Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Commits on Oct 6, 2015
  1. Add link to external miguel library

    Remove the merb_sequel link, as merb has been dead for years.
  2. Update CHANGELOG

  3. @skrobul

    call Model methods when serializing dataset

    skrobul authored
    This change will call the model method instead of returning raw value from the
    database. It will make the CSV serializer consistent in behavior with JSON
    serializer. Also, the InstanceMethod also calls the model methods rather than
    returning a value.
Commits on Oct 4, 2015
  1. @earaujoassis
Commits on Oct 1, 2015
  1. Raise Sequel::NoExistingObject instead of generic error if Model#refr…

    …esh can't find the related row
  2. Bump version to 4.27.0

Commits on Sep 29, 2015
  1. Remove busy-wait mention in thread-safety guide

    Sequel no longer busy-waits, unless you are using ruby 1.8.
  2. Update CHANGELOG

  3. @YorickPeterse

    Don't stub synchronization on MRI

    YorickPeterse authored
    Having a GIL does not mean you don't have to perform any synchronization
    at all. While most Ruby code won't be able to run in parallel native
    extensions _can_ run in parallel by calling functions such as
    rb_thread_check_ints() (which basically pauses the current thread when
    Depending on the Ruby environment to figure out if you need
    synchronization or not is _extremely_ dangerous. While this can work
    most of the time, once in a while you'll get weird NoMethodErrors or
    worse: segmentation faults, depending on the database adapter being
    used. In particular the PostgreSQL adapter (which uses the "pg" Gem)
    suffers from this as connection objects created by the "pg" Gem (or
    really any other object it provedes) are not thread-safe.
    While synchronizing access on MRI will have a slight performance impact,
    it's either that or errors/segmentation faults.
Commits on Sep 28, 2015
  1. Make bin/sequel warn if given arguments that it doesn't use

    In the future, this might raise an error.
Commits on Sep 25, 2015
  1. Fix the order of referenced composite keys returned by Database#forei…

    …gn_key_list on PostgreSQL (Fixes #1081)
Commits on Sep 24, 2015
Commits on Sep 17, 2015
  1. In the active model plugin, make Model#persisted? return false if the…

    … transaction used for creation is rolled back (Fixes #1076)
  2. Switch active_model plugin specs to spec format

    This removes a lot of the complexity, which is left over from when
    rspec was used for Sequel's tests.
Commits on Sep 15, 2015
Commits on Sep 13, 2015
  1. Use primary_key :keep_order option in the schema dumper if the auto i…

    …ncrementing column is not the first column in the table
  2. Set :auto_increment option correctly in the schema parser when the au…

    …to incrementing column is not the first column in the table
  3. Support :keep_order option to primary_key in schema generator, to not…

    … automatically make the primary key the first column
Commits on Sep 10, 2015
Commits on Sep 8, 2015
  1. Don't try to insert explicit value into autoincrementing column in specs

    Some databases such as MSSQL don't allow it.
Commits on Sep 5, 2015
  1. Add before_after_save plugin, for refreshing created objects and rese…

    …tting modified flag before calling after_create/update/save hooks
    I'd like to make this the default behavior, but I think it's too
    likely to break backwards compatibility for existing applications.
    So add a plugin for the desired behavior, and that plugin will
    become the default behavior in Sequel 5.
    If someone really desires the current behavior and is not just
    accidently relying on it, I can add a plugin for the current
    behavior to Sequel 5.
  2. Add Dataset#single_record! and #single_value! which don't require clo…

    …ning the receiver
    These should be faster due to the lack of cloning.  Additionally,
    this optimizes single_value and with_sql_single_value to use
    each{|_, v| return v} instead of values.first, which should be
    faster as it doesn't create an array, and additionally works
    correctly for model datasets.
    Note that this breaks backwards compatibility for code that is
    relying on with_sql_single_value on model datasets to return an
    array in [:column, value] format.
Commits on Sep 2, 2015
  1. Make postgres adapter work with postgres-pr 0.7.0 (Fixes #1074)

    This fixes an ArgumentError when trying to use Sequel with postgres-pr
    0.7.0, because that started shipping with a pg.rb file.
    Note that postgres-pr 0.7.0 doesn't work on ruby 2.0+ due to
    multibyte escape issues in a regexp.  For years I've only been
    testing using my postgres-pr fork.  Mention that fork in the doc
    so that people are aware of it.
Commits on Sep 1, 2015
  1. Bump version to 4.26.0

Commits on Aug 31, 2015
  1. Make Dataset#== not consider frozen status in determining equality

    Previously, since the frozen status was kept in @opts and #==
    compares @opts, a frozen dataset was not considered equal to an
    identical unfrozen dataset.  Pointed out by chanks.
Commits on Aug 25, 2015
  1. Make Database#alter_table_op_sql call op-specific method

    This dispatches to a new method based on the op.  It's generally
    cleaner and should be easier to override in adapters. Switch
    the adapters that overrode alter_table_op_sql to override the
    more specific methods.
Commits on Aug 20, 2015
  1. Add Dataset#grouping_sets to support GROUP BY GROUPING SETS on Postgr…

    …eSQL 9.5+, MSSQL 2008+, Oracle, DB2, and SQLAnywhere
Commits on Aug 17, 2015
  1. Fix handling of{set_dataset :table} on ruby 1.8

    This was broken with the implicit table name refactoring between
    4.24.0 and 4.25.0.
  2. @Larivact

    rewrite resources list in readme

    Larivact authored
    You shouldn't have to hover over every item in the resources list you should be able to just read it.
Something went wrong with that request. Please try again.