bug-texinfo
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [platform-testers] texinfo-6.0.90 pretest


From: Gavin Smith
Subject: Re: [platform-testers] texinfo-6.0.90 pretest
Date: Tue, 22 Dec 2015 19:25:56 +0000

On 22 December 2015 at 15:57, Nelson H. F. Beebe <address@hidden> wrote:
> All of the others seem to be failures like this:
>
> Can't call mro_method_changed_in() on anonymous symbol table at 
> ../tp/Texinfo/Convert/Paragraph.pm line 253.
> Compilation failed in require at ../tp/Texinfo/Convert/Plaintext.pm line 29.
> BEGIN failed--compilation aborted at ../tp/Texinfo/Convert/Plaintext.pm line 
> 29.
> Compilation failed in require at ../tp/Texinfo/Convert/Info.pm line 25.
> BEGIN failed--compilation aborted at ../tp/Texinfo/Convert/Info.pm line 25.
> Compilation failed in require at (eval 13) line 1.
> Makefile:1305: recipe for target 'texinfo.info' failed
>
> The failures occur on a mixture of platforms, including several
> GNU/Linux distributions, Solaris, FreeBSD 9 (but 10 and 11 succeed),
> and Mac OS X.

Thanks for running the tests. If it's the same error everywhere,
that's somewhat promising as it may only have to be fixed once. I'll
look at all the logs in detail later.

I found the following change that was made to the Perl interpreter:

http://grokbase.com/t/perl/perl5-changes/11apf1zrff/perl-git-branch-blead-updated-v5-15-4-12-g03d9f02

There's a lot in there about the Perl internals which we shouldn't
have to care about, but here's the relevant part:

======
This stops PL_curstash from pointing to a freed-and-reused scalar in
cases like ‘package Foo; BEGIN {*Foo:: = *Bar::}’.

In such cases, another BEGIN block, or any subroutine definition,
would cause a crash. Now it just happily proceeds.
======

This assignment of globs was exactly what was done in
Texinfo::Convert::Paragraph. It suggests that it might be alright if
the BEGIN block wasn't followed with anything, so I tried commenting
out the (empty) function that followed, and it worked. This was on
NetBSD 5.1; let's hope it fixes it for the other ones too.

(Like this:

# Used for debugging.  Not implemented.
##   sub dump($)
##   {
##   return "\n";
##   }



reply via email to

[Prev in Thread] Current Thread [Next in Thread]