aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Avoid slowing down AR object initializationGodfrey Chan2014-05-221-0/+1
| | | | | | | | | | | | | | | | | | 2d73f5a forces AR to enter the `define_attribute_methods` method whenever it instantiate a record from the `init_with` entry point. This is a potential performance hotspot, because `init_with` is called from all `find*` family methods, and `define_attribute_methods` is slow because it tries to acquire a lock on the mutex everytime it is entered. By using [DCL](http://en.wikipedia.org/wiki/Double-checked_locking), we can avoid grabbing the lock most of the time when the attribute methods are already defined (the common case). This is made possible by the fact that reading an instance variable is an atomic operation in Ruby. Credit goes to Aaron Patterson for pointing me to DCL and filling me in on the atomicity guarantees in Ruby. [*Godfrey Chan*, *Aaron Patterson*]
* Fixed serialization for records with an attribute named `format`.Godfrey Chan2014-05-225-0/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * * * This bug can be triggered when serializing record R (the instance) of type C (the class), provided that the following conditions are met: 1. The name of one or more columns/attributes on C/R matches an existing private method on C (e.g. those defined by `Kernel`, such as `format`). 2. The attribute methods have not yet been generated on C. In this case, the matching private methods will be called by the serialization code (with no arguments) and their return values will be serialized instead. If the method requires one or more arguments, it will result in an `ArgumentError`. This regression is introduced in d1316bb. * * * Attribute methods (e.g. `#name` and `#format`, assuming the class has columns named `name` and `format` in its database table) are lazily defined. Instead of defining them when a the class is defined (e.g. in the `inherited` hook on `ActiveRecord::Base`), this operation is deferred until they are first accessed. The reason behind this is that is defining those methods requires knowing what columns are defined on the database table, which usually requires a round-trip to the database. Deferring their definition until the last-possible moment helps reducing unnessary work, especially in development mode where classes are redefined and throw away between requests. Typically, when an attribute is first accessed (e.g. `a_book.format`), it will fire the `method_missing` hook on the class, which triggers the definition of the attribute methods. This even works for methods like `format`, because calling a private method with an explicit receiver will also trigger that hook. Unfortunately, `read_attribute_for_serialization` is simply an alias to `send`, which does not respect method visibility. As a result, when serializing a record with those conflicting attributes, the `method_missing` is not fired, and as a result the attribute methods are not defined one would expected. Before d1316bb, this is negated by the fact that calling the `run_callbacks` method will also trigger a call to `respond_to?`, which is another trigger point for the class to define its attribute methods. Therefore, when Active Record tries to run the `after_find` callbacks, it will also define all the attribute methods thus masking the problem. * * * The proper fix for this problem is probably to restrict `read_attribute_for_serialization` to call public methods only (i.e. alias `read_attribute_for_serialization` to `public_send` instead of `send`). This however would be quite risky to change in a patch release and would probably require a full deprecation cycle. Another approach would be to override `read_attribute_for_serialization` inside Active Record to force the definition of attribute methods: def read_attribute_for_serialization(attribute) self.class.define_attribute_methods send(attribute) end Unfortunately, this is quite likely going to cause a performance degradation. This patch therefore restores the behaviour from the 4-0-stable branch by explicitly forcing the class to define its attribute methods in a similar spot (when records are initialized). This should not cause any extra roundtrips to the database because the `@columns` should already be cached on the class. Fixes #15188.
* Merge pull request #15257 from jetthoughts/remove_redundant_codeRafael Mendonça França2014-05-221-1/+0
|\ | | | | Removed not used code
| * Removed not-used codePaul Nikitochkin2014-05-221-1/+0
| |
* | fix test from 7537057888d2d63c3b6c2019d5828bc445fbd6c9Aaron Patterson2014-05-221-0/+1
| |
* | Don't recurse infinitely when calling non-existent method on superKris Kemper2014-05-221-0/+12
| |
* | Merge pull request #15259 from sgrif/sg-push-limit-to-typesRafael Mendonça França2014-05-227-73/+67
|\ \ | | | | | | Push limit to type objects
| * | Push limit to type objectsSean Griffin2014-05-227-73/+67
|/ / | | | | | | | | Columns and injected types no longer have any conditionals based on the format of SQL type strings! Hooray!
