Skip Menu |

This queue is for tickets about the Net-OpenID-Common CPAN distribution.

Report information
The Basics
Id: 78218
Status: resolved
Priority: 0/
Queue: Net-OpenID-Common

People
Owner: Nobody in particular
Requestors: kevinbosak [...] gmail.com
Cc:
AdminCc:

Bug Information
Severity: Important
Broken in: 1.14
Fixed in: 1.15



Subject: Yahoo HTML parsing error
I have a Dancer application using Perl 5.14.2 and Net::OpenID::Consumer. While the other OpenID providers our site supports work fine, authenticating via Yahoo causes HTML::Parser to thow this error: "Parsing of undecoded UTF-8 will give garbage when decoding entities". This happens after the user logs into Yahoo and is sent back to our site with a token. More info, a code sample, and a possible fix here: http://stackoverflow.com/questions/11192557/how-can-i-avoid-encoding-errors-using- netopenidconsumer-with-yahoo-openids
Here's a git diff to apply that'll fix this bug. I don't have access to commit to: https://github.com/wrog/Net-OpenID- Common diff --git a/lib/Net/OpenID/URIFetch.pm b/lib/Net/OpenID/URIFetch.pm index 909c30a..41d7110 100644 --- a/lib/Net/OpenID/URIFetch.pm +++ b/lib/Net/OpenID/URIFetch.pm @@ -112,14 +112,10 @@ sub fetch { return $cached_response->(); } else { - my $content = $res->content; + my $content = $res->decoded_content; my $final_uri = $res->request->uri->as_string(); my $final_cache_key = "URIFetch:${prefix}:${final_uri}"; - if ($res->content_encoding && $res->content_encoding eq 'gzip') { - $content = Compress::Zlib::memGunzip($content); - } - if ($content_hook) { $content_hook->(\$content); }
ok, this is fixed in 1.15. Thanks for the patch and sorry this took so long (as it turned out this also required changes in Yadis.pm where we were also doing some manual decoding that now gets to go away. yay)