whole files if the encoding was invalid. It will now read
up to 10 bytes ahead to find a valid character boundary or
give up. [ruby-core:19465]
* test/csv/test_features.rb, test/csv/test_table.rb, test/csv/test_row.rb:
Loosened some tests to check for a compatible? Encoding instea
of an exact Encoding. [ruby-core:19470]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19931 b2dd03c8-39d4-4d8f-98ff-823fe69b080e