aboutsummaryrefslogtreecommitdiffstats
path: root/actionpack
Commit message (Collapse)AuthorAgeFilesLines
* A couple enhancements to the router:José Valim2010-06-302-27/+75
| | | | | | | | * Allow to use the get :symbol shortcut outside resources scopes as well; * Fix a bug where :action was not being picked from given options; * Giving :as option inside a resource now changes just the relative name instead of the full name;
* Clean up the logic to specify the name and path for action a bit.José Valim2010-06-301-87/+51
|
* We are trying to test that & escapes here not that & is being escaped, ↵Santiago Pastorino2010-06-291-3/+2
| | | | | | also added a cosmetic change to test_link_tag_with_query_and_no_name Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* s/escape_once/html_escape/, since html safety is the contract that now says ↵Xavier Noria2010-06-306-20/+25
| | | | whether something has to be escaped
* url_for no longer escapes HTML, the :escape option is also goneXavier Noria2010-06-305-33/+14
| | | | Rationale: url_for is just a path/URL generator, it is the responsability of the caller to escape conveniently HTML needs it, JavaScript needs different escaping, a text mail needs no escaping at all, etc.
* _snowman CHANGELOG (dobry pies)wycats2010-06-291-0/+3
|
* Revert "Revert "Name compiled render methods". This caused several failures ↵Jeremy Kemper2010-06-291-8/+11
| | | | | | | | on AP test suite for 1.9.2." This reverts commit 9013227e00895ac95f29077229ec2fb156f450b0. minitest was 'running' compiled templates beginning with test!
* Rename _snowman_ to _snowman to be in sync with _method and _csrf_token.José Valim2010-06-293-3/+3
|
* Revert "Name compiled render methods". This caused several failures on AP ↵José Valim2010-06-291-11/+8
| | | | | | test suite for 1.9.2. This reverts commit 2c4f8aae5ca8489bb3e47712f50b3f57015811d9.
* Use caller for helpers_dir deprecation warningsJeremy Kemper2010-06-281-2/+2
|
* Slice new buffer from beginning of old buffer to skip encoding special caseJeremy Kemper2010-06-281-3/+1
|
* Name compiled render methodsJeremy Kemper2010-06-281-8/+11
|
* returning -> tapJeremy Kemper2010-06-281-5/+4
|
* ETag: use body instead of @body since the method will always return a stringJeremy Kemper2010-06-281-2/+2
|
* Tidy up valid conditions in router a bit.José Valim2010-06-281-2/+4
|
* Add :controller and :action to the list of valid conditionsAndrew White2010-06-281-0/+1
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Restores the escaping of urls generated from hashes. [#4765 state:resolved]Andrew White2010-06-282-9/+5
| | | | | | | | | | | | | | HTML specifications recommend the escaping of urls in web pages, which url_for does by default for string urls and consquently urls generated by path helpers as these return strings. Hashes passed to url_for are not escaped by default and this commit reverses this default so that they are escaped. Undoes the changes of this commit: http://github.com/rails/rails/commit/1b3195b63ca44f0a70b61b75fcf4991cb2fbb944 Signed-off-by: José Valim <jose.valim@gmail.com>
* Merge :constraints from scope into resource options [#2694 state:resolved]Andrew White2010-06-282-4/+49
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Convert instance_variables to symbols before excluding internal vars [#4965 ↵rohit2010-06-281-1/+1
| | | | | | state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com>
* Remove invalid conditions from route [#4989 state:resolved]Andrew White2010-06-283-4/+26
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Deprecate the old router DSL. Since it is still used intensively across ↵José Valim2010-06-283-15/+25
| | | | ActionPack test suite, patches that translates Rails internal tests to the new router DSL are welcome (note though that a few tests shouldn't be translated since they are testing exactly the old mapper API, like the ones in actionpack/test/controller/resource_test.rb and actionpack/test/controller/routing_test.rb)
* Strip_tags never ending attribute should not raise a TypeError [#4870 ↵Bruno Michel2010-06-283-0/+6
| | | | | | state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com>
* Update CHANGELOGs.José Valim2010-06-281-1/+2
|
* Small typowycats2010-06-271-1/+1
|
* Fix several known web encoding issues:wycats2010-06-277-198/+318
| | | | | | | | | | | | | | | | | | | | | | | * Specify accept-charset on all forms. All recent browsers, as well as IE5+, will use the encoding specified for form parameters * Unfortunately, IE5+ will not look at accept-charset unless at least one character in the form's values is not in the page's charset. Since the user can override the default charset (which Rails sets to UTF-8), we provide a hidden input containing a unicode character, forcing IE to look at the accept-charset. * Now that the vast majority of web input is UTF-8, we set the inbound parameters to UTF-8. This will eliminate many cases of incompatible encodings between ASCII-8BIT and UTF-8. * You can safely ignore params[:_snowman_] TODO: * Validate inbound text to confirm it is UTF-8 * Combine the whole_form implementations in form_helper_test and form_tag_helper_test
