defmodule Mix.Tasks.Phx.Gen.Schema do @shortdoc "Generates an Ecto schema and migration file" @moduledoc """ Generates an Ecto schema and migration. $ mix phx.gen.schema Blog.Post blog_posts title:string views:integer The first argument is the schema module followed by its plural name (used as the table name). The generated schema above will contain: * a schema file in `lib/my_app/blog/post.ex`, with a `blog_posts` table * a migration file for the repository The generated migration can be skipped with `--no-migration`. ## Contexts Your schemas can be generated and added to a separate OTP app. Make sure your configuration is properly setup or manually specify the context app with the `--context-app` option with the CLI. Via config: config :marketing_web, :generators, context_app: :marketing Via CLI: $ mix phx.gen.schema Blog.Post blog_posts title:string views:integer --context-app marketing ## Attributes The resource fields are given using `name:type` syntax where type are the types supported by Ecto. Omitting the type makes it default to `:string`: $ mix phx.gen.schema Blog.Post blog_posts title views:integer The following types are supported: #{for attr <- Mix.Phoenix.Schema.valid_types(), do: " * `#{inspect attr}`\n"} * `:datetime` - An alias for `:naive_datetime` The generator also supports references, which we will properly associate the given column to the primary key column of the referenced table: $ mix phx.gen.schema Blog.Post blog_posts title user_id:references:users This will result in a migration with an `:integer` column of `:user_id` and create an index. Furthermore an array type can also be given if it is supported by your database, although it requires the type of the underlying array element to be given too: $ mix phx.gen.schema Blog.Post blog_posts tags:array:string Unique columns can be automatically generated by using: $ mix phx.gen.schema Blog.Post blog_posts title:unique unique_int:integer:unique Redact columns can be automatically generated by using: $ mix phx.gen.schema Accounts.Superhero superheroes secret_identity:redact password:string:redact Ecto.Enum fields can be generated by using: $ mix phx.gen.schema Blog.Post blog_posts title status:enum:unpublished:published:deleted If no data type is given, it defaults to a string. ## table By default, the table name for the migration and schema will be the plural name provided for the resource. To customize this value, a `--table` option may be provided. For example: $ mix phx.gen.schema Blog.Post posts --table cms_posts ## binary_id Generated migration can use `binary_id` for schema's primary key and its references with option `--binary-id`. ## repo Generated migration can use `repo` to set the migration repository folder with option `--repo`: $ mix phx.gen.schema Blog.Post posts --repo MyApp.Repo.Auth ## migration_dir Generated migrations can be added to a specific `--migration-dir` which sets the migration folder path: $ mix phx.gen.schema Blog.Post posts --migration-dir /path/to/directory ## prefix By default migrations and schemas are generated without a prefix. For PostgreSQL this sets the "SCHEMA" (typically set via `search_path`) and for MySQL it sets the database for the generated migration and schema. The prefix can be used to thematically organize your tables on the database level. A prefix can be specified with the `--prefix` flags. For example: $ mix phx.gen.schema Blog.Post posts --prefix blog > #### Warning {: .warning} > > The flag does not generate migrations to create the schema / database. > This needs to be done manually or in a separate migration. ## Default options This generator uses default options provided in the `:generators` configuration of your application. These are the defaults: config :your_app, :generators, migration: true, binary_id: false, timestamp_type: :naive_datetime, sample_binary_id: "11111111-1111-1111-1111-111111111111" You can override those options per invocation by providing corresponding switches, e.g. `--no-binary-id` to use normal ids despite the default configuration or `--migration` to force generation of the migration. ## UTC timestamps By setting the `:timestamp_type` to `:utc_datetime`, the timestamps will be created using the UTC timezone. This results in a `DateTime` struct instead of a `NaiveDateTime`. This can also be set to `:utc_datetime_usec` for microsecond precision. """ use Mix.Task alias Mix.Phoenix.Schema @switches [migration: :boolean, binary_id: :boolean, table: :string, web: :string, context_app: :string, prefix: :string, repo: :string, migration_dir: :string] @doc false def run(args) do if Mix.Project.umbrella?() do Mix.raise "mix phx.gen.schema must be invoked from within your *_web application root directory" end schema = build(args, []) paths = Mix.Phoenix.generator_paths() prompt_for_conflicts(schema) schema |> copy_new_files(paths, schema: schema) |> print_shell_instructions() end defp prompt_for_conflicts(schema) do schema |> files_to_be_generated() |> Mix.Phoenix.prompt_for_conflicts() end @doc false def build(args, parent_opts, help \\ __MODULE__) do {schema_opts, parsed, _} = OptionParser.parse(args, switches: @switches) [schema_name, plural | attrs] = validate_args!(parsed, help) opts = parent_opts |> Keyword.merge(schema_opts) |> put_context_app(schema_opts[:context_app]) |> maybe_update_repo_module() Schema.new(schema_name, plural, attrs, opts) end defp maybe_update_repo_module(opts) do if is_nil(opts[:repo]) do opts else Keyword.update!(opts, :repo, &Module.concat([&1])) end end defp put_context_app(opts, nil), do: opts defp put_context_app(opts, string) do Keyword.put(opts, :context_app, String.to_atom(string)) end @doc false def files_to_be_generated(%Schema{} = schema) do [{:eex, "schema.ex", schema.file}] end @doc false def copy_new_files(%Schema{context_app: ctx_app, repo: repo, opts: opts} = schema, paths, binding) do files = files_to_be_generated(schema) Mix.Phoenix.copy_from(paths, "priv/templates/phx.gen.schema", binding, files) if schema.migration? do migration_dir = cond do migration_dir = opts[:migration_dir] -> migration_dir opts[:repo] -> repo_name = repo |> Module.split() |> List.last() |> Macro.underscore() Mix.Phoenix.context_app_path(ctx_app, "priv/#{repo_name}/migrations/") true -> Mix.Phoenix.context_app_path(ctx_app, "priv/repo/migrations/") end migration_path = Path.join(migration_dir, "#{timestamp()}_create_#{schema.table}.exs") Mix.Phoenix.copy_from paths, "priv/templates/phx.gen.schema", binding, [ {:eex, "migration.exs", migration_path}, ] end schema end @doc false def print_shell_instructions(%Schema{} = schema) do if schema.migration? do Mix.shell().info """ Remember to update your repository by running migrations: $ mix ecto.migrate """ end end @doc false def validate_args!([schema, plural | _] = args, help) do cond do not Schema.valid?(schema) -> help.raise_with_help "Expected the schema argument, #{inspect schema}, to be a valid module name" String.contains?(plural, ":") or plural != Phoenix.Naming.underscore(plural) -> help.raise_with_help "Expected the plural argument, #{inspect plural}, to be all lowercase using snake_case convention" true -> args end end def validate_args!(_, help) do help.raise_with_help "Invalid arguments" end @doc false @spec raise_with_help(String.t) :: no_return() def raise_with_help(msg) do Mix.raise """ #{msg} mix phx.gen.schema expects both a module name and the plural of the generated resource followed by any number of attributes: mix phx.gen.schema Blog.Post blog_posts title:string """ end defp timestamp do {{y, m, d}, {hh, mm, ss}} = :calendar.universal_time() "#{y}#{pad(m)}#{pad(d)}#{pad(hh)}#{pad(mm)}#{pad(ss)}" end defp pad(i) when i < 10, do: << ?0, ?0 + i >> defp pad(i), do: to_string(i) end