Re: crash

Andy> At 02:35 PM 3/5/00 -0800, Kyle Jones wrote: Andy> Right, but I don't see how this can be achieved Andy> realistically. Every time you instantiate an image instance Andy> (and redisplay does this a lot) there is a chance that you Andy> will call lisp code. I guess I'm naive. It seems to me that redisplay should be given the capability to signal Lisp that it needs some function called, and wait on that function's completion. (Since XEmacs is single threaded, you can currently assume that it will, I suppose, but we probably want to reduce dependency on the assumption that XEmacs is single-threaded, too.) I don't know what exactly "wait" means in the single-threaded case, but I guess it would just return to Lisp at that point, which does the work, and then redisplay starts over. And the analogy to static images seems false, too, since tabs and progress bars are inherently dynamic; true, they don't get updated in real time, but they do get reinstantiated many times, multiplying the chance for a bug to manifest. Andy> I don't believe this is currently the case, but I'm prepared Andy> to be proved wrong. But if you are wrong but mostly right, these problems are going to be devilishly hard to debug, aren't they? Especially in a year when somebody decides to implement an oscilloscope in Lisp in the gutter and we've all pretty much forgotten this discussion.... I think you are violating the First Golden Law of Redisplay here. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."