aboutsummaryrefslogtreecommitdiffstats
path: root/activemodel
Commit message (Collapse)AuthorAgeFilesLines
* renamed mass-assignment scopes to roles, updated code, tests, docs and ↵Josh Kalderimis2011-05-082-25/+25
| | | | security guide
* Add documentation for new observer enable/disable feature.Myron Marston2011-05-063-3/+53
|
* Update CHANGELOGs.José Valim2011-05-061-0/+2
|
* Prepare for beta1David Heinemeier Hansson2011-05-041-1/+1
|
* Improved ActiveModel Observing docs.Sebastian Martinez2011-04-301-2/+4
|
* Merge branch 'master' of github.com:lifo/docrailsDan Pickett2011-04-293-4/+4
|\
| * Rephrased Dirty#changed? docsSebastian Martinez2011-04-261-1/+1
| |
| * Removed extra whitespaceSebastian Martinez2011-04-261-1/+1
| |
| * Better formatting on ActiveModel::TranslationSebastian Martinez2011-04-261-2/+2
| |
* | fix typoAditya Sanghi2011-04-291-1/+1
| |
* | :if should not fire on validations when not in context with :onAditya Sanghi2011-04-293-3/+17
| |
* | Add additional tests for AM::ObserverArray that I had missed yesterday.Myron Marston2011-04-281-1/+50
| |
* | Fix bug with AM::Observer disablement.Myron Marston2011-04-283-25/+27
| | | | | | | | | | Now that we propagate the enabling/disabling to descendants, we no longer have to check the disabled_observer Set on each superclass of the model class. This was causing a bug when disabling all observers at a superclass level and then enabling an individual observer at a subclass level. Plus the logic is simpler now :).
* | Fix dev env memory leaks by using AS::DescendantsTracker rather than keeping ↵Myron Marston2011-04-282-36/+12
| | | | | | | | | | track of subclasses manually. There's also no need to keep track of all ObserverArray instances in a hash, as this is likely to leak memory, too.
* | Revert "Revert "Handle enabling/disabling observers at different levels of ↵Myron Marston2011-04-283-3/+75
| | | | | | | | | | | | | | | | | | the class hierarchy."" This reverts commit 2a25c5818b03d7d6cd63aad180bff23479dbd861. I'm going to add another commit that keeps the same behavior of fixes the problems of leaking memory in development.
* | Revert "Handle enabling/disabling observers at different levels of the class ↵José Valim2011-04-283-75/+3
| | | | | | | | | | | | | | | | | | | | hierarchy." This reverts commit ad62f1928768bd2676958a4a08512bad342fe469 because the current subclasses implementation leaks memory in development. Instead of keeping an array of subclasses, the better solution is to ask the parent if an observer is being disabled or not.
* | Handle enabling/disabling observers at different levels of the class hierarchy.Myron Marston2011-04-283-3/+75
| | | | | | | | Last call wins.
* | Allow observers to be enabled and disabled.Myron Marston2011-04-285-3/+276
| | | | | | | | This is useful in situations like model unit tests and the occasional rake task to backfill old data.
* | Wrap line that is over 200 characters long. Now it's much easier to read.Myron Marston2011-04-281-1/+5
| |
* | bump the i18n gem to 0.6.0beta1 and update docs for the i18n exception ↵Sven Fuchs2011-04-261-1/+1
| | | | | | | | | | | | handler [#6739 state:committed] Signed-off-by: José Valim <jose.valim@gmail.com>
* | minor correction to the AMo mass-assignment security docsJosh Kalderimis2011-04-241-1/+1
| |
* | AM mass assignment security attr_accessible and attr_protected now allow for ↵Josh Kalderimis2011-04-244-45/+135
|/ | | | | | | | scopes using :as => scope eg. attr_accessible :name attr_accessible :name, :admin, :as => :admin
* minor correction to the Active Model instantiate_observers docJosh Kalderimis2011-04-221-1/+1
|
* let this regexp be stricterXavier Noria2011-04-191-1/+1
| | | | | | | | | The purpose of this regexp is to detect strings that are valid identifiers to be used as method names. The anchor $ allows a trailing newline, and both ^ and $ may be bypassed because Ruby regexps are multiline always. The anchors \A and \z are the proper ones to ensure string boundaries.
