Permalink
Commits on Dec 23, 2010
Commits on Dec 1, 2010
  1. fix a spec

    Andy Delcambre committed Dec 1, 2010
  2. Upgrade to bundler 1.0, use gemspec

    Andy Delcambre committed Dec 1, 2010
Commits on Nov 30, 2010
  1. Add bundler 1.0 stuff to .gitignore

    Andy Delcambre committed Nov 30, 2010
Commits on Aug 24, 2010
  1. Given DM1's preference of AS over Extlib, remove all {String,Symbol}#/

    jpr5 committed Aug 24, 2010
    references in favor of how DataMapper does it.
    
    Addresses issue: http://github.com/halorgium/dm-salesforce/issues/#issue/6
Commits on Aug 16, 2010
  1. Updated to support DM 1.x.

    jpr5 committed Aug 16, 2010
Commits on Mar 24, 2010
  1. Re-organized README.

    jpr5 committed Mar 24, 2010
  2. Remove a few custom Types' defaults to re-enable standard nil "defaul…

    jpr5 committed Mar 24, 2010
    …t value" behavior
    
    DataMapper explicitly disallows :default => nil in class property declarations.  When a
    DataMapper custom Type declares a non-nil default, there is no way to get nil property
    values "by default" upon object creation - they have to be set explicitly every time.
    
    This flies against long-standard DB and DM convention, where declared fields are nullable
    by default unless explicitly specified otherwise (SQL: NOT NULL, DM: :required => true).
    This is also true of the Salesforce object model.
    
    If default values are desired, they should be expressed via class property declarations,
    not underlying Type.  Maybe validation behavior should be considered separately?
  3. Fix formatting problem.

    jpr5 committed Mar 23, 2010
Commits on Mar 23, 2010
Commits on Feb 4, 2010
Commits on Jan 5, 2010
  1. next version is 0.10.4

    atmos committed Jan 5, 2010
Commits on Jan 4, 2010
  1. Remove do_sqlite3 dependency because it isn't needed for the core fun…

    codebrulee committed Jan 4, 2010
    …ctionality of the library.
Commits on Dec 18, 2009
  1. next version is 0.10.3

    ultraninja committed Dec 18, 2009
  2. Fix incompatibilities with DataMapper 0.10.2:

    ultraninja committed Dec 18, 2009
    Properly switch between 1/0 for writes and TRUE/FALSE for WHERE clauses.
    Support negation operator correctly. (i.e. :id.not => Account.engineyard.id)
    Refactored using DataObjectsAdapter as a template. More similarities
    between these modules now.
Commits on Dec 17, 2009
  1. Fixups to make dm-salesforce work with 0.10.2

    ultraninja committed Dec 17, 2009
    Eager loading will also receive Collections in addition to arrays
    Remove traces of nullable => false in favor of non-deprecated API
    Correct the behavior of the Salesforce Boolean type
Commits on Nov 5, 2009
  1. next version is 0.10.1

    ultraninja committed Nov 5, 2009
  2. adding 0.10.0 release, this needs evaluation and love but our interna…

    ultraninja committed Nov 5, 2009
    …l gems are running happily against it. a point release will followin a week or two after we've evaluated it
  3. Fixed association loading

    ultraninja committed Nov 5, 2009
Commits on Nov 4, 2009
  1. Queries working with SEL

    ultraninja committed Nov 4, 2009
  2. Fixed the server_unavailable? method, Fixed the Boolean type, refacto…

    ultraninja committed Nov 4, 2009
    …red an association spec
Commits on Nov 3, 2009
  1. Cleaned up the namespaces, Added tests for the SQL operators, removed…

    ultraninja committed Nov 3, 2009
    … a bunch of unused code, refactored error handling in the adapter
  2. gemfile updates, remove path pushing in favor of explicit inclusion, …

    ultraninja committed Nov 3, 2009
    …remove traces of auto_migrate, introduce custom types to ease intercepting BS between the adapter and model layer, test review to think about more than just EY's need.
Commits on Nov 2, 2009
Commits on Oct 30, 2009
Commits on Oct 24, 2009
  1. Improve/fix error handling for cases when multiple errors are reporte…

    jpr5 committed Oct 24, 2009
    …d by SF at the same time (i.e. >1 key collision).
    
    Also added some NOTE/TODO/FIXMEs for problems observed while debugging a related issue.  Still some bugs to track down.