Permalink
Commits on Aug 21, 2010
  1. @laserlemon
  2. @laserlemon

    Fixed the dummy service initialization in both the attributes and val…

    …idation specs so that it works within nested spec contexts.
    laserlemon committed with Jun 1, 2010
  3. @laserlemon

    Changed the name of the dummy service class in the attributes spec to…

    … be consistent with the validation spec.
    
    Having the same "Service" class name for both was causing problems.
    laserlemon committed with Jun 1, 2010
  4. @laserlemon
  5. @laserlemon
  6. @laserlemon

    Removed the ATTRIBUTE_TYPES constant from the dummy service class in …

    …the attributes spec that was leftover from some removed test code.
    laserlemon committed with Jun 1, 2010
  7. @laserlemon
  8. @laserlemon

    Changed the type of error that is raised when no service target exist…

    …s in order to pass a test.
    laserlemon committed with May 31, 2010
  9. @laserlemon
  10. @laserlemon

    Resolved the failing tests by undefining all but the double-underscor…

    …e and real_class methods from Shippinglogic::Proxy.
    laserlemon committed with May 31, 2010
  11. @laserlemon
  12. @laserlemon

    Added a new error spec.

    laserlemon committed with May 31, 2010
  13. @laserlemon

    Final tweaks to the attributes spec to ensure functionality with Ruby…

    … 1.8.6, 1.8.7, 1.9.1 and 1.9.2.
    laserlemon committed with May 28, 2010
  14. @laserlemon
  15. @laserlemon
  16. @laserlemon
  17. @laserlemon
  18. @laserlemon
  19. @laserlemon
  20. @laserlemon
  21. @laserlemon
  22. @laserlemon
  23. @laserlemon
  24. @laserlemon
  25. @laserlemon
  26. @laserlemon

    Changed sentence-form error messages to comma-separated.

    Didn't bother with replicating Array#to_sentence functionality.
    laserlemon committed with May 28, 2010
  27. @laserlemon
  28. @laserlemon
  29. @laserlemon
  30. @laserlemon

    Allowed attribute reading to properly parse and return Time objects f…

    …or :datetime type attributes.
    laserlemon committed with May 28, 2010
  31. @laserlemon
  32. @laserlemon
  33. @laserlemon

    Made the Attributes-defined initializer more robust so that it will n…

    …ot require a base argument.
    laserlemon committed with May 28, 2010
  34. @laserlemon

    Made the Attributes mixin reusable for non-proxy classes.

    The primary reason behind this is to focus testing on just the features of the Attributes module without requiring all the proxy-specific code.
    laserlemon committed with May 28, 2010
  35. @laserlemon

    Resetting the service target upon writing an attribute should be the …

    …responsibility of the service, not the Attributes mixin.
    laserlemon committed with May 28, 2010