[cgiapp] Re: peer review request for UTF patch: "Stray UTF8 bit on headers infects the whole output"

[Mark Stosberg]

> Mark Stosberg wrote:
> > Could someone who is more UTF-8 savvy than myself peer review this simple UTF-8
> > patch for CGI::Application for me?
> > 
> > http://rt.cpan.org/Public/Bug/Display.html?id=46513
> > 
> > You can comment through the ticket or here.
> > 
> >     Mark
> > 
> 
> You may find this thread on perlmonks relevant: 
> http://www.perlmonks.org/index.pl?node_id=683833
> 
> It also implies that the patch should probably test for the existence of the 
> utf8::downgrade function (for older perls).
> 
> My personal opinion: this patch is probably a good idea (given that LWP does 
> something similar), even though I have never had an issue with utf8 headers 
> myself (and I do output quite a bit of utf8 content). I'm by no means an 
> expert on encodings, but I do trust Juerds expertise on this subject (he's the 
> author of perlunitut).

So, the original patch loooked like this:

+	utf8::downgrade($headers, 1);

Based on Perlmonks thread, are you suggesting an updated patch should
look like this? :

+ *_utf8_downgrade = defined(&utf8::downgrade) 
+     ?  sub { utf8::downgrade($_[0], 1) or Carp::croak("content must be bytes") }
+     :  sub { }
+     ;
+	_utf8_downgrade($headers, 1);

    Mark