550a371218c9de50b2dfe1865dc99b878b514929 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/5e1668eadd6664e4d30fadab5333b7b0bc192b14 5e1668eadd6664e4d30fadab5333b7b0bc192b14

2009-04-03T10:13:31-07:00

Bump version to 2.12.0, add 2.12.0 release notes 947c977f39be7f3e7c298e9cd1b8bce38ba510d5 Jeremy Evans code@jeremyevans.net c879600684489e56ae8b37240cb21ad9d063e20e Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/550a371218c9de50b2dfe1865dc99b878b514929 550a371218c9de50b2dfe1865dc99b878b514929

2009-04-03T09:17:37-07:00

Add RDoc for the SEQUEL_NO_CORE_EXTENSIONS and SEQUEL_NO_ASSOCIATIONS options 945188ebfd6b65458d62ff512ee5b2c9b10198ed Jeremy Evans code@jeremyevans.net d2f3b9b5787ddad05cba9586206420f9f9614ac8 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/c879600684489e56ae8b37240cb21ad9d063e20e c879600684489e56ae8b37240cb21ad9d063e20e

2009-04-02T14:47:02-07:00

Cleanup Rakefile and use new rdoc format for website The website now will have 3 separate rdoc sections, one for core/model, one for adapters, and one for extensions/plugins. I think this will make the RDoc less cluttered and more approachable. The Rakefile website and website_rdoc tasks are now able to generate a fully functional local version of the site. In order for this to work, I changed all of the internal website links to relative links instead of the absolute links used previously. You can now use the Rakefile without rspec installed, though obviously you can't use the spec tasks. I refactored the Rakefile to use lambdas to reduce the amount of work done just by parsing the Rakefile, by only running code inside the tasks that need it. I removed the install_no_docs task. If anyone really wants it I can add it back. I made the .gitignore patterns absolute. I don't think this will cause any problems, and with normal usage of the sequel repository, there should be no effect. This commit changes the website task to only generate the base website, and adds a website_rdoc task to generate the 3 rdoc pages. You now need to use the website_rf task to update the sequel.rubyforge.org site. As I'm the only person that does that, I don't expect any complaints. I'm also switching the website upload to use rsync instead of scp, to improve the speed. f2d26d3600b19601e28ca60db71928f82368b5d7 Jeremy Evans code@jeremyevans.net c3e312359b6361dbb0e2b5f9cb81e9f94546fb71 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/d2f3b9b5787ddad05cba9586206420f9f9614ac8 d2f3b9b5787ddad05cba9586206420f9f9614ac8

2009-04-02T12:51:12-07:00

Fix bug in Model.def_dataset_method where it checked for instance methods existence instead of class method existence 17850cba3b2c0d714e4fae25c82e9885b6f57ba7 Jeremy Evans code@jeremyevans.net 76d7e74136cf5320837acd4b00e6b6218b41a7e5 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/c3e312359b6361dbb0e2b5f9cb81e9f94546fb71 c3e312359b6361dbb0e2b5f9cb81e9f94546fb71

2009-04-02T12:13:18-07:00

Fix rollback bug in Database#transaction in the PostgreSQL adapter When I changed the transaction method to only use savepoints when they were specifically requested, I broke the transaction method when Sequel::Rollback was raised. Scaringly, this wasn't caught be the specs or the app tests I run on a regular basis. I added an adapter spec for this, and also added an integration test suite related to database transactions, mostly copied from the PostgreSQL adapter spec. Hopefully this will ensure that breakage is caught earlier in the future (ideally before I commit). abc296556467ad1d990b7da53232e224f6e9c48b Jeremy Evans code@jeremyevans.net cabdbf40489013a0c17e578d6bb63bf4c47ee831 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/76d7e74136cf5320837acd4b00e6b6218b41a7e5 76d7e74136cf5320837acd4b00e6b6218b41a7e5

