220 12252 <35c8da4b-fb3e-4d79-8344-e414b6872554@isocpp.org> article Path: news.gmane.org!not-for-mail From: Douglas Boffey Newsgroups: gmane.comp.lang.c++.isocpp.proposals Subject: Re: Fwd: Study on usage of break label / continue label Date: Wed, 13 Aug 2014 00:40:53 -0700 (PDT) Lines: 294 Approved: news@gmane.org Message-ID: <35c8da4b-fb3e-4d79-8344-e414b6872554@isocpp.org> References: <25d71a9d-2112-4b70-9103-af95731ed766@isocpp.org> Reply-To: std-proposals@isocpp.org NNTP-Posting-Host: plane.gmane.org Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_2925_1600440311.1407915653301" X-Trace: ger.gmane.org 1407915661 29833 80.91.229.3 (13 Aug 2014 07:41:01 GMT) X-Complaints-To: usenet@ger.gmane.org NNTP-Posting-Date: Wed, 13 Aug 2014 07:41:01 +0000 (UTC) Cc: inkwizytoryankes@gmail.com To: std-proposals@isocpp.org Original-X-From: std-proposals+bncBDSIZVXVVUFBBBNNVSPQKGQETIJ6WNI@isocpp.org Wed Aug 13 09:40:56 2014 Return-path: Envelope-to: gclcip-std-proposals@m.gmane.org Original-Received: from mail-oa0-f72.google.com ([209.85.219.72]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1XHTB1-0001x0-CC for gclcip-std-proposals@m.gmane.org; Wed, 13 Aug 2014 09:40:55 +0200 Original-Received: by mail-oa0-f72.google.com with SMTP id m1sf51595919oag.3 for ; Wed, 13 Aug 2014 00:40:54 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:cc:message-id:in-reply-to:references:subject :mime-version:x-original-sender:reply-to:precedence:mailing-list :list-id:list-post:list-help:list-archive:list-subscribe :list-unsubscribe:content-type; bh=3ovS3KnQ7chO2ro92H+bz1ffDaI5Tc0tGrcxl00lMJY=; b=uqUxq1E6P/5M9wdq7e4OwqgO2njO9c0Xn+z2PWDoqFPz4IJ3UF+e3+lxqASGloL84s +Bz9JL3+poSiaW2PHm0CCQKsapOeqPFc0H39dui0p0bbDSdCM+VPylNEBqdP83OkrC4n D+rrWuz01wokS+GnuuSaKXQ6EerNed/UBDXmo+h+FgUPYbKA2jpP5wPCssRqKEL7MMIG VDs4VQnG5WJsajMZcEun5X7kjy3kIKho5frsZqTzYH3GzUlNk7NfMrohHYhXaAt+WMsH /1Jb4zqu5PIHtySqIrc+4zzmB9jE2UQhwke0PVHq7rfn7UkO5cGdjEbUM7pZoeui2XtM 1aag== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:date:from:to:cc:message-id:in-reply-to :references:subject:mime-version:x-original-sender:reply-to :precedence:mailing-list:list-id:list-post:list-help:list-archive :list-subscribe:list-unsubscribe:content-type; bh=3ovS3KnQ7chO2ro92H+bz1ffDaI5Tc0tGrcxl00lMJY=; b=d9ZWEeJ1GjN0AyBlBqzDJgxhrC7xLSmnyw8L7oPxQ1WW5XNtX+04Vvx7Z0v7nwlIIM 8C7DVnj5D9cNWdQ/y0qpULTXnLefCBScAkWJbb/lRoI8S+Y/14OSJ+O/iDwjwHkWHji9 yYvwOhhJL0ROfhw5nFHMHwGw34wE3ypu1NnSiVqR9Zh4As4Amoy/cM5L48o5JvuQH8Sj DhBiY2MkxnjMNH7UV9eQChMxMBD4ETRZ6IP1cf+lRfLz9Zkx4Te+Rulr960Jx2Cs9i+N x7XKF0QrN9nRmThLL1WGbwpryw+DoH8g1pHFdyXcxy6ZX4IRn18wxFYxdzeiNxWWZG/P ydxQ== X-Gm-Message-State: ALoCoQln+W2qu1KC/dMFkd7Uvyu6+e5vAZCDg6UzeJIzUzZy3P6gSMBCfS2qjspLCItO7TUylsfe X-Received: by 10.42.26.12 with SMTP id d12mr1380514icc.24.1407915654435; Wed, 13 Aug 2014 00:40:54 -0700 (PDT) X-BeenThere: std-proposals@isocpp.org Original-Received: by 10.140.41.211 with SMTP id z77ls483451qgz.54.gmail; Wed, 13 Aug 2014 00:40:53 -0700 (PDT) X-Received: by 10.140.20.151 with SMTP id 23mr16435qgj.13.1407915653658; Wed, 13 Aug 2014 00:40:53 -0700 (PDT) In-Reply-To: <25d71a9d-2112-4b70-9103-af95731ed766@isocpp.org> X-Original-Sender: Douglas.Boffey@gmail.com Precedence: list Mailing-list: list std-proposals@isocpp.org; contact std-proposals+owners@isocpp.org List-ID: X-Google-Group-Id: 399137483710 List-Post: , List-Help: , List-Archive: List-Subscribe: , List-Unsubscribe: , Xref: news.gmane.org gmane.comp.lang.c++.isocpp.proposals:12252 Archived-At: ------=_Part_2925_1600440311.1407915653301 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Friday, 8 August 2014 18:19:33 UTC+1, inkwizyt...