* Let's not dup because value may not always be duplicable.José Valim2011-04-191-1/+1
|
* Avoid define_method if possible.José Valim2011-04-191-12/+39
|
* bcrypt will encrypt anything, so validate_presence_of would not catch nil / ↵Aaron Patterson2011-04-142-1/+15
| | | | blank passwords. Thank you to Aleksander Kamil Modzelewski for reporting this
* Declaring the attr_accessor for password_confirmation is not necessaryBrian Cardarella2011-04-141-2/+1
| | | | as the confirmation validation already adds that attr_accessor
* Remove `#among?` from Active SupportPrem Sichanugrist2011-04-131-1/+1
| | | | | | After a long list of discussion about the performance problem from using varargs and the reason that we can't find a great pair for it, it would be best to remove support for it for now. It will come back if we can find a good pair for it. For now, Bon Voyage, `#among?`.
* Merge branch 'master' of git://github.com/lifo/docrailsXavier Noria2011-04-121-1/+1
|\
| * RewordingOge Nnadi2011-04-111-1/+1
| |
* | Change Object#either? to Object#among? -- thanks to @jamesarosen for the ↵David Heinemeier Hansson2011-04-121-1/+1
| | | | | | | | suggestion!
* | Remove code duplication.José Valim2011-04-113-20/+27
| |
* | No need to rescue here. Invoking the wrong method in an object can happen ↵José Valim2011-04-116-40/+0
| | | | | | | | anywhere if you are not careful. In other words, test your shit.
* | Revert "Add :use_include option to allow user to explicitly use ↵José Valim2011-04-114-37/+14
| | | | | | | | | | | | | | | | `Range#include?` method in Ruby 1.9" Use :with => range.to_a instead. This reverts commit f6540211b5b9133c9f93c11655a04d613c237e67.
* | Using Object#in? and Object#either? in various placesPrem Sichanugrist2011-04-112-2/+4
| | | | | | | | There're a lot of places in Rails source code which make a lot of sense to switching to Object#in? or Object#either? instead of using [].include?.
* | Add :use_include option to allow user to explicitly use `Range#include?` ↵Prem Sichanugrist2011-04-104-14/+37
| | | | | | | | | | | | method in Ruby 1.9 In Ruby 1.9 we're currently use `Range#cover?` to fix the performance problem. However, there might be the case that you want to use `Range#include?` instead. This patch will give you that option.
* | Add support for proc or lambda as an option for InclusionValidator, ↵Prem Sichanugrist2011-04-107-20/+167
| | | | | | | | | | | | ExclusionValidator, and FormatValidator You can now use a proc or lambda in :in option for InclusionValidator and ExclusionValidator, and :with, :without option for FormatValidator
* | Rename test directory to match code directoryJohn Firebaugh2011-04-062-0/+0
|/ | | | Hopefully "serializeration" is not an in joke that I missed.
* Merge branch 'master' of git://github.com/lifo/docrailsXavier Noria2011-04-031-11/+10
|\
| * Fix formatting and broken markupJohn Firebaugh2011-03-281-11/+10
| |
* | Gem::Specification#has_rdoc= is deprecated since rubygems 1.7.0Santiago Pastorino2011-04-021-2/+0
| |
* | Bring back AMo#i18n_key methodSantiago Pastorino2011-03-303-8/+13
|/
* use class_eval with a string when it's possibleSantiago Pastorino2011-03-221-2/+8
|
* Do not in place modify what table_name returnsSantiago Pastorino2011-03-222-12/+1
|
* define_attr_method should be able to define methods that returns nilSantiago Pastorino2011-03-172-2/+4
|
* remove misleading commentAaron Patterson2011-03-171-2/+0
|
* dup strings on return so that in place modifications do not break anything. ↵Aaron Patterson2011-03-172-1/+12
| | | | I am looking at you "compute_table_name"
* define_attr_method correctly defines methods with invalid identifiersSantiago Pastorino2011-03-162-5/+6
|
* Add a define_attr_method testSantiago Pastorino2011-03-161-0/+13
|