2009-04-02T11:03:40-07:00

Fix some bugs in the validation_helpers plugin when options are used 5b167b5b8bc46bc71f7327753e6b7309c1d3eec4 Jeremy Evans code@jeremyevans.net d261b1efbca1351f7bde1cdca0370959d3cb1a27 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/cabdbf40489013a0c17e578d6bb63bf4c47ee831 cabdbf40489013a0c17e578d6bb63bf4c47ee831

2009-04-02T09:00:04-07:00

Fix Database#blank_object for objects that don't respond to empty? 6ccdb7cb8ab9a4295278030f7355f76ae67f69e6 Jeremy Evans code@jeremyevans.net 8364e2ea423b44811dbe5a63e589342686ef95ca Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/d261b1efbca1351f7bde1cdca0370959d3cb1a27 d261b1efbca1351f7bde1cdca0370959d3cb1a27

2009-04-01T16:24:59-07:00

Update bullet points on front page, remove deprecated example, some other cleanup 0e04c00709c73fc2bce4d5514a5665c8bdbe8be4 Jeremy Evans code@jeremyevans.net 6b85f2ab43d8b15266936eb5e3f3651cf137193a Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/8364e2ea423b44811dbe5a63e589342686ef95ca 8364e2ea423b44811dbe5a63e589342686ef95ca

2009-04-01T16:24:16-07:00

Remove sequel_versioned plugin link, since it is broken 2f9054ef95ff5287b8dd3065d1a3fef12a486670 Jeremy Evans code@jeremyevans.net c8328a9b102acabfafdb5e2deddf94100b7f4242 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/6b85f2ab43d8b15266936eb5e3f3651cf137193a 6b85f2ab43d8b15266936eb5e3f3651cf137193a

2009-04-01T13:21:53-07:00

2009-04-01T13:19:00-07:00

Remove internal use of Symbol#like and Symbol#qualify 4811a535a6e2c4878c100df55e53d73897a0f666 Jeremy Evans code@jeremyevans.net 32f8d292430173cfd9f687a404fa45428e779bf3 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/c8328a9b102acabfafdb5e2deddf94100b7f4242 c8328a9b102acabfafdb5e2deddf94100b7f4242

2009-04-01T12:45:33-07:00

Fix RDoc for Sequel::Deprecation a1ec2f76214dabdc0a3c03721eb2f39f393fd452 Jeremy Evans code@jeremyevans.net 896dc72d8fd1463f26e1674998621fc3b83c5b7f Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/32f8d292430173cfd9f687a404fa45428e779bf3 32f8d292430173cfd9f687a404fa45428e779bf3

2009-03-31T18:12:59-07:00

Deprecate Java::JavaSQL::Timestamp#usec The JDBC adapter has returned Java::JavaSQL::Timestamp values as Time or DateTime for a few versions now, so we don't need to add a method to the Java class. 89b25d172967dea88179b679e14b69c8e4c8d7ec Jeremy Evans code@jeremyevans.net 2433bb38b84e1f8b5755896fc9fb651dfd2d117f 0cf1951fe77af033acc617eb3e47c0acef594427 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/896dc72d8fd1463f26e1674998621fc3b83c5b7f 896dc72d8fd1463f26e1674998621fc3b83c5b7f

2009-03-30T20:48:10-07:00

Merge dusty's method renaming patch with some updated documentation Conflicts: lib/sequel/adapters/shared/mysql.rb d8ca6b30844add70ab4872d4b8eaac99b57c5a65 Jeremy Evans code@jeremyevans.net 72909390efe8d3bca5ee9b5f27749c50b085ff15 Dusty Doris git@dusty.name http://github.com/jeremyevans/sequel/commit/0cf1951fe77af033acc617eb3e47c0acef594427 0cf1951fe77af033acc617eb3e47c0acef594427

2009-03-30T19:52:48-07:00

rename methods to be more descriptive of intent renamed multi_insert_update to on_duplicate_key_update renamed multi_insert_ignore to insert_ignore SUPPORT for ON DUPLICATE KEY UPDATE and INSERT IGNORE for mysql datasets d8ca6b30844add70ab4872d4b8eaac99b57c5a65 Dusty Doris git@dusty.name 0ea83550c1b3186afd8cb3a493d24795dca0012e Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/2433bb38b84e1f8b5755896fc9fb651dfd2d117f 2433bb38b84e1f8b5755896fc9fb651dfd2d117f

