On 11/29/06, a.r.ferreira@gmail.com via RT
<bug-B-Deobfuscate@rt.cpan.org> wrote:
Show quoted text>
> Wed Nov 29 07:30:59 2006: Request 23697 was acted upon.
> Transaction: Ticket created by a.r.ferreira@gmail.com
> Queue: B-Deobfuscate
> Subject: B-Deobfuscate 0.14+ broke Acme::Floral and Acme::MetaSyntactic::RefactorCode
> Broken in: 0.14, 0.15, 0.16
> Severity: Normal
> Owner: Nobody
> Requestors: a.r.ferreira@gmail.com
> Status: new
> Ticket <URL:
http://rt.cpan.org/Ticket/Display.html?id=23697 >
>
>
> Both modules, Acme::Floral and Acme::MetaSyntactic::RefactorCode, relied
> on the default behavior of B::Deobfuscate to read the data from
> *B::Deobfuscate::DATA and then these modules override it to enter their
> own stuff. As the default mechanism changed in 0.14 (which nows loads
> one of the default dictionary modules), these modules got broken. It may
> be understood as a bug in these modules. The bug is more a documentation
> issue and the effect of the change without notice.
Acme::Floral was doing very bad things and you shouldn't be surprised
that it broke. Acme::MetaSyntactic::RefactorCode is *also* doing very
bad things if it's using the same trick. Acme::Floral's dictionary got
subsumed into B::Deobfuscate so the "new" version should just be a
wrapper over B::O with the right parameters.
A::MS::RC should either use whatever got documented (assuming I
remembered to document something) or propose something back at me
about how to do this. I'm open to suggestions but kind of think this
new method is likely to be good for everyone.
Josh
