更多内容请访问 rubyonrails.org:

Active Record 和 PostgreSQL

本指南介绍了 PostgreSQL 中 Active Record 的特定用法。

阅读完本指南后，您将了解

  • 如何使用 PostgreSQL 的数据类型。
  • 如何使用 UUID 主键。
  • 如何在索引中包含非主键列。
  • 如何使用可延迟外键。
  • 如何使用唯一约束。
  • 如何实现排除约束。
  • 如何使用 PostgreSQL 实现全文搜索。
  • 如何使用数据库视图作为 Active Record 模型的底层实现。

要使用 PostgreSQL 适配器，您需要安装至少 9.3 版本的 PostgreSQL。不支持旧版本。

要开始使用 PostgreSQL，请查看 配置 Rails 指南。它描述了如何为 PostgreSQL 正确设置 Active Record。

1 数据类型

PostgreSQL 提供了许多特定数据类型。以下是 PostgreSQL 适配器支持的类型列表。

1.1 Bytea

# db/migrate/20140207133952_create_documents.rb
create_table :documents do |t|
  t.binary "payload"
end

# app/models/document.rb
class Document < ApplicationRecord
end

# Usage
data = File.read(Rails.root + "tmp/output.pdf")
Document.create payload: data

1.2 数组

# db/migrate/20140207133952_create_books.rb
create_table :books do |t|
  t.string "title"
  t.string "tags", array: true
  t.integer "ratings", array: true
end
add_index :books, :tags, using: "gin"
add_index :books, :ratings, using: "gin"

# app/models/book.rb
class Book < ApplicationRecord
end

# Usage
Book.create title: "Brave New World",
            tags: ["fantasy", "fiction"],
            ratings: [4, 5]

## Books for a single tag
Book.where("'fantasy' = ANY (tags)")

## Books for multiple tags
Book.where("tags @> ARRAY[?]::varchar[]", ["fantasy", "fiction"])

## Books with 3 or more ratings
Book.where("array_length(ratings, 1) >= 3")

1.3 Hstore

您需要启用 hstore 扩展才能使用 hstore。

# db/migrate/20131009135255_create_profiles.rb
class CreateProfiles < ActiveRecord::Migration[8.0]
  enable_extension "hstore" unless extension_enabled?("hstore")
  create_table :profiles do |t|
    t.hstore "settings"
  end
end

# app/models/profile.rb
class Profile < ApplicationRecord
end

irb> Profile.create(settings: { "color" => "blue", "resolution" => "800x600" })

irb> profile = Profile.first
irb> profile.settings
=> {"color"=>"blue", "resolution"=>"800x600"}

irb> profile.settings = {"color" => "yellow", "resolution" => "1280x1024"}
irb> profile.save!

irb> Profile.where("settings->'color' = ?", "yellow")
=> #<ActiveRecord::Relation [#<Profile id: 1, settings: {"color"=>"yellow", "resolution"=>"1280x1024"}>]>

1.4 JSON 和 JSONB

# db/migrate/20131220144913_create_events.rb
# ... for json datatype:
create_table :events do |t|
  t.json "payload"
end
# ... or for jsonb datatype:
create_table :events do |t|
  t.jsonb "payload"
end

# app/models/event.rb
class Event < ApplicationRecord
end

irb> Event.create(payload: { kind: "user_renamed", change: ["jack", "john"]})

irb> event = Event.first
irb> event.payload
=> {"kind"=>"user_renamed", "change"=>["jack", "john"]}

## Query based on JSON document
# The -> operator returns the original JSON type (which might be an object), whereas ->> returns text
irb> Event.where("payload->>'kind' = ?", "user_renamed")

1.5 范围类型

此类型映射到 Ruby Range 对象。

# db/migrate/20130923065404_create_events.rb
create_table :events do |t|
  t.daterange "duration"
end

# app/models/event.rb
class Event < ApplicationRecord
end

irb> Event.create(duration: Date.new(2014, 2, 11)..Date.new(2014, 2, 12))

irb> event = Event.first
irb> event.duration
=> Tue, 11 Feb 2014...Thu, 13 Feb 2014

## All Events on a given date
irb> Event.where("duration @> ?::date", Date.new(2014, 2, 12))

## Working with range bounds
irb> event = Event.select("lower(duration) AS starts_at").select("upper(duration) AS ends_at").first

irb> event.starts_at
=> Tue, 11 Feb 2014
irb> event.ends_at
=> Thu, 13 Feb 2014

1.6 组合类型

目前没有对组合类型的特殊支持。它们被映射到普通文本列

CREATE TYPE full_address AS
(
  city VARCHAR(90),
  street VARCHAR(90)
);

# db/migrate/20140207133952_create_contacts.rb
execute <<-SQL
  CREATE TYPE full_address AS
  (
    city VARCHAR(90),
    street VARCHAR(90)
  );
SQL
create_table :contacts do |t|
  t.column :address, :full_address
end

# app/models/contact.rb
class Contact < ApplicationRecord
end

irb> Contact.create address: "(Paris,Champs-Élysées)"
irb> contact = Contact.first
irb> contact.address
=> "(Paris,Champs-Élysées)"
irb> contact.address = "(Paris,Rue Basse)"
irb> contact.save!

