Question

使用小型Ruby脚本创建数据并将其存储到Postgres数据库。该应用程序在创建和删除数据库或运行迁移时工作得很好。但是，当我尝试运行rake db:seed时失败。该错误消息表明该表不存在...但同时如果我运行错误消息中引用的相同SQL脚本，但在pgAdmin中，它将返回有效结果。我想知道授予database.yml中命名的用户的权限是否正在发生，如果是，我将如何解决脚本中的问题？任何帮助理解这一点将不胜感激。

以下是我在终端中收到的错误消息：

acmecorp$ rake db:seed
rake aborted!
ActiveRecord::StatementInvalid: PG::UndefinedTable: ERROR:  relation "applications" does not exist
LINE 8:                WHERE a.attrelid = '"applications"'::regclass
                                          ^
:               SELECT a.attname, format_type(a.atttypid, a.atttypmod),
                     pg_get_expr(d.adbin, d.adrelid), a.attnotnull, a.atttypid, a.atttypmod,
                     c.collname, col_description(a.attrelid, a.attnum) AS comment
                FROM pg_attribute a
                LEFT JOIN pg_attrdef d ON a.attrelid = d.adrelid AND a.attnum = d.adnum
                LEFT JOIN pg_type t ON a.atttypid = t.oid
                LEFT JOIN pg_collation c ON a.attcollation = c.oid AND a.attcollation <> t.typcollation
               WHERE a.attrelid = '"applications"'::regclass
                 AND a.attnum > 0 AND NOT a.attisdropped
               ORDER BY a.attnum
/Users/thisguy/repositories/acmecorp/db/seeds.rb:1:in `<top (required)>'
/Users/thisguy/repositories/acmecorp/rakefile.rb:30:in `require_relative'
/Users/thisguy/repositories/acmecorp/rakefile.rb:30:in `block (2 levels) in <top (required)>'

Caused by:
PG::UndefinedTable: ERROR:  relation "applications" does not exist
LINE 8:                WHERE a.attrelid = '"applications"'::regclass
                                          ^
/Users/thisguy/repositories/acmecorp/db/seeds.rb:1:in `<top (required)>'
/Users/thisguy/repositories/acmecorp/rakefile.rb:30:in `require_relative'
/Users/thisguy/repositories/acmecorp/rakefile.rb:30:in `block (2 levels) in <top (required)>'
Tasks: TOP => db:seed
(See full trace by running task with --trace)

我的database.yml文件：

# config/database.yml
default: &default
  host: 'localhost'
  adapter: 'postgresql'
  encoding: utf-8

development:
  <<: *default
  database: 'acmecorp_dev'
  username: 'acmecorpapp'

Gemfile非常简单：

# Gemfile
source 'https://rubygems.org'

gem 'activerecord'
gem 'require_all'
gem 'pg'

我有一个rakefile.rb来控制各种任务（编辑删除不相关的项目）：

# rakefile.rb
require "active_record"
require "require_all"
require "yaml"

namespace :db do
  env = 'development' || ENV['env']

  db_config       = YAML::load(File.open('config/database.yml'))[env]
  db_config_admin = db_config.merge({'database' => 'postgres', 'schema_search_path' => 'public'})

  ...

  desc "Seed the database"
  task :seed do
    ActiveRecord::Base.establish_connection(db_config_admin)
    require_all 'models/*.rb'
    require_relative 'db/seeds.rb'
  end

  ...

end

如上所述，rakefile中的所有其他任务都可以正常工作 - 它只是打嗝的:seed任务。我非常有信心它与脚本与数据库交互的方式有关。

这是种子文件：

# db/seeds.rb
Application.create({
  ApplicationNumber:  1,
  AccountNumber: 1234,
  ApplVer:  1,
  CreateDateTime:  "1/2/2018",
  ExpirationDateTime:  "1/5/2019",
  ApplicationStatus:  "In process",
  ApprovedAmount:     1992.92,
  AcceptedAmount:     92.92,
  HomeAddressLine1:   "1 Main Street",
  HomeAddressLine2:   "",
  HomeAddressCity:    "Thibodaux",
  HomeAddressState:   "LA",
  HomeAddressZipCode: "12345"
})