* Upgrade to Rack 1.2.1Jeremy Kemper2010-06-272-9/+15
|
* Vendor unreleased rack-mount 0.6.6.pre dependencyJeremy Kemper2010-06-2736-2/+4015
|
* AV::logger returns AC::logger if it's defined, workaround meanwhile AV ↵Santiago Pastorino2010-06-281-2/+3
| | | | | | doesn't have it's own logger Signed-off-by: José Valim <jose.valim@gmail.com>
* Move Rails module to abstract_unit to make test in isolation workSantiago Pastorino2010-06-282-3/+3
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Fixed that an ArgumentError is thrown when request.session_options[:id] is ↵Michael Lovitt2010-06-276-36/+113
| | | | | | | | read in the following scenario: when the cookie store is used, and the session contains a serialized object of an unloaded class, and no session data accesses have occurred yet. Pushed the stale_session_check responsibility out of the SessionHash and down into the session store, closer to where the deserialization actually occurs. Added some test coverage for this case and others related to deserialization of unloaded types. [#4938] Signed-off-by: José Valim <jose.valim@gmail.com>
* Normalize recall params when the route is not a standard route otherwise ↵Andrew White2010-06-272-3/+48
| | | | | | :controller and :action may appear in the generated url [#4326 state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com>
* Change reference to Test::Unit::AssertionFailedError to the generic ↵David Trasbo2010-06-261-1/+1
| | | | | | ActiveSupport::TestCase::Assertion [#4987 state:commited] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* Small changes to responder:José Valim2010-06-261-5/+13
| | | | | | * resources is always an array; * Lazy retrieve request and formats; * Alias api_location and navigation_location to resource_location, making easier to change its behavior without affecting each other and without a need to reimplement any of the behavior methods.
* Support optional static segements as well [#4832 state:resolved]Andrew White2010-06-262-1/+35
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Fixed normalize_path in Routing::Mapper to handle optional prefix segments ↵Paul Barry2010-06-262-1/+47
| | | | | | with static and dynamic parts Signed-off-by: José Valim <jose.valim@gmail.com>
* Don't force pluralization of controller name when defining a resource [#4980 ↵Andrew White2010-06-262-1/+18
| | | | | | state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com>
* URL fragments should not have safe characters escaped. Ref: Appendix A, ↵Andrew White2010-06-252-4/+10
| | | | | | | | http://tools.ietf.org/rfc/rfc3986.txt [#4762 state:resolved] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* Failing test for whether glob parameters accept regexpsAndrew White2010-06-251-0/+9
| | | | | | [#4970 state:committed] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* Add failing test case for parameters with periodsAndrew White2010-06-251-0/+27
| | | | | | [#2536 state:resolved] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* Do not mark the session as loaded if an error happened while doing it.José Valim2010-06-251-2/+1
|
* Calling exists? in the session store, without checking for stale sessions, ↵José Valim2010-06-252-10/+10
| | | | was causing the cookie store to panic because we need to unpack the whole session to get its key. This commit fixes this issue and also caches exists calls for performance improvements.
* Do not wrap hidden fields with error proc [#4962 state:resolved]Carlos Antonio da Silva2010-06-253-3/+23
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Make sure that Rails doesn't resent session_id cookie over and over again if ↵Prem Sichanugrist2010-06-253-1/+27
| | | | | | | | it's already there [#2485 state:resolved] This apply to only Active Record store and Memcached store, as they both store only the session_id, which will be unchanged, in the cookie. Signed-off-by: José Valim <jose.valim@gmail.com>
* Just reading flash messages should not create a session if one does not ↵José Valim2010-06-253-5/+29
| | | | exist yet.
* Avoid deserializing cookies too early, which causes session objects to not ↵José Valim2010-06-243-21/+25
| | | | be available yet. Unfortunately, could not reproduce this in a test case.
* Fix namespaced redirects [#4389 state:resolved]Andrew White2010-06-242-1/+15
| | | | Signed-off-by: José Valim <jose.valim@gmail.com>
* Move Rails::LogSubscriber to ActiveSupport::LogSubscriber, allowing ↵José Valim2010-06-2411-104/+101
| | | | frameworks like ActiveRecord and ActiveResource to log outsude Rails::Application [#4816 state:resolved]
* Avoid using Pathname on Resolver and AS::Dependencies.José Valim2010-06-241-1/+1
|
* Sessions should not be created until written to and session data should be ↵Michael Lovitt2010-06-238-35/+190
| | | | | | | | destroyed on reset. [#4938] Signed-off-by: Jeremy Kemper <jeremy@bitsweat.net>
* Expose view via the view() method in AV::TestCase::BehaviorDavid Chelimsky2010-06-232-17/+24
| | | | | | | | | | - was exposed as _view, which suggested it was private - left _view as an alias of view as not to break any extensions that are relying on _view [#4932 state:resolved] Signed-off-by: José Valim <jose.valim@gmail.com>