2009-03-30T13:46:51-07:00

Fix some RDoc for multi_insert_ignore and multi_insert_update in the shared MySQL adapter d0fea49259fd2018eb6834469fe5a9a41e4c6866 Jeremy Evans code@jeremyevans.net 72909390efe8d3bca5ee9b5f27749c50b085ff15 e318b2d9b10ab1b3d979901d53034003a38100e4 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/0ea83550c1b3186afd8cb3a493d24795dca0012e 0ea83550c1b3186afd8cb3a493d24795dca0012e

2009-03-30T13:44:05-07:00

Merge commit 'dusty/master' a44a0746cc32c96fad689cf8a584bb500fb6793b Jeremy Evans code@jeremyevans.net 930f2b705ac548efde2ce042cf263553b948d815 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/72909390efe8d3bca5ee9b5f27749c50b085ff15 72909390efe8d3bca5ee9b5f27749c50b085ff15

2009-03-30T13:32:57-07:00

Fix deprecated use of Dataset#uniq in the oracle specs, thanks jacaetevha 7b68316f323dbc9a7c65ef997a7a0b68a058c3a0 Jeremy Evans code@jeremyevans.net 242bd8e257816bd16a73d832eaeea4371aa61ee7 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/930f2b705ac548efde2ce042cf263553b948d815 930f2b705ac548efde2ce042cf263553b948d815

2009-03-30T12:47:25-07:00

Fix Model.[] optimization introduced in 2.11.0 for databases that don't use LIMIT, thanks jacaetevha ffc9bfe15f1c588b17b1ed0589db3b5dfd2a7dff Jeremy Evans code@jeremyevans.net 3066d5d1750b4861bad1df246edc207fe9786936 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/242bd8e257816bd16a73d832eaeea4371aa61ee7 242bd8e257816bd16a73d832eaeea4371aa61ee7

2009-03-30T12:10:36-07:00

Use Model::Errors#add in the validation plugins c8a2e96c1c94ba9e16817d6447f174410eb74ed3 Jeremy Evans code@jeremyevans.net 4f5aebf41b62aee42e30a35b50f2f40e87b851f9 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/3066d5d1750b4861bad1df246edc207fe9786936 3066d5d1750b4861bad1df246edc207fe9786936

2009-03-30T12:05:19-07:00

RDoc updates in remaining model files, small fixes to reduce memory use The majority of the lines changed are due to using constants instead of creating new string and regexp objects in every call (which need to be garbage collected). This also fixes Model::Errors#count to not create full_messages, so it should be faster. This should be the last of the RDoc related commits. 819b86c37891b782b40f9555e7754b5a605ef905 Jeremy Evans code@jeremyevans.net fd618bab41035c5f8575f73e3ad253912fb19245 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/4f5aebf41b62aee42e30a35b50f2f40e87b851f9 4f5aebf41b62aee42e30a35b50f2f40e87b851f9

2009-03-30T11:19:45-07:00

RDoc updates in associations.rb a5b0f4837697a870b809e56d0b024585b2df3c9d Jeremy Evans code@jeremyevans.net a24f8d14e49c8a3e5e35aa212285f3e28b454d58 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/fd618bab41035c5f8575f73e3ad253912fb19245 fd618bab41035c5f8575f73e3ad253912fb19245

2009-03-30T08:08:27-07:00

Fix some deprecated use of multi_insert in the mysql specs to import aebab8219c6166cd98642a021a08162d3b4148fc Jeremy Evans code@jeremyevans.net 6749f16490fc57e4fb3344f0b1c5376b76a70689 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/a24f8d14e49c8a3e5e35aa212285f3e28b454d58 a24f8d14e49c8a3e5e35aa212285f3e28b454d58