* | Merge pull request #15256 from sgrif/sg-push-precision-to-typesRafael Mendonça França2014-05-226-22/+27
|\ \ | | | | | | Push precision to type objects
| * | Push precision to type objectsSean Griffin2014-05-226-22/+27
|/ /
* | Merge pull request #15250 from sgrif/sg-push-scale-to-typesRafael Mendonça França2014-05-227-20/+27
|\ \ | | | | | | Push scale to type objects
| * | Push scale to type objectsSean Griffin2014-05-227-20/+27
| | | | | | | | | | | | | | | | | | Ideally types will be usable without having to specify a sql type string, so we should keep the information related to parsing them on the adapter or another object.
* | | Merge pull request #15254 from DNNX/formatter-refactoring-3Rafael Mendonça França2014-05-221-4/+4
|\ \ \ | |/ / |/| | Rename `stack` to `queue`
| * | Rename `stack` to `queue`Viktar Basharymau2014-05-221-4/+4
| | | | | | | | | | | | | | | | | | | | | Because it is used as a queue (FIFO), not as a stack (LIFO). * http://en.wikipedia.org/wiki/Stack_(abstract_data_type) * http://en.wikipedia.org/wiki/Queue_(data_structure)
* | | Merge pull request #15239 from croaky/dc-sg-move-extract-precision-to-typesRafael Mendonça França2014-05-227-19/+15
|\ \ \ | | | | | | | | Move `extract_precision` onto type objects
| * | | Move `extract_precision` onto type objectsDan Croak and Sean Griffin2014-05-227-19/+15
|/ / /
* | | Merge pull request #15252 from DNNX/formatter-refactoring-2Rafael Mendonça França2014-05-221-1/+1
|\ \ \ | |_|/ |/| | Remove unnecessary `Hash#to_a` call
| * | Remove unnecessary `Hash#to_a` callViktar Basharymau2014-05-221-1/+1
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Inspired by https://github.com/rails/rails/commit/931ee4186b877856b212b0085cd7bd7f6a4aea67 ```ruby def stat(num) start = GC.stat(:total_allocated_object) num.times { yield } total_obj_count = GC.stat(:total_allocated_object) - start puts "#{total_obj_count / num} allocations per call" end h = { 'x' => 'y' } stat(100) { h. each { |pair| pair } } stat(100) { h.to_a.each { |pair| pair } } __END__ 1 allocations per call 2 allocations per call ```
* | Merge pull request #15247 from DNNX/activesupport-formatter-refactoringRafael Mendonça França2014-05-221-1/+1
|\ \ | | | | | | Use `break` instead of `next` in AD::Journey::Formatter#match_route
| * | Use `break` instead of `next` in AD::Journey::Formatter#match_routeViktar Basharymau2014-05-221-1/+1
| |/ | | | | | | | | | | The array is sorted in descending order, so there is no point in iterating further if we met a negative item - all the rest will be negative too.
* | Merge pull request #15249 from sgrif/sg-register-types-in-adapterRafael Mendonça França2014-05-224-84/+63
|\ \ | | | | | | Use the generic type map for all PG type registrations
| * | Use the generic type map for all PG type registrationsSean Griffin2014-05-224-84/+63
| | | | | | | | | | | | | | | | | | | | | We're going to want all of the benefits of the type map object for registrations, including block registration and real aliasing. Moves type name registrations to the adapter, and aliases the OIDs to the named types
* | | Merge pull request #15248 from sgrif/sg-additional-type-map-argsRafael Mendonça França2014-05-225-12/+29
|\ \ \ | | | | | | | | Allow additional arguments to be used during type map lookups
| * | | Allow additional arguments to be used during type map lookupsSean Griffin2014-05-225-12/+29
| |/ / | | | | | | | | | | | | | | | | | | | | | Determining things like precision and scale in postgresql will require the given blocks to take additional arguments besides the OID. - Adds the ability to handle additional arguments to `TypeMap` - Passes the column type to blocks when looking up PG types
* | | Merge pull request #15242 from JuanitoFatas/guides-anchorRafael Mendonça França2014-05-221-2/+1
|\ \ \ | | | | | | | | [Guides] Do not gsub non ASCII characters in header anchor.
| * | | Do not gsub non ASCII characters in header anchor.Juanito Fatas2014-05-221-2/+1
| | | |
* | | | Fix the hash key being delete from envRafael Mendonça França2014-05-221-1/+1
| |_|/ |/| | | | | | | | It was changed by mistake at c5d64b2b86aa42f57881091491ee289b3c489c7e.