1.7 枚举类型

该类型可以被映射为普通文本列，也可以映射为 ActiveRecord::Enum。

# db/migrate/20131220144913_create_articles.rb
def change
  create_enum :article_status, ["draft", "published", "archived"]

  create_table :articles do |t|
    t.enum :status, enum_type: :article_status, default: "draft", null: false
  end
end

您还可以创建枚举类型并将枚举列添加到现有表中

# db/migrate/20230113024409_add_status_to_articles.rb
def change
  create_enum :article_status, ["draft", "published", "archived"]

  add_column :articles, :status, :enum, enum_type: :article_status, default: "draft", null: false
end

以上迁移都是可逆的，但如果需要，您可以定义单独的 #up 和 #down 方法。在删除枚举类型之前，确保您删除了任何依赖枚举类型的列或表

def down
  drop_table :articles

  # OR: remove_column :articles, :status
  drop_enum :article_status
end

在模型中声明枚举属性会添加辅助方法，并防止将无效值分配给类的实例

# app/models/article.rb
class Article < ApplicationRecord
  enum :status, {
    draft: "draft", published: "published", archived: "archived"
  }, prefix: true
end

irb> article = Article.create
irb> article.status
=> "draft" # default status from PostgreSQL, as defined in migration above

irb> article.status_published!
irb> article.status
=> "published"

irb> article.status_archived?
=> false

irb> article.status = "deleted"
ArgumentError: 'deleted' is not a valid status

要重命名枚举，您可以使用 rename_enum，并同时更新模型中所有使用该枚举的地方

# db/migrate/20150718144917_rename_article_status.rb
def change
  rename_enum :article_status, :article_state
end

要添加新值，您可以使用 add_enum_value

# db/migrate/20150720144913_add_new_state_to_articles.rb
def up
  add_enum_value :article_state, "archived" # will be at the end after published
  add_enum_value :article_state, "in review", before: "published"
  add_enum_value :article_state, "approved", after: "in review"
  add_enum_value :article_state, "rejected", if_not_exists: true # won't raise an error if the value already exists
end

无法删除枚举值，这也意味着 add_enum_value 不可逆。您可以阅读 此处了解原因。

要重命名值，您可以使用 rename_enum_value

# db/migrate/20150722144915_rename_article_state.rb
def change
  rename_enum_value :article_state, from: "archived", to: "deleted"
end

提示：要显示所有枚举的所有值，您可以在 bin/rails db 或 psql 控制台中调用此查询

SELECT n.nspname AS enum_schema,
       t.typname AS enum_name,
       e.enumlabel AS enum_value
  FROM pg_type t
      JOIN pg_enum e ON t.oid = e.enumtypid
      JOIN pg_catalog.pg_namespace n ON n.oid = t.typnamespace

1.8 UUID

如果您使用的是 13.0 之前的 PostgreSQL 版本，则可能需要启用特殊扩展才能使用 UUID。启用 pgcrypto 扩展（PostgreSQL >= 9.4）或 uuid-ossp 扩展（适用于更早的版本）。

# db/migrate/20131220144913_create_revisions.rb
create_table :revisions do |t|
  t.uuid :identifier
end

# app/models/revision.rb
class Revision < ApplicationRecord
end

irb> Revision.create identifier: "A0EEBC99-9C0B-4EF8-BB6D-6BB9BD380A11"

irb> revision = Revision.first
irb> revision.identifier
=> "a0eebc99-9c0b-4ef8-bb6d-6bb9bd380a11"

您可以使用 uuid 类型在迁移中定义引用

# db/migrate/20150418012400_create_blog.rb
enable_extension "pgcrypto" unless extension_enabled?("pgcrypto")
create_table :posts, id: :uuid

create_table :comments, id: :uuid do |t|
  # t.belongs_to :post, type: :uuid
  t.references :post, type: :uuid
end

# app/models/post.rb
class Post < ApplicationRecord
  has_many :comments
end

# app/models/comment.rb
class Comment < ApplicationRecord
  belongs_to :post
end

有关将 UUID 用作主键的更多详细信息，请参阅 本节。

1.9 位字符串类型

# db/migrate/20131220144913_create_users.rb
create_table :users, force: true do |t|
  t.column :settings, "bit(8)"
end

# app/models/user.rb
class User < ApplicationRecord
end

irb> User.create settings: "01010011"
irb> user = User.first
irb> user.settings
=> "01010011"
irb> user.settings = "0xAF"
irb> user.settings
=> "10101111"
irb> user.save!

