Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | break each polymorphic type to it's own method | Aaron Patterson | 2014-05-12 | 1 | -24/+81 | |
| | ||||||
* | remove dead code | Aaron Patterson | 2014-05-12 | 1 | -12/+0 | |
| | ||||||
* | only test `persisted?` on the record once | Aaron Patterson | 2014-05-12 | 1 | -3/+7 | |
| | | | | | also avoid the case of calling `persisted?` on something that doesn't respond to it | |||||
* | only send options if there are any options | Aaron Patterson | 2014-05-12 | 1 | -5/+7 | |
| | ||||||
* | remove should_pop variable | Aaron Patterson | 2014-05-12 | 1 | -4/+1 | |
| | ||||||
* | only pop records when we get an array | Aaron Patterson | 2014-05-12 | 1 | -6/+7 | |
| | ||||||
* | remove unnecessary array allocations | Coraline Ada Ehmke + Aaron Patterson | 2014-05-12 | 1 | -5/+1 | |
| | ||||||
* | simplified route method name generation | Coraline Ada Ehmke + Aaron Patterson | 2014-05-12 | 1 | -27/+17 | |
| | ||||||
* | small refactoring | Coraline Ada Ehmke + Aaron Patterson | 2014-05-12 | 1 | -6/+2 | |
| | ||||||
* | pulling helpermethods up before refactoring | Coraline Ada Ehmke + Aaron Patterson | 2014-05-12 | 1 | -20/+46 | |
| | ||||||
* | push all is_a tests up, always pass options to the named route | Aaron Patterson | 2014-05-12 | 1 | -28/+27 | |
| | ||||||
* | pull nil checks up | Aaron Patterson | 2014-05-12 | 1 | -3/+5 | |
| | ||||||
* | skip another Array is_a check | Aaron Patterson | 2014-05-12 | 1 | -4/+6 | |
| | ||||||
* | only do the Array is_a check once | Aaron Patterson | 2014-05-12 | 1 | -4/+3 | |
| | ||||||
* | we always convert this variable to a list, so we can remove the unconversion | Aaron Patterson | 2014-05-12 | 1 | -1/+0 | |
| | ||||||
* | record as a hash is not possible, so rm branch | Aaron Patterson | 2014-05-12 | 1 | -2/+0 | |
| | ||||||
* | stop popping in build_named_route_call | Aaron Patterson | 2014-05-12 | 1 | -4/+3 | |
| | ||||||
* | remove extract_record since we have extracted the list | Aaron Patterson | 2014-05-12 | 1 | -12/+1 | |
| | ||||||
* | always pass a list to build_named_route_call | Aaron Patterson | 2014-05-12 | 1 | -10/+19 | |
| | ||||||
* | add a test for passing two models to the polymorphic_url method | Aaron Patterson | 2014-05-12 | 1 | -1/+1 | |
| | ||||||
* | reduce conditionals in url_for | Aaron Patterson | 2014-05-12 | 1 | -3/+6 | |
| | ||||||
* | extract route key translation to a method and reuse it | Aaron Patterson | 2014-05-12 | 1 | -17/+14 | |
| | ||||||
* | make the module version quack the same as the instance | Aaron Patterson | 2014-05-12 | 1 | -3/+10 | |
| | | | | this way we can stop checking respond_to on every call. | |||||
* | Hash#except is expensive and this is a hotspot, so use a dup + delete | Aaron Patterson | 2014-05-12 | 1 | -1/+2 | |
| | ||||||
* | Always use the provided port for protocol relative urls | Andrew White | 2014-05-11 | 1 | -0/+2 | |
| | | | | | There may be situations where you need to tunnel SSL connections over port 80 so we shouldn't remove it if it has been explicitly provided. | |||||
* | ActionDispatch::Http::URL.normalize_port should not strip port for protocol ↵ | Guilherme Cavalcanti | 2014-05-11 | 1 | -2/+0 | |
| | | | | relative URL. | |||||
* | Merge branch 'master' of github.com:rails/docrails | Vijay Dev | 2014-05-10 | 1 | -1/+22 | |
|\ | ||||||
| * | copy edits [ci skip] | Vijay Dev | 2014-05-10 | 1 | -8/+8 | |
| | | ||||||
| * | [ci skip] doc Http::Headers methods | schneems | 2014-05-09 | 1 | -1/+17 | |
| | | ||||||
| * | copy edits [ci skip] | Vijay Dev | 2014-05-08 | 1 | -2/+1 | |
| | | ||||||
| * | [ci skip] document ActionDispatch::HTTP::Headers | schneems | 2014-05-07 | 1 | -0/+6 | |
| | | ||||||
