Skip Menu |

This queue is for tickets about the Text-FIGlet CPAN distribution.

Report information
The Basics
Id: 122198
Status: resolved
Priority: 0/
Queue: Text-FIGlet
People
Owner: Nobody in particular
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:
Bug Information
Severity: (no value)
Broken in: 2.19.3
Fixed in: 2.19.4

History Show all quoted text
  Sat Jun 24 03:31:03 2017 SREZIC [...] cpan.org - Ticket created
Subject: Use of strings with code points over 0xFF ... is not allowed
perl 5.27.1 introduced a new fatal deprecation which is causing the test suite to fail: ... Use of strings with code points over 0xFF as arguments to 1's complement (~) operator is not allowed at t/2-extended.t line 59, <GEN4> line 456. # Looks like your test exited with 255 just after 3. t/2-extended.t .. Dubious, test returned 255 (wstat 65280, 0xff00) Failed 4/7 subtests Use of strings with code points over 0xFF as arguments to 1's complement (~) operator is not allowed at t/3-ransom.t line 58. # Looks like your test exited with 9 just after 2. t/3-ransom.t .... Dubious, test returned 9 (wstat 2304, 0x900) Failed 2/4 subtests ...
  Wed Aug 23 21:10:26 2017 jpierce [...] cpan.org - Correspondence added
It's a shame p5p did that ten years after adopting Unicode, I don't see the benefit. I cannot test it (or some other recent CPANTS failures) either since they still haven't fixed 5.26 to build on older systems https://blogs.perl.org/users/yuki_kimoto/2017/05/i-cant-install-perl-5260-rc1-in-centos-5.html I'll look into it once I can get a working test perl. On Sat Jun 24 03:31:03 2017, SREZIC wrote: Show quoted text
> perl 5.27.1 introduced a new fatal deprecation which is causing the > test suite to fail: > > ... > Use of strings with code points over 0xFF as arguments to 1's > complement (~) operator is not allowed at t/2-extended.t line 59, > <GEN4> line 456. > # Looks like your test exited with 255 just after 3. > t/2-extended.t .. > Dubious, test returned 255 (wstat 65280, 0xff00) > Failed 4/7 subtests > Use of strings with code points over 0xFF as arguments to 1's > complement (~) operator is not allowed at t/3-ransom.t line 58. > # Looks like your test exited with 9 just after 2. > t/3-ransom.t .... > Dubious, test returned 9 (wstat 2304, 0x900) > Failed 2/4 subtests > ...
  Wed Aug 23 21:10:26 2017 The RT System itself - Status changed from 'new' to 'open'
  Wed Aug 23 21:38:04 2017 jpierce [...] cpan.org - Correspondence added
Ahh, found some details here: https://groups.google.com/forum/#!topic/perl.perl5.changes/Sxw_tcHvv7c Although still seems unwarranted. In particular, it breaks things like this where bitwise negation is not the goal but instead ~~ is used as an idiom for forcing scalar context. Perhaps they could add a new ~~ operator at the same time as the deprecation... On Wed Aug 23 21:10:26 2017, JPIERCE wrote: Show quoted text
> It's a shame p5p did that ten years after adopting Unicode, I don't > see the benefit. I cannot test it (or some other recent CPANTS > failures) either since they still haven't fixed 5.26 to build on older > systems > > https://blogs.perl.org/users/yuki_kimoto/2017/05/i-cant-install-perl- > 5260-rc1-in-centos-5.html > > I'll look into it once I can get a working test perl. > > On Sat Jun 24 03:31:03 2017, SREZIC wrote:
> > perl 5.27.1 introduced a new fatal deprecation which is causing the > > test suite to fail: > > > > ... > > Use of strings with code points over 0xFF as arguments to 1's > > complement (~) operator is not allowed at t/2-extended.t line 59, > > <GEN4> line 456. > > # Looks like your test exited with 255 just after 3. > > t/2-extended.t .. > > Dubious, test returned 255 (wstat 65280, 0xff00) > > Failed 4/7 subtests > > Use of strings with code points over 0xFF as arguments to 1's > > complement (~) operator is not allowed at t/3-ransom.t line 58. > > # Looks like your test exited with 9 just after 2. > > t/3-ransom.t .... > > Dubious, test returned 9 (wstat 2304, 0x900) > > Failed 2/4 subtests > > ...
