aboutsummaryrefslogtreecommitdiffstats
path: root/actionpack/lib/action_dispatch
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #35229 from rails/fix-35222Aaron Patterson2019-02-131-2/+2
|\ | | | | Convert lookup context's to a stack for fixing #35222 and #34138
| * Turn lookup context in to a stack, push and pop if formats changeAaron Patterson2019-02-111-2/+2
| | | | | | | | | | | | | | | | | | | | | | This commit keeps a stack of lookup contexts on the ActionView::Base instance. If a format is passed to render, we instantiate a new lookup context and push it on the stack, that way any child calls to "render" will use the same format information as the parent. This also isolates "sibling" calls to render (multiple calls to render in the same template). Fixes #35222 #34138
* | Merge pull request #35175 from drn/create-sessionGuillermo Iguaran2019-02-121-1/+1
|\ \ | |/ |/| Support testing of non-ActionDispatch-routed apps.
| * Support testing of non-ActionDispatch-routed apps.Darren Cheng2019-02-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The [Grape API framework](https://github.com/ruby-grape/grape) regularly writes tests like [spec/grape/api_spec.rb](https://github.com/ruby-grape/grape/blob/master/spec/grape/api_spec.rb). When attempting to write a test in a Rails environment similar to the following: ``` describe Grape::Api, type: :request do let(:app) { Class.new(Grape::API) do get 'test' do { foo: 'bar' } end end } it '200s' do get 'test' end end ``` The following exception is thrown: ``` NoMethodError: undefined method `url_helpers' for #<Array:0x00007fb4e4bc4c88> -- 0: .../lib/action_dispatch/testing/integration.rb:330:in `block in create_session' 1: .../lib/action_dispatch/testing/integration.rb:326:in `initialize' 2: .../lib/action_dispatch/testing/integration.rb:326:in `new' 3: .../lib/action_dispatch/testing/integration.rb:326:in `create_session' 4: .../lib/action_dispatch/testing/integration.rb:316:in `integration_session' 5: .../lib/action_dispatch/testing/integration.rb:348:in `block (2 levels) in <module:Runner>' ``` This change explicitly ensures that `app.routes` is an `ActionDispatch::Routing::RouteSet` instance.
* | Merge pull request #35036 from rails/av-base-subclassAaron Patterson2019-02-061-0/+4
|\ \ | |/ |/| Move compiled ERB to an AV::Base subclass
| * Pull generated methods up in to the anonymous subclassAaron Patterson2019-02-061-6/+1
| | | | | | | | Then we don't need the extra module.
| * Move templates to an anonymous subclass of AV::BaseAaron Patterson2019-02-061-0/+9
| | | | | | | | | | Now we can throw away the subclass and the generated methods will get GC'd too
* | Update terminal-to-html inline images url in ScreenshotHelperbetterzega2019-02-051-1/+1
| |
* | Merge pull request #35162 from silppuri/fix-incorrectly-matching-unachored-pathsAaron Patterson2019-02-051-1/+1
|\ \ | |/ |/| Fix incorrectly matching unanchored paths
| * Restrict matching with word boundary or end of stringPetri Avikainen2019-02-051-1/+1
| |
| * Define word boundary for unanchored path regexpPetri Avikainen2019-02-051-1/+1
| |
* | Merge pull request #35139 from 7coAim/fix_debug_exceptionsGeorge Claghorn2019-02-052-2/+2
|\ \ | | | | | | Fix NameError : Make debug exceptions works in an environment where ActiveStorage is not loaded.
