Erlang driver and Ecto adapter for Oracle Database
Using prepared statement functionality, the SQL statement you want to run is precompiled and stored in a database object, and you can run it as many times as required without compiling it every time it is run. If the data in the statement changes, you can use bind variables as placeholders for the data and then provide literal values at run time.
Using bind variables:
{"insert into tabl values (:1)", [<<16#E7,16#99,16#BE>>]}
{"insert into tabl values (:id, :dat)", [#{dat => {2023, 1, 1}, id => 1}]}
Calling stored procedure:
{"begin proc(:1, :2, :3); end;"
, [1.0, 2.0, 3.0]}
Calling stored function:
{"begin :1 := func(:2); end;"
, [{:out, :varchar}, "one hundred"]}
Using cursor variable:
{"begin open :1 for select * from tabl where dat>:2; end;"
, [:cursor, {2016, 8, 1}]}
Using returning clause:
{"insert into tabl values (tablid.nextval, sysdate) return id into :1"
, [{:out, :number}]}
Update batching:
{:batch, "insert into tabl values (:1, :2, :3)"
, [[1, 2, 3],[4, 5, 6],[7, 8, 9]]}
Row prefetching:
{:fetch, "select * from tabl where id>:1"
, [1]}
{:fetch, cursor, row_format, last_row}
Adapter options split in different categories described below. All options can be given via the repository configuration:
config :your_app, YourApp.Repo,
...
:hostname
- Server hostname (Name or IP address of the database server):port
- Server port (Number of the port where the server listens for requests):database
- Database (Database service name or SID with colon as prefix):username
- Username (Name for the connecting user):password
- User password (Password for the connecting user):parameters
- Keyword list of connection parameters:socket_options
- Options to be given to the underlying socket:timeout
- The default timeout to use on queries, defaults to 15000
:pool
- The connection pool module, defaults to DBConnection.ConnectionPool
:pool_size
- The size of the pool, defaults to 1
:idle_interval
- The ping interval to validate an idle connection, defaults to 5000
:autocommit
- Mode that issued an automatic COMMIT operation:fetch
- Number of rows to fetch from the server:sdu
- Size of session data unit:read_timeout
- Read timeout while reading from the socket, defaults to 500
:role
- Mode that is used in an internal logon:prelim
- Mode that is permitted when the database is down:newpassword
- User new password (Change password for the connecting user):description
- Connect descriptor:charset
- Client character set, defaults to UTF-8[{:out, :number}, {:out, :varchar}]
[{:out, :number}, {:out, :date}]
Oracle types | Literal syntax in params |
---|---|
NUMBER ,FLOAT ,BINARY_FLOAT |
:number , :integer , :float , :decimal |
CHAR , VARCHAR2 , CLOB |
:varchar , :char , :clob , :string |
NCHAR , NVARCHAR2 , NCLOB |
:nvarchar , :nchar , :nclob |
RAW , BLOB |
:raw , :blob , :binary , :hexstring |
DATE |
:date |
TIMESTAMP |
:timestamp |
TIMESTAMP WITH TIME ZONE |
:timestamptz |
SYS_REFCURSOR |
:cursor |
Using query options: [in: [:number, :binary]]
The primitive types are:
Ecto types | Oracle types | Literal syntax in params |
---|---|---|
:id , :integer |
NUMBER (*,0) , INTEGER |
1, 2, 3 |
:float |
NUMBER ,FLOAT ,BINARY_FLOAT |
1.0, 2.0, 3.0 |
:decimal |
NUMBER ,FLOAT ,BINARY_FLOAT |
Decimal |
:string |
CHAR , VARCHAR2 , CLOB |
"one hundred" |
:string |
NCHAR , NVARCHAR2 , NCLOB |
"百元", "万円" |
:binary |
RAW , BLOB |
"E799BE" (base 16 encoded) |
:binary |
RAW , BLOB |
<<0xE7,0x99,0xBE>> (option [in: [:binary]) |
:binary_id |
RAW , BLOB |
<<231,153,190, ...>> (option [in: [:binary_id]) |
:boolean |
CHAR , VARCHAR2 , NUMBER |
true, false |
:map |
CLOB , NCLOB |
%{"one" => 1, "hundred" => "百"} |
:naive_datetime |
DATE , TIMESTAMP |
NaiveDateTime |
:utc_datetime |
TIMESTAMP WITH TIME ZONE |
DateTime |
String
in Elixir is UTF-8 encoded binary.
:us7ascii
, :we8iso8859p1
, :ee8iso8859p2
, :nee8iso8859p4
, :cl8iso8859p5
, :ar8iso8859p6
,
:el8iso8859p7
,:iw8iso8859p8
, :we8iso8859p9
, :ne8iso8859p10
, :th8tisascii
, :vn8mswin1258
,
:we8iso8859p15
,:blt8iso8859p13
, :ee8mswin1250
, :cl8mswin1251
, :el8mswin1253
, :iw8mswin1255
,
:tr8mswin1254
,:we8mswin1252
, :blt8mswin1257
, :ar8mswin1256
:ja16euc
, :ja16sjis
, :ja16euctilde
,:ja16sjistilde
,:ko16mswin949
,
:zhs16gbk
, :zht32euc
, :zht16big5
, :zht16mswin950
, :zht16hkscs
iex> Ecto.Adapters.SQL.query(YourApp.Repo, "select 1+:1,sysdate,rowid from dual where 1=:1 ", [1])
{:ok, %{num_rows: 1, rows: [[2, ~N[2016-08-01 13:14:15], "AAAACOAABAAAAWJAAA"]]}}
iex> row = [%Ecto.Query.Tagged{value: <<0xE7,0x99,0xBE>>, type: :binary}]
iex> Ecto.Adapters.SQL.query(YourApp.Repo, "insert into tabl values (:1)", row)
iex> row = [%Ecto.Query.Tagged{value: %{dat: {2023, 1, 1}, id: 1}, type: :map}]
iex> Ecto.Adapters.SQL.query(YourApp.Repo, "insert into tabl values (:id, :dat)", row)
iex> opts = [batch: true, in: [Ecto.UUID, :number]]
iex> row = [Ecto.UUID.bingenerate, 1]
iex> Ecto.Adapters.SQL.query(YourApp.Repo, "insert into tabl values (:1, :2)",
...> [row, row], opts)
iex> opts = [returning: false, out: [:integer]]
iex> row = [Date.utc_today]
iex> Ecto.Adapters.SQL.query(YourApp.Repo, "insert into tabl (dat) values (:1) return id into :2",
...> row, opts)
Using quoted identifiers:
defmodule YourApp.Users do
use Ecto.Schema
schema "\\"USERS\\"" do
field :id, :integer
field :uuid, :binary_id
field :name, :string, source: :'"NAME"'
field :namae, :string, source: :'"名まえ"'
end
end
iex> YourApp.Repo.all(from(u in "\\"USERS\\"", select: u.'"NAME"', where: u.id == 1))
iex> YourApp.Repo.all(from(u in YourApp.Users, select: u.namae, where: u.id == 1))
iex> uuid = "601d74e4-a8d3-4b6e-8365-eddb4c893327"
iex> YourApp.Repo.all(from(u in YourApp.Users, select: u.name,
iex> where: u.uuid == type(^uuid, :binary_id)), [in: [:binary_id]])
Imagine you have this migration:
defmodule YourApp.Repo.Migrations.Users do
use Ecto.Migration
def change do
create table(:users, comment: "users table") do
add :name, :string, comment: "name column"
add :namae, :string, national: true
add :custom_id, :uuid
timestamps()
end
end
end
To migrate you'd do it normally:
$ mix ecto.migrate