public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: nobody@gcc.gnu.org
Cc: gcc-prs@gcc.gnu.org,
Subject: Re: libstdc++/8195: n-th algorithm (STL) doesn`t work properly
Date: Fri, 11 Oct 2002 08:56:00 -0000	[thread overview]
Message-ID: <20021011155602.2584.qmail@sources.redhat.com> (raw)

The following reply was made to PR libstdc++/8195; it has been noted by GNATS.

From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: "Belov, Eugeny" <Eugeny_Belov@stl.sarov.ru>
Cc: gcc-bugs@gcc.gnu.org, <gcc-gnats@gcc.gnu.org>
Subject: Re: libstdc++/8195: n-th algorithm (STL) doesn`t work properly
Date: Fri, 11 Oct 2002 10:50:51 -0500 (CDT)

 >   I am understanding that the principle of this algorithm (in simple 
 > terms) is to divide the sequence into 2 parts like this:
 > smaller elements , n-th element, bigger elements,  and in the general 
 > case both parts can be unsorted.
 
 Correct.
 
 >   But the real problem in this case is that smaller elements comes after 
 > n-th ("n-th" is the integer with value 12 - the biggiest of all other 
 > elements), i.e. the sequence after using the nth_element() is "4 4 5 6 6 
 > 6 9 _12_ 8 10" and in this case I think the n-th algorithm 
 > implementation make a mistake.
 
 Why? The signature of the function is
   void nth_element(RandomAccessIterator first, 
                    RandomAccessIterator nth,
                    RandomAccessIterator last);
 and you call it like
   nth_element(a, a+2, a+10);
 so the result is that elements 0 and 1 will be smaller than the one at 
 position 2, and that elements 3 through 9 will be larger. This is what 
 happens in your output.
 
 Regards
   Wolfgang
 
 PS-1: Note that a+2 points to "9" in your example, not "12" as you claim, 
 since indices are zero-based.
 
 PS-2: The algorithm sorts such that the values before and after the 
 _pointer_ are smaller and larger, respectively, not those elements that 
 are smaller and larger than the _element pointed to_! I think this is the 
 basic misunderstanding on your behalf. The standard on this reads
 
   After nth_element the element in the position pointed to by nth is the
   element that would be in that position if the whole range were sorted.
   Also  for  any iterator i in the range [first, nth) and any iterator j
   in the range [nth, last) it holds that: !(*i > *j) or comp(*j, *i)  ==
   false.
 
 -------------------------------------------------------------------------
 Wolfgang Bangerth              email:           bangerth@ticam.utexas.edu
                                www: http://www.ticam.utexas.edu/~bangerth
 
 


             reply	other threads:[~2002-10-11 15:56 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-10-11  8:56 Wolfgang Bangerth [this message]
  -- strict thread matches above, loose matches on Subject: below --
2002-10-11 16:12 paolo
2002-10-11  8:46 Belov, Eugeny
2002-10-11  7:46 Wolfgang Bangerth
2002-10-11  4:16 johnb

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20021011155602.2584.qmail@sources.redhat.com \
    --to=bangerth@ticam.utexas.edu \
    --cc=gcc-prs@gcc.gnu.org \
    --cc=nobody@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).