No takers? I've documented this at http://ironruby.net/Documentation/Ruby_Standard_Library/Modifications so if anyone wants to bite the info is there.
~js > -----Original Message----- > From: Jimmy Schementi > Sent: Tuesday, June 23, 2009 2:45 PM > To: Tomas Matousek; IronRuby External Code Reviewers > Cc: [email protected] > Subject: RE: Code Review: HashFixes > > How are we going to make sure this gets back into REXML and RubyGems? > We shouldn't ship significantly different versions of the Ruby stdlib, as > people will expect to switch to any version of the stdlib without issues. > > My first thought is to have someone in the community step up to propose > these changes to RubyGems and REXML for us, and get them checked in. > Anyone interested? > > ~js > > > -----Original Message----- > > From: Tomas Matousek > > Sent: Tuesday, June 23, 2009 1:37 PM > > To: IronRuby External Code Reviewers > > Cc: [email protected] > > Subject: Code Review: HashFixes > > > > tfpt review "/shelveset:HashFixes;REDMOND\tomat" > > > > Fixes implementation of hash methods in REXML and RubyGems. The > > current implementation is incorrect since it sums hash codes of > > subcomponents potentially overflowing to Bignum. This issue is mainly > > exposed when calculating hash of an array. Array#hash calculates its > > hash code by calling hash on each item and converting the result of > > the hash method into Fixnum. So if the result if the hash method is a > > Bignum an error is reported: > > > > class C > > def hash > > 100000000000000000000 > > end > > end > > > > [C.new].hash # => in `hash': bignum too big to convert into `long' > > (RangeError) > > > > Tomas _______________________________________________ Ironruby-core mailing list [email protected] http://rubyforge.org/mailman/listinfo/ironruby-core