1.10 网络地址类型

类型 inet 和 cidr 被映射到 Ruby IPAddr 对象。macaddr 类型被映射到普通文本。

# db/migrate/20140508144913_create_devices.rb
create_table(:devices, force: true) do |t|
  t.inet "ip"
  t.cidr "network"
  t.macaddr "address"
end

# app/models/device.rb
class Device < ApplicationRecord
end

irb> macbook = Device.create(ip: "192.168.1.12", network: "192.168.2.0/24", address: "32:01:16:6d:05:ef")

irb> macbook.ip
=> #<IPAddr: IPv4:192.168.1.12/255.255.255.255>

irb> macbook.network
=> #<IPAddr: IPv4:192.168.2.0/255.255.255.0>

irb> macbook.address
=> "32:01:16:6d:05:ef"

1.11 几何类型

除 points 外的所有几何类型都被映射到普通文本。一个点被转换为包含 x 和 y 坐标的数组。

1.12 时间间隔

此类型被映射到 ActiveSupport::Duration 对象。

# db/migrate/20200120000000_create_events.rb
create_table :events do |t|
  t.interval "duration"
end

# app/models/event.rb
class Event < ApplicationRecord
end

irb> Event.create(duration: 2.days)

irb> event = Event.first
irb> event.duration
=> 2 days

2 UUID 主键

您需要启用 pgcrypto（仅 PostgreSQL >= 9.4）或 uuid-ossp 扩展来生成随机 UUID。

# db/migrate/20131220144913_create_devices.rb
enable_extension "pgcrypto" unless extension_enabled?("pgcrypto")
create_table :devices, id: :uuid do |t|
  t.string :kind
end

# app/models/device.rb
class Device < ApplicationRecord
end

irb> device = Device.create
irb> device.id
=> "814865cd-5a1d-4771-9306-4268f188fe9e"

如果未将 :default 选项传递给 create_table，则假定使用 gen_random_uuid()（来自 pgcrypto）。

要为使用 UUID 作为主键的表使用 Rails 模型生成器，请将 --primary-key-type=uuid 传递给模型生成器。

例如

$ rails generate model Device --primary-key-type=uuid kind:string

在构建具有引用此 UUID 的外键的模型时，将 uuid 视为本机字段类型，例如

$ rails generate model Case device_id:uuid

3 索引

PostgreSQL 包含各种索引选项。除了 常见索引选项之外，PostgreSQL 适配器还支持以下选项

3.1 包含

在创建新索引时，可以使用 :include 选项包含非主键列。这些键不用于索引扫描以进行搜索，但可以在索引扫描期间读取，而无需访问关联的表。

# db/migrate/20131220144913_add_index_users_on_email_include_id.rb

add_index :users, :email, include: :id

支持多个列

# db/migrate/20131220144913_add_index_users_on_email_include_id_and_created_at.rb

add_index :users, :email, include: [:id, :created_at]

4 生成列

从 PostgreSQL 12.0 版本开始支持生成列。

# db/migrate/20131220144913_create_users.rb
create_table :users do |t|
  t.string :name
  t.virtual :name_upcased, type: :string, as: "upper(name)", stored: true
end

# app/models/user.rb
class User < ApplicationRecord
end

# Usage
user = User.create(name: "John")
User.last.name_upcased # => "JOHN"

5 可延迟外键

默认情况下，PostgreSQL 中的表约束会在每个语句执行后立即检查。它有意不允许创建引用记录尚未存在的表的记录。但是，可以通过在 foreign key 定义中添加 DEFERRABLE 来在事务提交时运行此完整性检查。要默认延迟所有检查，可以将其设置为 DEFERRABLE INITIALLY DEFERRED。Rails 通过在 add_reference 和 add_foreign_key 方法中将 :deferrable 键添加到 foreign_key 选项来公开此 PostgreSQL 功能。

一个例子是在事务中创建循环依赖，即使您已经创建了外键

add_reference :person, :alias, foreign_key: { deferrable: :deferred }
add_reference :alias, :person, foreign_key: { deferrable: :deferred }

如果引用是使用 foreign_key: true 选项创建的，则以下事务将在执行第一个 INSERT 语句时失败。但是，如果设置了 deferrable: :deferred 选项，则不会失败。

ActiveRecord::Base.lease_connection.transaction do
  person = Person.create(id: SecureRandom.uuid, alias_id: SecureRandom.uuid, name: "John Doe")
  Alias.create(id: person.alias_id, person_id: person.id, name: "jaydee")
end

当 :deferrable 选项设置为 :immediate 时，让外键保持默认行为，即立即检查约束，但允许在事务中使用 set_constraints 手动延迟检查。这将导致外键在事务提交时进行检查