| * | fix NameErrorkurosawat2019-02-052-2/+2
| | | | | | | | | | | | NameError: uninitialized constant ActionView::CompiledTemplates::ActiveStorage
* | | Merge pull request #35134 from Edouard-chin/ec-cookie-expiry-regressionRafael França2019-02-041-7/+2
|\ \ \ | | | | | | | | Cookie doesn't expire anymore unless a flag is set:
| * | | Cookie doesn't expire anymore unless a flag is set:Edouard CHIN2019-02-041-7/+2
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - There is a regression in 6.0 introduced by #32937 where cookie doesn't expire anymore unless the new `use_cookies_with_metadata` configuration is set to `true`. This causes issue for app migration from 5.2 to 6.0 because the `use_cookies_with_metadata` flag can't be set to true until all servers are running on 6.0. Here is a small reproduction script that you can run in the console ```ruby ActionDispatch::Cookies request = ActionDispatch::Request.empty request.env["action_dispatch.key_generator"] = ActiveSupport::KeyGenerator.new('1234567890') request.env["action_dispatch.signed_cookie_salt"] = 'signed cookie' request.env["action_dispatch.cookies_rotations"] = ActiveSupport::Messages::RotationConfiguration.new request.env["action_dispatch.use_authenticated_cookie_encryption"] = true signed_cookie = request.cookie_jar.signed signed_cookie[:foobar] = { value: '123', expires: 1.day.ago } p signed_cookie[:foobar] ```
* | | Merge pull request #35086 from gsamokovarov/cleanup-whitelisting-refsGannon McGibbon2019-02-041-2/+2
|\ \ \ | |_|/ |/| | Cleanup the whitelisting references after #33145
| * | Cleanup the whitelisting references after #33145Genadi Samokovarov2019-02-031-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | During the development of #33145, I have named a few concepts in the code as `whitelisted`. We decided to stay away from the term and I adjusted most of the code afterwards, but here are the cases I forgot to change. I also found a case in the API guide that we could have cleaned up as well. [ci skip]
* | | Use consistent hash syntax for routesAndy Waite2019-02-011-14/+14
| |/ |/| | | | | [ci skip]
* | Fix doc of `ActionDispatch::SystemTestCase` [ci skip]yuuji.yaginuma2019-02-011-7/+7
| | | | | | | | | | | | * Fix broken format. * Need to specify driver to the first argument of `driven_by`. * `add_emulation` doesn't have `device` keyword. Ref: https://github.com/SeleniumHQ/selenium/blob/master/rb/lib/selenium/webdriver/chrome/options.rb#L142-L162
* | Merge pull request #35093 from rails/av-base-constructorAaron Patterson2019-01-301-1/+3
|\ \ | | | | | | Tighten up the AV::Base constructor
| * | Tighten up the AV::Base constructorAaron Patterson2019-01-291-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The AV::Base constructor was too complicated, and this commit tightens up the parameters it will take. At runtime, AV::Base is most commonly constructed here: https://github.com/rails/rails/blob/94d54fa4ab641a0ddeb173409cb41cc5becc02a9/actionview/lib/action_view/rendering.rb#L72-L74 This provides an AV::Renderer instance, a hash of assignments, and a controller instance. Since this is the common case for construction, we should remove logic from the constructor that handles other cases. This commit introduces special constructors for those other cases. Interestingly, most code paths that construct AV::Base "strangely" are tests.
* | | selenium-webdriver is not always required for system testingRyuta Kamizono2019-01-301-4/+2
| | | | | | | | | | | | | | | | | | But `NameError: uninitialized constant ActionDispatch::SystemTesting::Browser::Selenium` is pretty confused. I've little improved missing constant error to `NameError: uninitialized constant Selenium`.
* | | Fix system testing failureRyuta Kamizono2019-01-301-8/+11
|/ / | | | | | | https://travis-ci.org/rails/rails/jobs/486155626#L1317-L1335
* | Rename methods and update docsEileen Uchitelle2019-01-293-19/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a minor update to the named methods for the following: - s/desired_capabilities/capabilities - s/driver_options/capabilities Since they are all the same thing we should keep the name the same throughout the feature. Updated docs to match / be a little bit clearer Also updated the Gemfile for selenium-webdriver.
