mariastream

Simplified stream interface for mariaDB, build upon the mariasql module.

Installation

npm install mariastream

Example

// Other modules
var typepoint  = require('typepoint');
var inspectpoint = require('inspectpoint');
 
// This module
var mariastream = require('mariastream');
 
// Create client and connect
var client = mariastream();
    client.connnect({
      host: '127.0.0.1',
      user: 'root',
      password: ''
    });
 
// There are connection and close events, but you don't have to wait for that
 
client
  // Create a statement object, this is like prepare query that you can use
  // multiply times. Except it also works for normal querys without paramenters.
  .statement('SELECT id, act FROM database.table WHERE name = :name')
 
  // Create a readable object stream, there are also writable stream, duplex stream,
  // and a simple execute method there will buffer the result.
  .readable({name: 'andreas'})
 
  // This module don't do typecasting, but there are modules for that purpose.
  // NOTE: if you know of a fast way to do automatic typecasting please let me know
  .pipe(typepoint({id: Number, act: String}))
 
  // Each `data` event or `.read()` call gives an object,
  // run util.inspect and output it.
  .pipe(inspectpoint())
  .pipe(process.stdout);
 
  // Outputs:
  // {id: 1, act: 'sleeping'}
  // {id: 2, act: 'eating'}
  // {id: 3, act: 'programming'}
  // PS: Yeah, I can't eat and do programming at once, I'm a disgrace.

TODO

• Complete tests for abort once mariasql has been fixed
• Complete tests for connected property once mariasql has been fixed
• Write external stream module for type conversion

Documentation

This module is a wrapper around the mariasql module. Therefore if you are in doubt, you should consult thier documentation for more information.

The mariastream module returns an object constructor that you should use to create or wrap a connection.

var mariastream = require('mariastream');

Class: Client

Client = mariastream([options])

To create a client simple call the mariastream client constructor.

var client = mariastream();

The constructor takes an optional options object, the idea is that these options are inherited by each new Statement object. At the moment only the option useArray exists. For more information about this option visit the mariasql documentation.

var client = mariastream({useArray: true}); // Default is false

Client.connect(settings, [callback])

Client.connect(mariasql, [callback])

When client object is created it is not connected anything before the .connect method is executed. The connect methods requires a first argument, there can either be an object there takes the same options as [mariasql](https://github.com/mscdex/node-mariasql).connect() method, or it can be a mariasql instance. In case of the latter, you must call or have called the .connect() on the mariasql instance.

The second argument is an optional callback and will execute once the client is connected. In case you connected using an already connected mariasql instance the callback won't be called.

client.connect({
  host: '127.0.0.1',
  user: 'root',
  password: '',
  multiStatements: true,
  db: 'my-database'
}, function () {
  console.log('connected');
});

Client.close([callback])

Closes the connection. Note that all querys must be completed before the connection will be closed. The optional callback argument is called when connection is actually closed.

client.close(function () {
  console.log('connection is closed');
});

Client.connected

Returns true if the client is connected.

Client.threadId

If the client is connected, this is the thread id for the connection on the server.

Event: Client.on('error')

Errors can happen (e.q. connection error), unless its form a statement the error will emit on this object.

Event: Client.on('connect')

Emitted when a connection is made.

Event: Client.on('close')

Emitted when the connection is closed.

Class: Statement

Statement = Client.statement(sql, [options])

When you want to query something from a database, you have to create a statement object. If the SQL query is the same, you can reused this object. This is especially useful in case the query contains placeholders. The the internal prepare function won't have to be created multiply times. Once statement object has been made, you can then call one of its methods in order to execute the SQL query.

The .statement method, requires an SQL query, there can contain multiply commands. The query can also contain placeholders, the syntax of those are defined by the mariasql module. But put in a simple way if you use ? signs, then use arrays. If you use :placeholder then use objects.

var query = client.statement(
  'UPDATE mariastream.test SET value=:after WHERE value=:before;' +
  'SELECT value FROM mariastream.test WHERE value=:after'
);

The second argument is an optional object, there take the same properties as the client constructor. If a property is not defined here, it inherits from the options set in client constructor.

Abortable = Statement.execute([params], callback);

Unlike the other statement methods, this method don't return an object stream interface. Instread it simply executes the statement and buffer up the resutlt. Once the statement has been processed or an error occurred the callback will be called.

query.execute(function (err, rows, info) {
  console.log('statement execution is done');
});

The callback takes 3 arguments, the first is an Error or null, the second and array of the rows, the thrid is an info object there contain the following properties:

{
  queries: Number, // Number of successful queries
  insertId: Number, // Last inserted id
  affectedRows: Number, // Total affected rows
  numRows: Number // Total selected rows
}

It should be noted that unlike the mariasql module, the info object will contain the summed information for all subquerys in the statement.

The .execute method also takes an optional params object before the callback. This object contain parameters for the prepare statement to use.

ReadStream = Statement.readable([params])

The .readable method, returns an object ReadStream, where each fetched row can be accessed from the data event or the .read() method.

The method takes an optional params object, there sets the parameters in the prepare statement to use.

query.readable(['value'])
  .pipe(inspectpoint())
  .pipe(process.stdout);

DublexStream = Statement.duplex()

The .duplex method, returns an object DuplexStream. This means that each .write(params) will result in the statement to be processed with the required params objects specifing the parameters for the prepare statement to use. The rows from each statement can then be read though the data event or a .read() call.

Its important to note that since 10 writes will be processed simultaneously, there are no garantie for the order of the returned rows. If the order is important it is therefore recomended to return a static value (e.q. id) even if it is known in advance and only return one row (e.q. LIMIT 1).

var updater = client.statement(
  'SELECT id, value FROM mariastream.test WHERE id=:id LIMIT 1;' +
  'UPDATE mariastream.test SET value=:value WHERE id=:id'
).writable();
 
updater.write({id: 2, value: 'bar'});
updater.once('readable', function () {
  updater.read(); // {id: '2', value: 'foo'}
});

WriteStream = Statement.writable()

This works very much like the .duplex stream. The important diffrence is that this is a WriteStream only, meaning that it won't buffer up rows if you don't need them. Also you can't accidently pipe from this stream only too.

It should be noted that if the statement this WriteStream will execute do return rows, you are mostlikely doing it wrong. But now you have all the stream types you could want.

Class: Abortable

This object is returned by the statement.execute method. It is the same object that the [mariasql](https://github.com/mscdex/node-mariasql).query method will return.

Abortable.abort()

The only reason to use this object, is if you would abort the statement execution.

Class: Stream

There are 3 methods there return a stream object, they are called statement.readable, statement.duplex and statement.writable. Some of them can be read from others can be written to.

They all have the classic (stream)[http://nodejs.org/api/stream.html] methods, but futhermore they also share these methods:

Stream.info

Contains the same information as the info object in statement.execute but is updated continuely.

{
  queries: Number, // Number of successful queries
  insertId: Number, // Last inserted id
  affectedRows: Number, // Total affected rows
  numRows: Number // Total selected rows
}

Stream.destroy()

This will abort all active statements in progress related to the stream. In case of the statement.readable this is only one active query. But in the case of stream.wirtable and steam.duplex there can up to 10 active queries.

Event: Stream.on('close')

When the stream is done, either becase no more data can be read or written, or if the stream has been aborted this event will emit.

Event: Stream.on('error')

If an error happen while processing the query, this event will emit.

License

The software is license under "MIT"

Copyright (c) 2013 Andreas Madsen

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.