Since on day 2 I stayed awake until 6am, I woke up on 1pm, a little late.
Ingy wasn't sleeping at all (at least it looked like), he was about to release his new YAML and I helped him testing them on Win32 systems. Apparently it was broken, as it's supposed to be. But the breakage is really small (just 1 test of thousands of tests is failing) and we found that even Test::More is broken on my Windows + ActivePerl. heh.
Updated Catalyst::View::JSON to support Safari utf-8 problems and Opera MIME-type gotcha.
Meanwhile I threw a question on #catalyst IRC channel about the possibility to include charset/encoding handling on Catalyst core, so all the plugins can use the information sanely. I quickly brainstoemed with nothingmuch about the real world use cases, etc.
I released Text::Emoticon which is just a few lines of code. It correctly dispatches MSN and Yahoo subclass and provide the same interface for it. I tried to create Kwiki::Emoticon plugin to hook Emoticon format into Kwiki's wiki format. But obviously adding new Wiki format was not the easiest thing to do in the world. I'll hack with gugod and Ingy on the 4th day.
I bugged Audrey about the weird test failure and JSON incompatiblities of her YAML::Syck module and we finally released JSON::Syck module! Well, this is under my CPAN directory but all I did was just to bug Audrey and made some test cases :) Currently JSON::Syck doesn't handle non-ASCII characters very well. I've been talking with Douglas Crockford (the JSON guy) and why the lucky stiff asking what's the right thing to do, to encode/decode Unicode strings bewtween javascript and Perl (UTF-8 bytes or "\uXXXX"?)
Updated Catalyst::View::JSON so that you can use JSON::Syck as a backend converter. I'll make Data::Serializer driver (plugin) to absorb the API difference between JSON and JSON::Syck.
Audrey came back to her home @ Taipei since she'll have $work on Wednesday. After she went back we (Ingy, gugod and I) went to the nice Cafe called "Dali". We had a nice coffee/tea there and off course hacked together. I took a quick look on Sekimura-san's Template::Provider::Encode and came along with a better module Template::Provider::Encoding (I know it's a little confusing). Template::Provider::Encoding forces a small convention on template files like [% USE encoding 'euc-jp' %] at the top of the tempalte (otherwise it's assumed in UTF-8) and it's a good convention I think.
17 svn commits, 7 CPAN uploads.