@gmail.com wrote: > > Some time ago was proposition to allow "chain" breaks. > Something like that: > for(...) > { > while(...) > { > if(...) > break break; //break while and for > else if(...) > break continue; //break while but continue for > } > } > But I think that label version is better. Mainly because is more secure= =20 > when you add or remove nested loops. > Label break could detect situation when you move big block of code to=20 > outside of loop its belongs. > Chained breaks will accept new position as long nested level is deep=20 > enough even if new context is completely different. > > Hmm, so you need to count the number of breaks and continues, and match= =20 them to the nesting level? Does that help produce more legible code? I would prefer some sort of named control statements, e.g. for [foo] (=E2=80=A6) { // =E2=80=A6 while [bar] (=E2=80=A6) { // =E2=80=A6 if (baz) break foo; // =E2=80=A6 } // =E2=80=A6 } =20 > > On Friday, August 8, 2014 3:20:55 AM UTC+2, Zhihao Yuan wrote: >> >> On Thu, Aug 7, 2014 at 8:03 PM, wrote: >> >>> >>> case ...: >>>> break loop: >>>> ... >>>> continue loop: // `loop` is a contextual keyword >>>> >>> >>> That's a simple creative idea I hadn't thought of but no doubt the rest= =20 >>> of the world did. >>> >> >> If you read those code, you are going to immediate have this >> idea. Because in the code, I keep seeing >> >> loop: for/while >> ... =20 >> switch (...) { >> ... >> break loop; // and even `continue loop;` >> >> And I feel "What's a pattern!", but later, in (B), I see >> >> lab07: ... >> switch ... >> break lab07; >> ... >> lab08: ... >> break lab08; >> >> Then I realized that the pattern above doesn't always work, >> because unlike variable names, label names are not scoped, >> so that you can't have two `loop` in one function. >> >> Then I say "OK, let's forget about the label name..." :) >> >> Can C be broken down more, does your analysis invite "break switch" for= =20 >>> when you are in a loop in a case statement and want out? >>> >>> >> I did not see a use complex like that, but one of the two >> undetermined samples might be the case. >> >> However, it's a interesting question that a`break loop` >> inside a loop inside a `switch` inside a loop should break >> which loop. I think it should still break the outer one. That is: >> >> for (...) { // loop1 >> switch (...) { >> case ...: >> while (...) { // loop2 >> break; // get out of loop2, continue switching >> break loop; // get out of loop1 >> } >> } >> } >> >> --=20 >> Zhihao Yuan, ID lichray >> The best way to predict the future is to invent it. >> ___________________________________________________ >> 4BSD -- http://bit.ly/blog4bsd=20 >> > --=20 ---=20 You received this message because you are subscribed to the Google Groups "= ISO C++ Standard - Future Proposals" group. To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposals+unsubscribe@isocpp.org. To post to this group, send email to std-proposals@isocpp.org. Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposa= ls/. ------=_Part_2925_1600440311.1407915653301 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