* | driver_option -> driver_optionsEdouard CHIN2019-01-292-7/+7
| |
* | Implement a way to add browser capabilities:Edouard CHIN2019-01-293-10/+40
|/ | | | | | | | | | | | | | | | | | * There is currently no way to define specific browser capabilities since our SystemTest driver override the `option` key [Ref](https://github.com/rails/rails/blob/a07d0680787ced3c04b362fa7a238c918211ac70/actionpack/lib/action_dispatch/system_testing/driver.rb#L35) This option key is used internally by selenium to add custom capabilities on the browser. Depending on the Browser, some option are allowed to be passed inside a hash, the driver takes care of setting whatever you passed on the driver option. An example [here](https://github.com/rails/rails/blob/a07d0680787ced3c04b362fa7a238c918211ac70/actionpack/lib/action_dispatch/system_testing/driver.rb#L35) where you are allowed to pass args such as `--no-sandbox` etc However this behavior was only meant for backward compatibility and as you can see it's deprecated. The non-deprecated behavior is to create a `<Driver>::Option` object containing all the capabilities we want. This is what we [currently do](https://github.com/rails/rails/blob/a07d0680787ced3c04b362fa7a238c918211ac70/actionpack/lib/action_dispatch/system_testing/browser.rb#L34-L36) when chrome or firefox are in headless mode. This PR allows to pass a block when calling `driven_by`, the block will be pased a `<Driver>::Option` instance. You can modify this object the way you want by adding any capabilities. The option object will be then passed to selenium. ```ruby driven_by :selenium, using: :chrome do |driver_option| driver_option.add_argument('--no-sandbox') driver_option.add_emulation(device: 'iphone 4') end ```
* Fixed a bug where the debug view does not show the error page properlyYuki Nishijima2019-01-242-11/+15
| | | | | | | | | | | | | | | | | | | | | There are two cases where the debug view does not show the error details properly: * When the cause is mapped to an HTTP status code the last exception is unexpectedly uwrapped * When the last error is thrown from a view template the debug view is not using the `rescues/template_error.html.erb` to generate the view Both the cases could be fixed by not unwrapping the exception. The only case where the exception should be unwrapped is when the last error is an `ActionView::Template::Error` object. In this case the HTTP status code is determined based on the cause. There are actually more wrapper exceptions that are intentionally thrown. However, there is a consistent pattern of setting the original message and original backtrace to the wrapper exception implemented, so the debug view will not lose the information about what went wrong eariler.
* Merge pull request #34994 from schuetzm/host-authorization-only-in-developmentRafael França2019-01-242-4/+4
|\ | | | | Recommend adding the requested domain to hosts whitelist only in deve…
| * Recommend adding the requested domain to hosts whitelist only in developmentMarc Schütz2019-01-232-4/+4
| |
* | Merge pull request #34952 from rails/template-stuffAaron Patterson2019-01-221-1/+8
|\ \ | |/ |/| Template Handler Refactoring
| * Only cache the view_context_class in one placeAaron Patterson2019-01-181-1/+8
| | | | | | | | | | | | | | This patch removes the instance writer of view_context_class. Subclasses may override it, but it doesn't need to be written. This also eliminates the need to cache the return value of the class level `view_context_class` method.
* | Remove secret_token rack env and cookie upgrade codeRafael Mendonça França2019-01-172-39/+1
| | | | | | | | Now that secret_token was removed all this code is now dead.
* | Remove deprecated methods in ActionDispatch::TestResponseRafael Mendonça França2019-01-171-27/+0
| | | | | | | | | | `#success?`, `missing?` and `error?` were deprecated in Rails 5.2 in favor of `#successful?`, `not_found?` and `server_error?`.
* | Revert "Don't handle params option in a special way in url_for helper"Rafael Mendonça França2019-01-162-0/+5
|/ | | | | | | | | | | This reverts commit e385e4678fc64be6e176c3bdac6641db9fe48d85. While this option was undocumented it exists to make possible to pass parameters to the route helpers that are reserved like `:domain`. While `url_for(domain: 'foo.com')` would generate a URL in the `foo.com` domain `url_for(params: { domain: 'foo.com' })` would generate a URL with `?domain=foo.com`.
* Allow using combine the Cache-Control `public` and `no-cache` headersyuuji.yaginuma2019-01-071-4/+6
| | | | | | | | | | | | | | | | | Since #30367, if `no-cache` includes Cache-Control headers, special keys like `public`, `must-revalidate` are ignored. But in my understanding, `public` still need in case of want to cache authenticated pages. The authenticated pages to be cacheable, but still authenticated for every user, need to specify the `Cache-Control: public, no-cache`. For keys other than `public`, I did not know the case where it was necessary to use it in combination with `no-cache`, so I fixed that can be used only for `public`. Ref: https://www.mnot.net/cache_docs/#CACHE-CONTROL Fixes #34780.
* Enable `Style/RedundantBegin` cop to avoid newly adding redundant begin blockRyuta Kamizono2018-12-214-24/+16
| | | | | | | | | | Currently we sometimes find a redundant begin block in code review (e.g. https://github.com/rails/rails/pull/33604#discussion_r209784205). I'd like to enable `Style/RedundantBegin` cop to avoid that, since rescue/else/ensure are allowed inside do/end blocks in Ruby 2.5 (https://bugs.ruby-lang.org/issues/12906), so we'd probably meets with that situation than before.
