From owner-cvs-all Sun May 21 14:47: 1 2000 Delivered-To: cvs-all@freebsd.org Received: from tomts3-srv.bellnexxia.net (tomts3.bellnexxia.net [209.226.175.141]) by hub.freebsd.org (Postfix) with ESMTP id 48C7937B54B; Sun, 21 May 2000 14:46:54 -0700 (PDT) (envelope-from hoek@FreeBSD.org) Received: from localhost.nowhere ([209.226.183.150]) by tomts3-srv.bellnexxia.net (InterMail vM.4.01.02.17 201-229-119) with ESMTP id <20000521214651.QBOK25676.tomts3-srv.bellnexxia.net@localhost.nowhere>; Sun, 21 May 2000 17:46:51 -0400 Received: (from tim@localhost) by localhost.nowhere (8.9.3/8.9.1) id RAA10471; Sun, 21 May 2000 17:46:27 -0400 (EDT) (envelope-from tim) Date: Sun, 21 May 2000 17:46:23 -0400 From: Tim Vanderhoek To: Brian Feldman Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/usr.bin/more prim.c Message-ID: <20000521174623.A10303@mad> References: <200005211835.LAA63624@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95i In-Reply-To: <200005211835.LAA63624@freefall.freebsd.org>; from Brian Feldman on Sun, May 21, 2000 at 11:35:28AM -0700 Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sun, May 21, 2000 at 11:35:28AM -0700, Brian Feldman wrote: > Log: > In the modern world, things are much faster than when more(1) was created. > Scrolling sideways is fast, and a "...skipping..." message making everything > blink does much more harm than good. I agree, but this commit kinda breaks reading short files when -e is specified. It probably also breaks tags and searches in some instances, but I didn't bother testing. It needs to be done differently. My plans, short-term, have been to keep the "...skipping..." for the sake of short files, and also modify more(1) to pretend -c was specified if the user redraws the screen with ^L. Under syscons I never see the "...skipping...". Does the message actually harm output on some terminals? If so, the correct fix is to add a heuristic to guess when the output will fill a full screen or not. Such a heuristic would have a very low failure rate. I am not the maintainer of more, but, I suggest that you either implement such a heuristic, encourage someone else to implement such a heuristic, or leave it with the previous behaviour. By "kinda breaks" I mean that you cannot see where the top of the file is without either a line-by-line analysis or, in some cases, probably a semantical analysis would be required to determine the top of the file. Actually, semantical analysis is always potentially necessary, since there's no reason "...skipping..." can't occur within the file being viewed, but this makes it more likely to be needed (think tags and searchs). -- Signature withheld by request of author. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message