最后，我有一个简单的应用程序模型，seeds.rb文件引用：

# models/application.rb
class Application < ActiveRecord::Base
end

现在上面提到的错误信息表明`关系＆＃34;应用程序＆＃34;不存在＆＃34; ...但是当我在消息中运行相同的SQL时，直接在pgAdmin中，这是我得到的（对于格式化道歉 - 无法弄清楚如何将其引入有效的表格：

attname|format_type|pg_get_expr|attnotnull|atttypid|atttypmod|collname|comment
id|integer|nextval('applications_id_seq'::regclass)|t|23|-1||
ApplicationNumber|integer||f|23|-1||
AccountNumber|integer||f|23|-1||
ApplVer|integer||f|23|-1||
CreateDateTime|timestamp without time zone||f|1114|-1||
ExpirationDateTime|timestamp without time zone||f|1114|-1||
ApplicationStatus|character varying||f|1043|-1||
ApprovedAmount|numeric||f|1700|-1||
AcceptedAmount|numeric||f|1700|-1||
HomeAddressLine1|character varying||f|1043|-1||
HomeAddressLine2|character varying||f|1043|-1||
HomeAddressCity|character varying||f|1043|-1||
HomeAddressState|character varying||f|1043|-1||
HomeAddressZipCode|character varying||f|1043|-1||
FileCreatedDate|timestamp without time zone||f|1114|-1||
created_at|timestamp without time zone||t|1114|-1||
updated_at|timestamp without time zone||t|1114|-1||

所以底线似乎是： - 脚本已正确配置为与PG一起使用 - 至少就创建，删除和迁移数据库而言 - rake db:seed任务失败 - 在命令行上的错误消息中报告的SQL在通过pgAdmin

运行时工作正常

我不确定为什么会这个问题。是否与acmecorpapp中分配给数据库的database.yml用户可用的权限有关？

Answer 1

请检查档案rakefile.rb。它似乎合并了错误的数据库。用于开发的数据库是acmecorp_dev，您已经在db_config中拥有它。我想你不必再合并database了。试试这个：

require "active_record"
require "require_all"
require "yaml"

namespace :db do
  env = 'development' || ENV['env']

  db_config       = YAML::load(File.open('config/database.yml'))[env]
  db_config_admin = db_config.merge({ 'schema_search_path' => 'public'})

  ...

  desc "Seed the database"
    task :seed do
    ActiveRecord::Base.establish_connection(db_config_admin)
    require_all 'models/*.rb'
    require_relative 'db/seeds.rb'
 end

 ...

end

Answer 2

好的，在Chivorn Kouch上面的建议之后想出来了。解决方案需要额外的db_config变量，以及对:seed任务的修改。

在rakefile.rb：

中

# rakefile.rb
namespace :db do
  env = 'development' || ENV['env']

  db_config       = YAML::load(File.open('config/database.yml'))[env]
  db_config_admin = db_config.merge({ 'database' => 'postgres','schema_search_path' => 'public' })
  # add following line to create variable for the :seed task
  db_config_seed = db_config.merge({ 'schema_search_path' => 'public' })

  ...

  desc "Seed the database"
  task :seed do
    # change variable referenced in the connection line below
    ActiveRecord::Base.establish_connection(db_config_seed)
    require_all 'models/*.rb'
    require_relative 'db/seeds.rb'
    puts "Database seeded."
  end

  ...

end

通过这些更改，命令全部按预期工作。

没有Rails的Ruby / Postgres ActiveRecord：未定义的表错误：关系＆＃34; ...＆＃34;不存在

2 个答案: