I alluded to this in a panel discussion at APL'91 on parallelism in APL. I was making the argument that parallelism raises the level of complexity by introducing possibilities of interaction absent from serial processing. The example I gave was doing a search and replace on "AA" by "B" when the target includes "AAA" - do we give the result of "BA" or "BB"? The latter is a possibility - and may be desired - if a parallel search finds the two matches starting at the first and second positions in "AAA", e.g. as in
'AA' E. 'AAA' 1 1 0 On Fri, Jan 15, 2010 at 11:00 PM, <[email protected]> wrote: > ... > 3. Re: bug in regexp support distributed with J (Oleg Kobchenko) > ... > Message: 3 > Date: Fri, 15 Jan 2010 13:27:45 -0800 (PST) > From: Oleg Kobchenko <[email protected]> > Subject: Re: [Jbeta] bug in regexp support distributed with J > To: Beta forum <[email protected]> > Message-ID: <[email protected]> > Content-Type: text/plain; charset=us-ascii > > ... > Is there any theory or references or code samples of > "overlapped" matches? > ... > > -- Devon McCormick, CFA ^me^ at acm. org is my preferred e-mail ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