* | | skip dealing with params if none are provided | Aaron Patterson | 2014-05-09 | 1 | -4/+11 | |
| | | | | | | | | | | | | | | | | | | | | | | This lets us avoid 1. A slow call to Hash#slice 2. An is_a? test 3. Extra hash allocations (from slice) 4. String allocations etc. | |||||
* | | don't mutate the options hash, so we don't have to dup | Aaron Patterson | 2014-05-09 | 1 | -4/+3 | |
| | | | | | | | | avoids extra hash allocations on each call | |||||
* | | use unless and || since these options are boolean | Aaron Patterson | 2014-05-09 | 1 | -1/+1 | |
| | | ||||||
* | | HTTP::Headers#key? correctly converts | schneems | 2014-05-07 | 1 | -1/+3 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously if you were looking for a given key, the header may incorrectly tell you that it did not exist even though it would return a valid value: ```ruby env = { "CONTENT_TYPE" => "text/plain" } headers = ActionDispatch::Http::Headers.new(env) headers["Content-Type"] # => "text/plain" headers.key?("Content-Type") # => false ``` This PR fixes that behavior by converting the key before checking for presence | |||||
* | | Do not use short-circuit return | Rafael Mendonça França | 2014-05-04 | 1 | -2/+3 | |
| | | ||||||
* | | Merge pull request #11166 from xavier/callable_constraint_verification | Rafael Mendonça França | 2014-05-04 | 1 | -0/+7 | |
|\ \ | | | | | | | | | | | | | | | | | | | Callable route constraint verification Conflicts: actionpack/CHANGELOG.md | |||||
| * | | Verify that route constraints respond to the expected messages instead of ↵ | Xavier Defrang | 2013-06-28 | 1 | -0/+7 | |
| | | | | | | | | | | | | silently failing to enforce the constraint | |||||
* | | | Use #include? instead of #any?, make it simpler | Akshay Vishnoi | 2014-05-03 | 1 | -1/+1 | |
| | | | ||||||
* | | | passing a nil in the polymorphic array is not supported. remove nils before ↵ | Aaron Patterson | 2014-05-02 | 1 | -0/+4 | |
| | | | | | | | | | | | | you call the method | |||||
* | | | passing a nil should always raise an ArgumentError | Aaron Patterson | 2014-05-02 | 1 | -1/+0 | |
| | | | ||||||
* | | | Tiny follow up to #14915 [ci skip] | Robin Dupret | 2014-05-02 | 1 | -19/+18 | |
| | | | ||||||
* | | | only add the optiosn if they are not empty | Aaron Patterson | 2014-05-01 | 1 | -1/+3 | |
| | | | | | | | | | | | | | | | | | | I think this is wrong, but it gets the build passing for now. We should always add options, but we need to make more guarantees about how the underlying url helper is called | |||||
* | | | always pass options to the _url method | Aaron Patterson | 2014-05-01 | 1 | -3/+1 | |
| | | | ||||||
* | | | never merge url options in to the first data hash | Aaron Patterson | 2014-05-01 | 1 | -1/+1 | |
| | | | | | | | | | | | | | | | if you want options, don't mix them with the first hash, just pass them all in with the second hash | |||||
* | | | Merge pull request #14915 from juanpastas/patch-1 | Rafael Mendonça França | 2014-05-01 | 1 | -20/+37 | |
|\ \ \ | | | | | | | | | Update mapper.rb | |||||
| * | | | [skip ci] Document: required `via` option in `match` routing method. | Juan David Pastas | 2014-04-30 | 1 | -20/+37 | |
| | |/ | |/| | ||||||
* | | | Merge pull request #12651 from cespare/ipv6-remote-ip-fixes | Rafael Mendonça França | 2014-05-01 | 1 | -1/+1 | |
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | Make remote_ip detection properly handle private IPv6 addresses Conflicts: actionpack/CHANGELOG.md | |||||
| * | | | Make remote_ip detection properly handle private IPv6 addresses | Caleb Spare | 2013-10-26 | 1 | -1/+1 | |
| | | | | | | | | | | | | | | | | Fixes #12638. | |||||
* | | | | avoid calling extract_record multiple times | Aaron Patterson | 2014-04-30 | 1 | -3/+2 | |
| | | | |