2009-03-29T17:48:33-07:00

Minor code changes in and major RDoc updates to model.rb, model/base.rb, model/plugins.rb Change the @models cache under Sequel to use a Model::ANONYMOUS_MODEL_CLASSES constant instead. Rename DATASET_METHOD_RE to NORMAL_METHOD_NAME_REGEXP, better reflecting current usage and purpose. Add SETTER_METHOD_REGEXP instead of creating a new Regexp every time setter_methods is used, should save memory. Simplify create to just call new with the args and block and then call save. Allow def_dataset_method with a block to work without a dataset, since it stores the information to apply to a future dataset given to set_dataset anyway. Use the ternary operator instead of if/else inside save to save a few lines. Make destroy for model datasets faster by not incrementing a counter. 536b2bbb4f5a272f913a5324b49dbcfe6810ff94 Jeremy Evans code@jeremyevans.net 6e4043d992b4fdf0b0d3b046bcaf414431dfa9a6 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/6749f16490fc57e4fb3344f0b1c5376b76a70689 6749f16490fc57e4fb3344f0b1c5376b76a70689

2009-03-29T11:59:59-07:00

Don't use the model association plugin if SEQUEL_NO_ASSOCIATIONS constant or environment variable is defined If you don't need associations for your application, Sequel won't force them on you any more. This should cut down on memory usage when associations are not needed. This commit removes the little remaining coupling between the base model plugin and the caching, validation class methods, and association plugins. This commit moves the association_reflection.rb file into associations.rb. Associations could be just like any other plugin if I wanted to change the namespace hierarchy, but I don't think it is worth it. Almost everyone using models is going to want associations of some kind, this is for the rare soul that doesn't want it. This commit allows you to have associations for only certain classes by doing: require 'sequel/model/associations' MyModel.plugin Sequel::Model::Associations This commit probably causes some ruby -w issues because of the removal of the code that sets @transform to nil. Transforms are going away soon, so I'm willing to make this change now, so I don't forget to make it later. In order to fix a ruby -w issue in validation class methods plugin, the skip_superclass_validations? was added to the plugin's class methods. 59782adfa6382bbae1ebf133cab7d8e727eac6ca Jeremy Evans code@jeremyevans.net f91d716f69f78b5aba129e8af1c3c9521dee55f4 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/6e4043d992b4fdf0b0d3b046bcaf414431dfa9a6 6e4043d992b4fdf0b0d3b046bcaf414431dfa9a6

2009-03-26T21:54:48-07:00

Don't require core_sql if SEQUEL_NO_CORE_EXTENSIONS constant or environment variable is defined This commit enables the user to have Sequel not add any extensions to the core classes. Sequel should still work correctly without the core extensions, otherwise there is a bug. All the core extensions give you is a nice DSL, they are not required. 4cba1abfcb29325f17f0069beb581c145710e85e Jeremy Evans code@jeremyevans.net ac1e97f1bc0454eb3fa1fcd8d7a393d809807427 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/f91d716f69f78b5aba129e8af1c3c9521dee55f4 f91d716f69f78b5aba129e8af1c3c9521dee55f4

2009-03-26T21:30:47-07:00

Break Database#typecast_value into separate functions for each type d905a52c6c5acf0374ccac4de50e6a824e4c1928 Jeremy Evans code@jeremyevans.net 8feee320f33e2798963237771465ea1f035e9a39 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/ac1e97f1bc0454eb3fa1fcd8d7a393d809807427 ac1e97f1bc0454eb3fa1fcd8d7a393d809807427

2009-03-26T21:12:42-07:00

Change the deprecation messages to specify that removal/changes will happen in 3.0 6cc2e62c8ffc7eb77f2f022b64866ddfb5c70bd1 Jeremy Evans code@jeremyevans.net 9b429ccc50502ddf1933e5d6871afed1c79837e6 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/8feee320f33e2798963237771465ea1f035e9a39 8feee320f33e2798963237771465ea1f035e9a39