* | | pg test, extract tsvector tests into `postgresql/full_text_test.rb`Yves Senn2014-05-222-27/+31
| | |
* | | pg test, move timestamp tests over to `postgresql/timestamp_test.rb`.Yves Senn2014-05-222-43/+53
| | |
* | | pg test, extract network address type tests into separate file.Yves Senn2014-05-223-51/+79
| | |
* | | pg test, remove unused code.Yves Senn2014-05-221-3/+0
|/ /
* | Merge pull request #15101 from tgxworld/improve_performance_of_building_uriAndrew White2014-05-223-13/+16
|\ \ | |/ |/| Form full URI as string to be parsed in Rack::Test.
| * Form full URI as string to be parsed in Rack::Test.Guo Xiang Tan2014-05-213-13/+16
|/ | | | There are performance gains to be made by avoiding URI setter methods.
* Merge pull request #14803 from kuldeepaggarwal/null_relation_sum_fixMatthew Draper2014-05-223-3/+66
|\ | | | | | | Fixed a problem where `sum` used with a `group` was not returning a Hash.
| * Fixed a problem where `sum`, `size`, `average`, `minimum` and `maximum` usedKuldeep Aggarwal2014-05-153-3/+66
| | | | | | | | with a grouping was not returning a Hash.
* | Update CHANGELOG.mdGuillermo Iguaran2014-05-211-4/+4
| | | | | | Most recent change should be moved to the top
* | Merge pull request #15215 from JohnKellyFerguson/rename-posts-to-articlesGuillermo Iguaran2014-05-2120-655/+658
|\ \ | | | | | | Documentation: Rename Posts to Articles
| * | Rename Posts to Articles in Guides, continuation of 2d446e77 / #13774 [ci skip]John Kelly Ferguson2014-05-2120-655/+658
|/ /
* | Merge pull request #15237 from sgrif/sg-move-extract-scaleRafael Mendonça França2014-05-214-8/+7
|\ \ | | | | | | Move extract_scale to decimal type
| * | Move extract_scale to decimal typeSean Griffin2014-05-214-8/+7
| | | | | | | | | | | | | | | | | | The only type that has a scale is decimal. There's a special case where decimal columns with 0 scale are type cast to integers if the scale is not specified. Appears to only affect schema dumping.
* | | Merge pull request #15218 from sgrif/sg-move-oid-typesRafael Mendonça França2014-05-2123-353/+532
|\ \ \ | | | | | | | | Move PG OID types to their own files
| * | | Move PG OID types to their own filesSean Griffin2014-05-2123-353/+532
| | | | | | | | | | | | | | | | | | | | | | | | | | | | As we promote these classes to first class concepts, these classes are starting to gain enough behavior to warrant being moved into their own files. Many of them will become quite large as we move additional behavior to the type objects.
* | | | drop hash allocations during matchAaron Patterson2014-05-211-2/+2
| | | |
* | | | fewer object allocations and method calls during route matchAaron Patterson2014-05-211-5/+5
| | | |
* | | | middle variable is never used, so rmAaron Patterson2014-05-213-16/+16
| | | |
* | | | use the existing constant rather than the hardcoded stringAaron Patterson2014-05-211-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | so tht I don't go totally insane with THsi crazy hash driven development. why is Everything a hash? Why do people think hashes in hashes with random keys is a Good API? You can't find things or deprecate them or control access whatsoever, you just have to hope that everyone is like "oh, you want to change that? that's cool! we all know it's hashes so go for it!" The End.
* | | | remove dead code. @klass isn't used anymoreAaron Patterson2014-05-211-1/+0
| | | |
* | | | we can just use Ruby hereAaron Patterson2014-05-211-2/+2
| | | |
* | | | reuse path formatter from the non-optimized path.Aaron Patterson2014-05-213-39/+3
| | | | | | | | | | | | | | | | | | | | The optimized and non-optimized path share more code now without significant performance degretation
* | | | reduce object allocationsAaron Patterson2014-05-211-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Example: x = [1,2,3,4] y = [3,2,1] def test x, y hash = {} x.zip(y) { |k,v| hash[k] = v } hash end def test2 x, y Hash[x.zip(y)] end def test3 x, y x.zip(y).each_with_object({}) { |(k,v),hash| hash[k] = v } end def stat num start = GC.stat(:total_allocated_object) num.times { yield } total_obj_count = GC.stat(:total_allocated_object) - start puts "#{total_obj_count / num} allocations per call" end stat(100) { test(x,y) } stat(100) { test2(x,y) } stat(100) { test3(x,y) } __END__ 2 allocations per call 7 allocations per call 8 allocations per call
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-07 18:14:13 +0000