* Module#{define_method,alias_method,undef_method,remove_method} become public ↵Ryuta Kamizono2018-12-211-4/+4
| | | | | | since Ruby 2.5 https://bugs.ruby-lang.org/issues/14133
* Merge branch 'master' into host-authorizationEileen M. Uchitelle2018-12-171-6/+5
|\
| * Allow using parsed_body in ActionController::TestCaseTobias Bühlmann2018-12-161-6/+5
| | | | | | | | | | | | | | | | | | | | | | | | … by switching the initialzation of an appropriate response parser in `ActionDispatch::TestResponse` from eagerly to lazily. By doing so, the response parser can be correctly set for `ActionController::TestCase`, which doesn't include the content type header in the constructor but only sets it at a later time. Fixes #34676.
* | Introduce a guard against DNS rebinding attacksGenadi Samokovarov2018-12-155-41/+169
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ActionDispatch::HostAuthorization is a new middleware that prevent against DNS rebinding and other Host header attacks. By default it is included only in the development environment with the following configuration: Rails.application.config.hosts = [ IPAddr.new("0.0.0.0/0"), # All IPv4 addresses. IPAddr.new("::/0"), # All IPv6 addresses. "localhost" # The localhost reserved domain. ] In other environments, `Rails.application.config.hosts` is empty and no Host header checks will be done. If you want to guard against header attacks on production, you have to manually permit the allowed hosts with: Rails.application.config.hosts << "product.com" The host of a request is checked against the hosts entries with the case operator (#===), which lets hosts support entries of type RegExp, Proc and IPAddr to name a few. Here is an example with a regexp. # Allow requests from subdomains like `www.product.com` and # `beta1.product.com`. Rails.application.config.hosts << /.*\.product\.com/ A special case is supported that allows you to permit all sub-domains: # Allow requests from subdomains like `www.product.com` and # `beta1.product.com`. Rails.application.config.hosts << ".product.com"
* Upgrade Rubocop to 0.61.1 and fix offensesVinicius Stock2018-12-101-79/+79
|
* Expand metaprogramming for Symbol, Slash and Dot.Alberto Almagro2018-12-071-6/+7
| | | | | | | | | | | | | | This first started with moving type method inside `ActionDispatch::Journey::Nodes::Symbol`. `AD::Journey::Nodes::Symbol#type` was generated dynamically with an `each` block. While this is OK for classes like `AD::Journey::Nodes::Slash` or `AD::Journey::Nodes::Dot` which don't have further implementation, all other classes containing more logic have this method defined in their class body. This patch does the same in this case. On code review process @kamipo suggested to fully expand over metaprogramming for Slash and Dot classes, a topic on which I agree with him.
* Remove unnecessary variable routeAlberto Almagro2018-12-021-11/+2
| | | | | The variable `route` was only allocated to hold an object that was immediately returned. This patch removes that variable.
* Avoid extra array allocationsSamuel Cochran2018-11-291-2/+2
|
* Log exceptions atomicallySamuel Cochran2018-11-281-5/+8
| | | | | | When distributed over multiple logger calls the lines can become intermixed with other log statements. Combining them into a single logger call makes sure they always get logged together.
* Raise an error on root route naming conflicts.Gannon McGibbon2018-11-201-4/+2
| | | | | Raises an ArgumentError when multiple root routes are defined in the same context instead of assigning nil names to subsequent roots.
* Merge pull request #20865 from colavitam/only-except-behaviorRafael Mendonça França2018-11-191-2/+8
|\ | | | | | | :only and :except are now chained for routing resource(s)
| * :only and :except are now chained for routing resource(s)Michael Colavita2015-07-131-3/+9
| | | | | | | | | | | | | | | | | | | | | | Allow chaining the :only and :except options for routing resource(s). Previously, the following yielded routes for both show and destroy: resource :account, :only => [:show, :destroy], :except => :destroy This now yields only the show action. This chaining can be useful for passing optional :except options to code that makes use of the :only option (e.g. for a gem with its own routing methods).
* | Remove unused `Journey::Router::RoutingError`yuuji.yaginuma2018-11-181-3/+0
| | | | | | | | `Journey::Router::RoutingError` is no longer used since db06d128262b49c8b02e153cf95eb46f4eff364b.
* | Allow rescue from parameter parse errorsGannon McGibbon2018-11-134-7/+21
| | | | | | | | [Gannon McGibbon + Josh Cheek]