2009-03-26T21:07:22-07:00

Move Model.extract_options! from base to the validation_class_method plugin, since that is the only user d916f3d7692152a658365cc3a6e14e83f5cd382a Jeremy Evans code@jeremyevans.net 0c9fa7c5411b69587b2a2261e081081754342212 Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/9b429ccc50502ddf1933e5d6871afed1c79837e6 9b429ccc50502ddf1933e5d6871afed1c79837e6

2009-03-26T21:01:07-07:00

Add validation_helpers model plugin, which adds instance level validation support similar to previously standard validations, with a different API This plugin should satisfy most validation needs. All previously standard validations were ported except for acceptance_of and confirmation_of, both of which are anti-patterns, as they encourage error checking in the model that should be done in the controller. Model validations should deal with the errors in the model's data, not whether a user checked a box or put an identical value in a confirmation field. validates_each was not ported as it is much easier to just write your own validation code if a standard validation doesn't handle it: # Old class level validation validates_each(:date) do |o,a,v| o.errors[a] << '...' unless v > Date.today end # New instance level validation def validate errors[:date] << '...' unless date > Date.today end The other validations were mostly renamed: # Old class level validations validates_format_of :col, :with=>/.../ validates_length_of :col, :maximum=>5 validates_length_of :col, :minimum=>3 validates_length_of :col, :is=>4 validates_length_of :col, :within=>3..5 validates_not_string :col validates_numericality_of :col validates_numericality_of :col, :only_integer=>true validates_presence_of :col validates_inclusion_of :col, :in=>[3, 4, 5] validates_uniqueness_of :col, :col2 validates_uniqueness_of([:col, :col2]) # New instance level validations def validate validates_format /.../, :col validates_max_length 5, :col validates_min_length 3, :col validates_exact_length 4, :col validates_length_range 3..5, :col validates_not_string :col validates_numeric :col validates_integer :col validates_presence :col validates_includes([3,4,5], :col) validates_unique :col, :col2 validates_unique([:col, :col2]) end Another change made is to specify the same type of validation on multiple attributes, you must use an array: # Old validates_length_of :name, :password, :within=>3..5 # New def validate validates_length_range 3..5, [:name, :password] end The :message, :allow_blank, :allow_missing, and :allow_nil options are still respected. The :tag option is not needed as instance level validations work with code reloading without workarounds. The :if option is also not needed for instance level validations: # Old validates_presence_of :name, :if=>:new? validates_presence_of :pass, :if=>{flag > 3} # New def validate validates_presence(:name) if new? validates_presence(:pass) if flag > 3 end validates_unique is a little different than the other new validations. It doesn't accept the :allow_* options, and you can specify multiple attributes instead of using an array. If you use an array, it is intepreted that you want the combination of values unique, instead of each value unique, which is how it operated previously. The new uniqueness logic is also much simpler and hopefully more robust, though you should still have a unique database index for integrity purposes. The validation_helpers plugin is half the size of the validation_class_methods plugin, and has better specs, IMO. This commit also removes some duplicative code from the validation_class_methods specs. ec4babeb747e0d9b56f5b5781f6ea214a0300d04 Jeremy Evans code@jeremyevans.net 4faa6f50c3a853e020b0b62eaf6a43c73bcbb31d Jeremy Evans code@jeremyevans.net http://github.com/jeremyevans/sequel/commit/0c9fa7c5411b69587b2a2261e081081754342212 0c9fa7c5411b69587b2a2261e081081754342212

2009-03-26T14:40:04-07:00

Split multi_insert into 2 methods with separate APIs, keep (deprecated) backwards compatibility The API for multi_insert has always bothered me. It's too complex, and the method should have been split from the beginning. import was an alias to multi_insert, so I've split the method and used import as the underlying method that takes an array columns and an array or arrays of values or a dataset, and had multi_insert call import. Backwards compatibility was kept, but it's deprecated and will be removed before 3.0. While here, fix a very misleading RDoc example for multi_insert, probably added by me about a year ago because I didn't understand the method's API (using *args when you don't actually accept an arbitrary number of arguments is a bad idea). bea8ba6b4867a2eff7a081031128488c2987931b Jeremy Evans code@jeremyevans.net