Skip Menu |

This queue is for tickets about the Any-Moose CPAN distribution.

Report information
The Basics
Id: 120203
Status: rejected
Priority: 0/
Queue: Any-Moose

People
Owner: Nobody in particular
Requestors: DGHOLZ [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: Normal
Broken in: 0.27
Fixed in: (no value)



Subject: Please make the latest version of Any::Moose not warn about deprecation
Hello, Please release version 0.28 without a deprecation warning (and optionally release 0.29 with the warning) so that users who want to use the fixes released in 0.27 can do that without disabling deprecation warnings. As I understand the plans for the future of Any::Moose, the intent is not to do any more releases, maintenance, or support for the module & to point users to Moo as an actively-supported superior alternative. I see that version 0.27 now causes a deprecation warning to be shown, echoing that recommendation. To me, deprecation warnings indicate that there is some functionality which is going to be removed in a future release. I have trouble matching that up with the plans for Any::Moose, as I think it won't have another release removing features, and it will continue to be available for installation from CPAN. Could the readme & documentation be changed to declare that Any::Moose has reached its End Of Life, and that no work will be done on it in the future, and that Moo is the recommended actively-maintained alternative? Users of Any::Moose could then continue using it (if they were satisfied with its features), and naturally migrate to Moo as they outgrew A::M. I'm a downstream user of Any::Moose (via Method::Signatures), so I would have to apply the workaround of "no warnings 'deprecated';" to each of my modules using M::S (until they release a version doing the same); I'm hesitant to do this since it will mask deprecation warnings from M::S. I'm planning to pin my code to use 0.26 of A::M, and review the incidental changes made for 0.27 to see if I'm missing out on anything. Daniel
On 2017-02-11 03:00:16, DGHOLZ wrote: Show quoted text
> Hello, > > Please release version 0.28 without a deprecation warning (and > optionally release 0.29 with the warning) so that users who want to > use the fixes released in 0.27 can do that without disabling > deprecation warnings.
Can you clarify which changes in 0.27 you were looking for? The other changes in 0.27 were all tooling or documentation-related and mostly benefit cpan clients and static analyzers.
On Mon Feb 13 17:33:12 2017, ETHER wrote: Show quoted text
> Can you clarify which changes in 0.27 you were looking for? The other > changes in 0.27 were all tooling or documentation-related and mostly > benefit cpan clients and static analyzers.
Thank you for your summary of the other changes in 0.27. That helps me feel confident that I will be OK pinning Any::Moose to 0.26 for my project. I'm not a direct consumer of Any::Moose, so I'm not looking for any of its features. I did a cursory inspection of the commits added in 0.27, and I quickly notice that there were a few on topics other than adding the deprecation warning. I don't plan to check how those changes (or lack of) will affect the behaviour of Method::Signatures. I see that M::S have a patch to put the recommended workaround in place, and I'll wait for it to be released.
Subject: Re: [rt.cpan.org #120203] Please make the latest version of Any::Moose not warn about deprecation
Date: Thu, 16 Feb 2017 17:51:52 -0800
To: bug-Any-Moose [...] rt.cpan.org
From: Karen Etheridge <karen [...] froods.org>
I'll continue to put pressure on the maintainers to resolve https://rt.cpan.org/Ticket/Display.html?id=88716, the root issue -- patching around the deprecation warning isn't that great as we really *really* want to get everyone off of Any::Moose :) On Thu, Feb 16, 2017 at 3:08 PM, Daniel George Holz via RT < bug-Any-Moose@rt.cpan.org> wrote: Show quoted text
> Queue: Any-Moose > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=120203 > > > On Mon Feb 13 17:33:12 2017, ETHER wrote: >
> > Can you clarify which changes in 0.27 you were looking for? The other > > changes in 0.27 were all tooling or documentation-related and mostly > > benefit cpan clients and static analyzers.
> > Thank you for your summary of the other changes in 0.27. That helps me > feel confident that I will be OK pinning Any::Moose to 0.26 for my project. > > I'm not a direct consumer of Any::Moose, so I'm not looking for any of its > features. I did a cursory inspection of the commits added in 0.27, and I > quickly notice that there were a few on topics other than adding the > deprecation warning. > > I don't plan to check how those changes (or lack of) will affect the > behaviour of Method::Signatures. I see that M::S have a patch to put the > recommended workaround in place, and I'll wait for it to be released. >
It sounds like this is not needed; closing.
RT-Send-CC: karen [...] froods.org
Dana Sub 11. Velj 2017, 06:00:16, DGHOLZ reče: Show quoted text
> Hello, >
[...] Show quoted text
> > I'm a downstream user of Any::Moose (via Method::Signatures), so I > would have to apply the workaround of "no warnings 'deprecated';" to > each of my modules using M::S (until they release a version doing the > same); I'm hesitant to do this since it will mask deprecation warnings > from M::S. I'm planning to pin my code to use 0.26 of A::M, and review > the incidental changes made for 0.27 to see if I'm missing out on > anything.
Actually I don't think that using "no warnings 'deprecated';" would work at all --- warnings.pm is a lexical pragma, but the warning is emitted in Method::Signatures, not in your M::S-using module. So you have actually to patch M::S (see https://github.com/evalEmpire/method-signatures/pull/129/files) or put something like $SIG{__WARN__} = sub { warn $_[0] if $_[0] !~ /^Any::Moose is deprecated/ }; into your module (but I don't see a possibility to localize this pseudo signal handler).