Skip Menu |

This queue is for tickets about the Math-Pari CPAN distribution.

Report information
The Basics
Id: 61271
Status: open
Priority: 0/
Queue: Math-Pari

People
Owner: Nobody in particular
Requestors: zefram [...] fysh.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



Subject: PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sat, 11 Sep 2010 16:35:16 +0100
To: bug-Math-Pari [...] rt.cpan.org
From: Zefram <zefram [...] fysh.org>
The effect of the PERL_POLLUTE macro, which Pari.xs defines, has been removed from Perl 5.13.4. As a result, Math::Pari no longer compiles. I suggest that you switch to the 5.6 symbol names. -zefram
CC: undisclosed-recipients: ;
Subject: Re: [rt.cpan.org #61271] PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sat, 11 Sep 2010 13:12:58 -0700
To: Zefram via RT <bug-Math-Pari [...] rt.cpan.org>
From: Ilya Zakharevich <nospam-abuse [...] ilyaz.org>
On Sat, Sep 11, 2010 at 11:35:31AM -0400, Zefram via RT wrote: Show quoted text
> The effect of the PERL_POLLUTE macro, which Pari.xs defines, has been > removed from Perl 5.13.4. As a result, Math::Pari no longer compiles.
What are the messages? Thanks, Ilya
Subject: Re: [rt.cpan.org #61271] PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sat, 11 Sep 2010 21:18:25 +0100
To: Ilya Zakharevich via RT <bug-Math-Pari [...] rt.cpan.org>
From: Zefram <zefram [...] fysh.org>
Ilya Zakharevich via RT wrote: Show quoted text
>What are the messages?
/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -DVERSION=\"2.01080604\" -DXS_VERSION=\"2.01080604\" -fPIC "-I/home/zefram/usr/perl/perl_install/perl-5.13.4-i32-f52/lib/5.13.4/i386-linux-thread-multi/CORE" -DPARI_VERSION_EXP=2001007 -DDEBUG_PARI -DUSE_SLOW_ARRAY_ACCESS -DUSE_SLOW_NARGS_ACCESS -DLSB_in_U32=0 -Derr=pari_err -DGCC_INLINE Pari.c Pari.xs: In function 'findVariable': Pari.xs:567: error: 'na' undeclared (first use in this function) Pari.xs:567: error: (Each undeclared identifier is reported only once Pari.xs:567: error: for each function it appears in.) Pari.xs: In function 'sv2pari': Pari.xs:850: error: 'na' undeclared (first use in this function) Pari.xs: In function 'installPerlFunctionCV': Pari.xs:1228: error: 'na' undeclared (first use in this function) Pari.xs: In function 'fill_argvect': Pari.xs:1581: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface22': Pari.c:3354: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface27': Pari.c:3532: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface28': Pari.c:3579: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface37': Pari.c:3881: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface47': Pari.c:3927: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface48': Pari.c:3973: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface49': Pari.c:4028: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface83': Pari.c:4073: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface84': Pari.c:4111: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface73': Pari.c:4329: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface86': Pari.c:4380: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_interface87': Pari.c:4419: error: 'na' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari__2bool': Pari.c:4466: error: 'sv_yes' undeclared (first use in this function) Pari.c:4466: error: 'sv_no' undeclared (first use in this function) Pari.c: In function 'XS_Math__Pari_pari2bool': Pari.c:4494: error: 'sv_yes' undeclared (first use in this function) Pari.c:4494: error: 'sv_no' undeclared (first use in this function) Pari.xs: In function 'XS_Math__Pari_loadPari': Pari.xs:3594: error: 'na' undeclared (first use in this function) make: *** [Pari.o] Error 1 -zefram
CC: undisclosed-recipients: ;
Subject: Re: [rt.cpan.org #61271] PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sat, 11 Sep 2010 14:31:55 -0700
To: Zefram via RT <bug-Math-Pari [...] rt.cpan.org>
