroups: comp.arch,comp.lang.ada Subject: Re: Language support for flexible handling of system-detected errors. Date: 29 Mar 2002 16:56:38 GMT Organization: IBM_Research Message-ID: References: NNTP-Posting-Host: starcraft.watson.ibm.com X-Trace: news.btv.ibm.com 1017420998 14562 9.2.8.37 (29 Mar 2002 16:56:38 GMT) X-Complaints-To: news@btv.ibm.com NNTP-Posting-Date: 29 Mar 2002 16:56:38 GMT Xref: archiver1.google.com comp.arch:26105 comp.lang.ada:21848 Date: 2002-03-29T16:56:38+00:00 List-Id: In article , Larry Kilgallen wrote: >In article , "Marin David Condic" writes: >> This is a fair objection and one I have leveled myself from time to time. It >> would be really nice to have a highly integrated IDE - as well as >> significant libraries to speed development with leverage. >> >> Here's a question that might help the Ada crowd out by getting a critic's >> perspective: If you want a spiffy IDE (and possibly libraries) what would it >> look like? (Its fair to point to some other language IDE and say "Something >> like that...", but it might be more helpful to describe some specific >> features.) > >Think Pascal (formerly Lightspeed Pascal ?). It features (featured?) > > incremental recompilation > debugging > >both right from the editor window. If it allowed me to do this from *my* editor's window it would be just wonderful. All that's needed for that is a well-defined interface that allows the compiler to read the editor's file buffer, and to report back to the editor what areas to highlight and what messages to display. Such an interface can be defined generically, so whether I use emacs or some other programmable editor wouldn't matter. It would also mean that the compiler developer would not have to re-invent what it takes to make a good program editor. Not only would it be different from other editors -- it would most likely have fewer general-purpose features (non-trivial undo, non-trivial macro support), though it might have certain nice language-specific features such as syntax-sensitive layout control. Many years ago I saw a programming-language-specific editor whose internal representation was the parse tree. This permitted a nice integration with language fatures, debugging and incremental recompilation. It may have been the one you mention (it was early 80s). One significant problem with this approach was that only syntactically complete program fragments could be cut or pasted from other programs or files, so one had to have a second editor to put the pieces together before one could import something! A general-purpose editor has no problems with arbitrary program text updates. The language-sensitive part of the environment, interfaced to the compiler, could highlight complete and incomplete syntactic groupings, but would not interfere with the incremental nature of composing a program. Michel.