Path: csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!aioe.org!feeder.news-service.com!feeds.phibee-telecom.net!talisker.lacave.net!lacave.net!not-for-mail From: Su Zhang Newsgroups: comp.lang.ruby Subject: Re: Exclusive float range, Range#step result in counterintuitive result Date: Thu, 14 Apr 2011 13:01:42 -0500 Organization: Service de news de lacave.net Lines: 40 Message-ID: References: NNTP-Posting-Host: bristol.highgroove.com Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Trace: talisker.lacave.net 1302804203 68837 65.111.164.187 (14 Apr 2011 18:03:23 GMT) X-Complaints-To: abuse@lacave.net NNTP-Posting-Date: Thu, 14 Apr 2011 18:03:23 +0000 (UTC) In-Reply-To: X-Received-From: This message has been automatically forwarded from the ruby-talk mailing list by a gateway at comp.lang.ruby. If it is SPAM, it did not originate at comp.lang.ruby. Please report the original sender, and not us. Thanks! For more details about this gateway, please visit: http://blog.grayproductions.net/categories/the_gateway X-Mail-Count: 381551 X-Ml-Name: ruby-talk X-Rubymirror: Yes X-Ruby-Talk: Xref: x330-a1.tempe.blueboxinc.net comp.lang.ruby:2866 This looks like a bug to me. (The problem exists as of 1.9.2-p180.) In , `range_step' is the implementation of Range#step. It calls `ruby_float_step', which is implemented in , to handle float stepping. The call looks like this: else if (ruby_float_step(b, e, step, EXCL(range))) { /* done */ } The problem is that it passes the `exclude_end?' property of the range to `ruby_float_step', in our case it is false. Inside the implementation of `ruby_float_step' is something like: n = floor(n + err); if (!excl) n++; for (i=0; i [1.0, 2.0, 3.0, 4.0, 5.0, 6.0] This works since `num_step', again implemented in , explicitly passes a false to excl: else if (!ruby_float_step(from, to, step, FALSE)) { ... -- Posted via http://www.ruby-forum.com/.