From: Ilya Zakharevich <nospam-abuse [...] ilyaz.org>
On Sat, Sep 11, 2010 at 11:35:31AM -0400, Zefram via RT wrote: Show quoted text
> Sat Sep 11 11:35:29 2010: Request 61271 was acted upon. > Transaction: Ticket created by zefram@fysh.org > Queue: Math-Pari > Subject: PERL_POLLUTE breakage of Math::Pari on 5.13.4 > Broken in: (no value) > Severity: (no value) > Owner: Nobody > Requestors: zefram@fysh.org > Status: new > Ticket <URL: http://rt.cpan.org/Ticket/Display.html?id=61271 > > > > The effect of the PERL_POLLUTE macro, which Pari.xs defines, has been > removed from Perl 5.13.4.
And now, when p5p in its infinite wisdom removed these (automatically generated) defines, every extension would need to add them back by hand. Well, would one expect anything else? Show quoted text
> As a result, Math::Pari no longer compiles.
The following patch makes things work without PERL_POLLUTE... Thanks, Ilya --- Pari.xs-pre Wed Mar 3 13:53:32 2010 +++ Pari.xs Sat Sep 11 14:21:38 2010 @@ -34,6 +34,12 @@ extern "C" { #ifdef __cplusplus } #endif + +#if !defined(na) && defined(PERL_VERSION) && (PERL_VERSION > 7) /* Added in 6 (???), Removed in 13 */ +# define na PL_na +# define sv_no PL_sv_no +# define sv_yes PL_sv_yes +#endif #if PARI_VERSION_EXP < 2002012 void init_defaults(int force); /* Probably, will never be fixed in 2.1.* */
Subject: Re: [rt.cpan.org #61271] PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sat, 11 Sep 2010 22:41:24 +0100
To: Ilya Zakharevich via RT <bug-Math-Pari [...] rt.cpan.org>
From: Zefram <zefram [...] fysh.org>
Ilya Zakharevich via RT wrote: Show quoted text
> every extension would need to add them back by >hand.
The thinking is that compatibility of XS to 5.005 is no longer any practical use, so no one should still be using the old names. Math::Pari is the only module affected of all the ones I test. By all means petition perl5-porters for PERL_POLLUTE to be reinstated, if you have sufficient ire. -zefram
On 2010-09-11 17:41:37, zefram@fysh.org wrote: Show quoted text
> Ilya Zakharevich via RT wrote:
> > every extension would need to add them back by > >hand.
> > The thinking is that compatibility of XS to 5.005 is no longer > any practical use, so no one should still be using the old names. > Math::Pari is the only module affected of all the ones I test. By all > means petition perl5-porters for PERL_POLLUTE to be reinstated, if you > have sufficient ire. >
I just want to remind that perl 5.14.0 will happen in a month or so, so probably the Pari.xs patch should be applied and a new release be done? Regards, Slaven
CC: undisclosed-recipients: ;
Subject: Re: [rt.cpan.org #61271] PERL_POLLUTE breakage of Math::Pari on 5.13.4
Date: Sun, 27 Mar 2011 07:44:10 -0700
To: Slaven_Rezic via RT <bug-Math-Pari [...] rt.cpan.org>
From: Ilya Zakharevich <nospam-abuse [...] ilyaz.org>
On Thu, Mar 24, 2011 at 03:37:45AM -0400, Slaven_Rezic via RT wrote: Show quoted text
> Queue: Math-Pari > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=61271 > > > On 2010-09-11 17:41:37, zefram@fysh.org wrote:
> > Ilya Zakharevich via RT wrote:
> > > every extension would need to add them back by > > >hand.
> > > > The thinking is that compatibility of XS to 5.005 is no longer > > any practical use, so no one should still be using the old names. > > Math::Pari is the only module affected of all the ones I test. By all > > means petition perl5-porters for PERL_POLLUTE to be reinstated, if you > > have sufficient ire. > >
> > I just want to remind that perl 5.14.0 will happen in a month or so, so > probably the Pari.xs patch should be applied and a new release be done?
I really appreciate your prodding me in this regard, thanks! However, given the annoyance factor of this particular edit, I do not know when I would be able to find time to touch it... If I knew a reliable indicator when a module must do POLLUTE definitions itself... Drat, Ilya