On Friday, 8 August 2014 18:19:33 UTC+1, inkwizyt.= ...@gmail.com wrote:
Some time ago was proposition to allow "chain" breaks.
Somethin= g like that:
for(...)
{
    while(...)
    {
        if(...)
            = break <= span style=3D"color:#008">break; //break while a= nd for
        else if(...)
            = break <= /span>continue= ; //bre= ak while but continue for
   = ; }
= }
But I think that label version is better. Mainly be= cause is more secure when you add or remove nested loops.
Label break co= uld detect situation when you move big block of code to outside of loop its= belongs.
Chained breaks will accept new position as long nested level i= s deep enough even if new context is completely different.

Hmm, so you need to count the number of breaks and continues= , and match them to the nesting level?  Does that help produce more le= gible code?

I would prefer some sort of named control statements, e.= g.

for [foo] (=E2=80=A6) {<= /span>
  = // =E2=80= =A6
&nb= sp; while<= /span> [bar] (=E2=80=A6) {
    // =E2=80=A6
    if (b= az)=
   =   br= eak foo;
    &nbs= p; // =E2= =80=A6
=   }
  <= /span>// =E2=80= =A6
} 

On Friday, August 8, 2014 3:20:55 AM UTC+2, Zhihao Yuan wrote:
On Thu, Aug 7, 2014 at 8:= 03 PM, <gmis...@gmail.com> wrote:

  cas= e ...:
    break loop:
    ...
    continue loop:  // `loop` is a contextual keyword

That's a simple creati= ve idea I hadn't thought of but no doubt the rest of the world did.

If you read those code, you are going to immediate have= this
idea.  Because in the code, I keep seeing

<= /div>
  loop: for/while
  ... 
 &= nbsp;  switch (...) {
    ...
      break loop;  // and even `co= ntinue loop;`

And I feel "What's a pattern!", but later, = in (B), I see

  lab07: ...
  = ;  switch ...
      break lab07;
  ...
&nb= sp; lab08: ...
      break lab08;
Then I realized that the pattern above doesn't always work,=
because unlike variable names, label names are not scoped,
so that you can't have two `loop` in one function.

=
Then I say "OK, let's forget about the label name..." :)

Can C be broken down more, does your analysis in= vite "break switch" for when you are in a  loop in a case st= atement and want out?


I did not see a use complex l= ike that, but one of the two
undetermined samples might be th= e case.

However, it's a interesting question that a`break loop`
inside a loop in= side a `switch` inside a loop should break
which loop.  I think it = should still break the outer one.  That is:

  for (...) {  // loop1
    switch (= ....) {
    case ...:
  = ;    while (...) {  // loop2
        break;    // get = out of loop2, continue switching
    &nbs= p;   break loop;    // get out of loop1
 &= nbsp;    }
    }
  }

--
Zhihao Yuan, ID lichray
The best way to predict the future is to invent it.
____________________= _______________________________
4BSD -- http://bit.ly/blo= g4bsd

--

---
You received this message because you are subscribed to the Google Groups &= quot;ISO C++ Standard - Future Proposals" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to std-proposa= ls+unsubscribe@isocpp.org.
To post to this group, send email to std-proposals@isocpp.org.
Visit this group at http://groups.google.com/a/isocpp.org/group/std-proposals/<= /a>.
------=_Part_2925_1600440311.1407915653301-- .