  Sat Nov 11 10:44:26 2017 SREZIC [...] cpan.org - Correspondence added
I see that you're mixing usages of scalar() and ~~ in the test scripts. So maybe it's not too bad if you just stick to scalar() here. As for p5p breaking backward compatibility --- maybe you should speak up and ask p5p about this issue? Regards, Slaven On 2017-08-23 21:38:04, JPIERCE wrote: Show quoted text
> Ahh, found some details here: > > https://groups.google.com/forum/#!topic/perl.perl5.changes/Sxw_tcHvv7c > > Although still seems unwarranted. In particular, it breaks things like > this where bitwise negation is not the goal but instead ~~ is used as > an idiom for forcing scalar context. Perhaps they could add a new ~~ > operator at the same time as the deprecation... > > On Wed Aug 23 21:10:26 2017, JPIERCE wrote:
> > It's a shame p5p did that ten years after adopting Unicode, I don't > > see the benefit. I cannot test it (or some other recent CPANTS > > failures) either since they still haven't fixed 5.26 to build on > > older > > systems > > > > https://blogs.perl.org/users/yuki_kimoto/2017/05/i-cant-install-perl- > > 5260-rc1-in-centos-5.html > > > > I'll look into it once I can get a working test perl. > > > > On Sat Jun 24 03:31:03 2017, SREZIC wrote:
> > > perl 5.27.1 introduced a new fatal deprecation which is causing the > > > test suite to fail: > > > > > > ... > > > Use of strings with code points over 0xFF as arguments to 1's > > > complement (~) operator is not allowed at t/2-extended.t line 59, > > > <GEN4> line 456. > > > # Looks like your test exited with 255 just after 3. > > > t/2-extended.t .. > > > Dubious, test returned 255 (wstat 65280, 0xff00) > > > Failed 4/7 subtests > > > Use of strings with code points over 0xFF as arguments to 1's > > > complement (~) operator is not allowed at t/3-ransom.t line 58. > > > # Looks like your test exited with 9 just after 2. > > > t/3-ransom.t .... > > > Dubious, test returned 9 (wstat 2304, 0x900) > > > Failed 2/4 subtests > > > ...
  Sat Nov 11 13:58:15 2017 belg4mit [...] pthbb.org - Correspondence added
Subject: Re: [rt.cpan.org #122198] Use of strings with code points over 0xFF ... is not allowed
Date: Sat, 11 Nov 2017 13:27:06 -0500
To: bug-Text-FIGlet [...] rt.cpan.org
From: Jerrad Pierce <belg4mit [...] pthbb.org>
Yes, somestimes I type one, sometimes the other. Double unary negation (~~) is an old and long-standing trick for terse scalarification. I'll consider bringing it to p5p, but I've not had much luck before, and it's such a high volume list that I am reluctant to subscribe.
  Thu Nov 30 11:55:03 2017 $_ = 'spro^^*%*^6ut# [...] &$%*c>#!^!#&!pan.org'; y/a-z.@//cd; print - Correspondence added
On Sat Nov 11 13:58:15 2017, belg4mit@pthbb.org wrote: Show quoted text
> Yes, somestimes I type one, sometimes the other. Double unary negation (~~) > is an old and long-standing trick for terse scalarification. > > I'll consider bringing it to p5p, but I've not had much luck before, > and it's such a high volume list that I am reluctant to subscribe.
I brought it up in this ticket: https://rt.perl.org/Ticket/Display.html?id=131594 The response I got from Zefram was as follows: Show quoted text
> No, it's not so sensible that it would be worth preserving. It doesn't > actually work, even on those Perls where it doesn't error (take > ~~"\x{ffffffffffffff80}" or ~~"\x{ffffff80}", depending on your word > size). Stringification and scalar context are both operations that we > provide proper operators for. Double complementation is an excessively > cute hack, and the semantics on which it depends are being broken for > good reason.
  Sat Sep 07 16:58:16 2019 jpierce [...] cpan.org - Status changed from 'open' to 'resolved'
  Sat Sep 07 16:58:16 2019 jpierce [...] cpan.org - Fixed in 2.19.4 added