ActiveRecord::Base.lease_connection.transaction do
  ActiveRecord::Base.lease_connection.set_constraints(:deferred)
  person = Person.create(alias_id: SecureRandom.uuid, name: "John Doe")
  Alias.create(id: person.alias_id, person_id: person.id, name: "jaydee")
end

默认情况下，:deferrable 为 false，并且约束始终立即检查。

6 唯一约束

# db/migrate/20230422225213_create_items.rb
create_table :items do |t|
  t.integer :position, null: false
  t.unique_constraint [:position], deferrable: :immediate
end

如果您要将现有的唯一索引更改为可延迟的，您可以使用 :using_index 来创建可延迟的唯一约束。

add_unique_constraint :items, deferrable: :deferred, using_index: "index_items_on_position"

与外键一样，可以通过将 :deferrable 设置为 :immediate 或 :deferred 来延迟唯一约束。默认情况下，:deferrable 为 false，并且约束始终立即检查。

7 排除约束

# db/migrate/20131220144913_create_products.rb
create_table :products do |t|
  t.integer :price, null: false
  t.daterange :availability_range, null: false

  t.exclusion_constraint "price WITH =, availability_range WITH &&", using: :gist, name: "price_check"
end

与外键一样，可以通过将 :deferrable 设置为 :immediate 或 :deferred 来延迟排除约束。默认情况下，:deferrable 为 false，并且约束始终立即检查。

8 全文搜索

# db/migrate/20131220144913_create_documents.rb
create_table :documents do |t|
  t.string :title
  t.string :body
end

add_index :documents, "to_tsvector('english', title || ' ' || body)", using: :gin, name: "documents_idx"

# app/models/document.rb
class Document < ApplicationRecord
end

# Usage
Document.create(title: "Cats and Dogs", body: "are nice!")

## all documents matching 'cat & dog'
Document.where("to_tsvector('english', title || ' ' || body) @@ to_tsquery(?)",
                 "cat & dog")

可以选择将向量存储为自动生成的列（来自 PostgreSQL 12.0）

# db/migrate/20131220144913_create_documents.rb
create_table :documents do |t|
  t.string :title
  t.string :body

  t.virtual :textsearchable_index_col,
            type: :tsvector, as: "to_tsvector('english', title || ' ' || body)", stored: true
end

add_index :documents, :textsearchable_index_col, using: :gin, name: "documents_idx"

# Usage
Document.create(title: "Cats and Dogs", body: "are nice!")

## all documents matching 'cat & dog'
Document.where("textsearchable_index_col @@ to_tsquery(?)", "cat & dog")

9 数据库视图

假设您需要使用包含以下表的旧数据库

rails_pg_guide=# \d "TBL_ART"
                                        Table "public.TBL_ART"
   Column   |            Type             |                         Modifiers
------------+-----------------------------+------------------------------------------------------------
 INT_ID     | integer                     | not null default nextval('"TBL_ART_INT_ID_seq"'::regclass)
 STR_TITLE  | character varying           |
 STR_STAT   | character varying           | default 'draft'::character varying
 DT_PUBL_AT | timestamp without time zone |
 BL_ARCH    | boolean                     | default false
Indexes:
    "TBL_ART_pkey" PRIMARY KEY, btree ("INT_ID")

此表根本不符合 Rails 约定。由于简单的 PostgreSQL 视图默认情况下是可更新的，因此我们可以按如下方式对其进行包装

# db/migrate/20131220144913_create_articles_view.rb
execute <<-SQL
CREATE VIEW articles AS
  SELECT "INT_ID" AS id,
         "STR_TITLE" AS title,
         "STR_STAT" AS status,
         "DT_PUBL_AT" AS published_at,
         "BL_ARCH" AS archived
  FROM "TBL_ART"
  WHERE "BL_ARCH" = 'f'
SQL

# app/models/article.rb
class Article < ApplicationRecord
  self.primary_key = "id"
  def archive!
    update_attribute :archived, true
  end
end

irb> first = Article.create! title: "Winter is coming", status: "published", published_at: 1.year.ago
irb> second = Article.create! title: "Brace yourself", status: "draft", published_at: 1.month.ago

irb> Article.count
=> 2
irb> first.archive!
irb> Article.count
=> 1

此应用程序只关心未归档的 Articles。视图也允许使用条件，因此我们可以直接排除已归档的 Articles。

10 结构转储

如果您的 config.active_record.schema_format 为 :sql，Rails 将调用 pg_dump 来生成结构转储。

您可以使用ActiveRecord::Tasks::DatabaseTasks.structure_dump_flags配置pg_dump。 例如，要从结构转储中排除注释，请将其添加到初始化程序中

ActiveRecord::Tasks::DatabaseTasks.structure_dump_flags = ["--no-comments"]
返回顶部