Permalink
Commits on Sep 10, 2010
  1. 1.1.2 release.

    committed Sep 10, 2010
  2. Documentation update.

    committed Sep 10, 2010
Commits on Sep 9, 2010
  1. Don't extend a Net::HTTP response with our module when WebMock has al…

    …ready extended it with its own module.
    
    These modules duplicate the same behavior and stomp on each other's toes.
    committed Sep 9, 2010
Commits on Sep 8, 2010
  1. When the result of a request is an exception, raise it, rather than t…

    …rying to read a body from it.
    committed Sep 2, 2010
Commits on Aug 28, 2010
  1. List patron and em-http-request in our Gemfile so that our Gemfile.lo…

    …ck doesn't change everytime we run something on JRuby.
    committed Aug 28, 2010
Commits on Aug 27, 2010
  1. 1.1.1 release.

    committed Aug 27, 2010
  2. The newest ruby-debug-base19 isn't working on 1.9.1 for me, so lock t…

    …o the previous release.
    committed Aug 27, 2010
  3. Lock dev dependency to WebMock 1.3.3.

    On 1.3.4 I get a failing cucumber scenario, but I believe it's a bug in WebMock.  I'm going to investigate more.
    committed Aug 27, 2010
Commits on Aug 26, 2010
  1. Updated Gemfile/gemspec/Rakefile, etc to use latest best practices ba…

    …sed on the gem skeleton that bundler generates.
    committed Aug 26, 2010
  2. Refactored and updated #check_version! on FakeWeb/WebMock adapters so…

    … that a warning is printed for a version with a greater major or minor version number.
    
    This simulates the ~> gem comparison operator, and is inspired by wycats' recent blog post:
    
    http://yehudakatz.com/2010/08/21/using-considered-harmful-or-whats-wrong-with/
    committed Aug 26, 2010
  3. Change http stubbing adapters to be modules rather than classes.

    They should never be instantiated and don't really hold state.  A module is more appropriate here.
    committed Aug 26, 2010
Commits on Aug 24, 2010
  1. Use a localhost regex with FakeWeb.allow_net_connect for the ignore_l…

    …ocalhost feature.
    
    This is simpler than changing allow_net_connect from false to true and back again in our Net::HTTP extension for a localhost request.
    committed Aug 24, 2010
Commits on Aug 23, 2010
  1. Use FakeWeb::Utility.request_uri_as_string rather than our own logic …

    …to create the request uri.
    committed Aug 22, 2010
  2. Upgrade to FakeWeb 1.3.0.

    committed Aug 23, 2010
  3. Optimized #stub_requests a bit.

    Appending to an array with `<< value` is faster than with `+= [value]`, as it doesn't create a new array object for each append.
    
    A benchmark shows there is about a 10x performance difference:
    
    http://gist.github.com/544809
    committed Aug 23, 2010
Commits on Aug 22, 2010
  1. 1.1.0 release.

    committed Aug 22, 2010
  2. Readme updates.

    committed Aug 22, 2010
  3. Lock rspec-core to one of my branches that fixes an issue with shared…

    … example groups on 1.8.6.
    committed Aug 22, 2010
  4. 1.9.2 compatibility fix: in an override, use an alias to refer to the…

    … original method rather than super.
    
    This works around an error I'm getting on 1.9.2:
    
    "super from singleton method that is defined to multiple classes is not supported; this will be fixed in 1.9.3 or later"
    committed Aug 22, 2010
  5. 1.9.2 compatibility fix: don't try to use ruby-debug on 1.9.2.

    On 1.9.2, I'm getting errors like these:
    
    dyld: lazy symbol binding failed: Symbol not found: _rb_method_node
      Referenced from: /Users/mmarston/.rvm/gems/ruby-1.9.2-p0/gems/ruby-debug-base19-0.11.23/lib/ruby_debug.bundle
      Expected in: flat namespace
    
    dyld: Symbol not found: _rb_method_node
      Referenced from: /Users/mmarston/.rvm/gems/ruby-1.9.2-p0/gems/ruby-debug-base19-0.11.23/lib/ruby_debug.bundle
      Expected in: flat namespace
    committed Aug 22, 2010
  6. 1.9.2 compatibility fix: sort attributes before putting them in a Set.

    On 1.9.2, Set.new([1, 2]).hash != Set.new([2, 1]).hash, but Set.new([1, 2]).hash == Set.new([1, 2]).hash.
    committed Aug 22, 2010