1. String
2. Symbol

What is a symbol?

A symbol is a string which is prefixed colon at starting. In other words, if you take ruby string such as “gaurish”, ‘gaurish sharma’ or “gaurish-sharma” & prefix a colon(‘:’) at te start, you get a symbol.  here is can example.

Two ways to do the essentially the same thing. Why?

Yes, you guessed it right. Strings & Symbols are essentially the thing. so why have two diffident “types” to represent string? The reason is how strings are implemented in ruby.

In Ruby, Everything is a Object including Strings

Due to SmallTalk influence, Ruby is fully object oriented language & there are no primitive types. Everything is implemented as Objects. so, when ever you try to assigned a string value to an variable like this x = “foo”, you are creating a new object in memory. Here is an small program which allocates same string to a variable 1,000 times & prints object_id of each string

As you may notice from terminal output of this program/script that each string has different object_id. ruby does creates a new object in memory every single time even thought the string is the same.

Wonder, if there is a another way?

How can I have strings which create object in memory once & all later operations refer to that previously created object with same object_id?

Glad: you asked. Enter Symbols! here is the same program with symbols.

In contrast with string, Notice how object_id is same for symbols which means its the same symbol.

What are the differences between Symbols & Strings?

1. Symbols are immutable: Their value remains constant.
2. Multiple uses of the same symbol have the same object ID and are the same object compared to string which will be a different object with unique object ID, everytime.
3. You can’t call any of the String methods like #upcase, #split on Symbols.

When to use symbol instead of string?

Symbols generally have performance benefits. so will see their usage almost everywhere including getters & setters in classes, hash keys etc. here is few examples

#you will NEVER see this
class Foo
  attr_reader "name"
end

# Instead, you will ALWAYS see a symbol
class Foo
  attr_reader :name
end
#same thing but symbols are faster than strings

to use strings when:

1. you want to use any of string methods like #upcase, #split, #downcase etc.
2. you want to change/mutate the string

to use Symbols when:

1. symbols can’t be changed at runtime. If you need something that absolutely, positively must remain constant
2. Places where same string is going to be repeatably used, example hash keys are pretty good candidate for symbols. so instead of string keys, hash["key"] = value. you should  use symbols like this  hash[:key] = value 

To conclude, strings & symbols in ruby are similar but differences given above. But the main difference is that symbols are immutable & slightly more performant than strings so you should you them place where you know same string is likely to be repeated again & again.

Related posts:

1. New Ruby 1.9 hash syntax
2. Ruby vs Java: Why do I like ruby more
3. Contributing to Ruby on Rails – A Guide

Step 0 – Decide what to contribute

It could be a bug or feature that you want aka starch your itch. It could a bug fix in which cause check github issue tracker for existing issues & see if you can fix anyone of them and send a Pull Request. It could be removing a warning from test. Or it would contributing documentation.

Step 1 – Fork & Change

This is rather simple but you must know how to use git – the popular distributed version control system. If you don’t then please following this guide closely & read multiple times until you understand.

• Goto rails page on github & click the “Fork” button on the top. requires github account. so register if not already done. Its free!
• You’ve successfully forked rails, but so far it only exists on GitHub. To be able to work on the project, you will need to clone it to your local machine.

  git clone https://github.com/your_github_username_goes_here/rails.git
  # Clones your fork of the repo into the current directory in terminal
  

• Create a new branch

  $ git checkout -b my_branch
  

• Create new rails app based on your fork

          $ /replace/with/path/to/your/rails/fork/railties/bin/rails new app_name_goes_here --dev
          

• Do your changes. Now whatever changes you do, would instantly get reflected into your rails app. Once you are okay with said changed, commit them

              $ git commit -a -m "Describe your change in this message"
              # commit your changes. make sure to follow the guidance of a good commit message
         

Step 2 – Create a Pull Request

so you have made your change, tested it.Sweet! Now, its time to send your awesome changes to rails core team for review.

• Add Upstream:

   $ git add remote git@github.com:rails/rails.git 

• switch back to master, pull latest changes

           $ git checkout master
  	 $ git pull --rebase upstream master
  

• Switch back your branch & Reapply your changes ontop of latest changes of the rails’ upstream master

   $ git checkout my_branch
  $ git rebase master
  

• In the process of the rebase, it may discover conflicts. In that case it will stop and allow you to fix the conflicts. After fixing conflicts, use git add . to update the index with those contents, and then just run:

  $ git rebase --continue
  

• Push branch to GitHub

  $ git push origin my_branch --force
  

• Open Pull request using github’s web-interface.

Congratulation on your contribution! Now, wait for comments on the Pull request.

Step – 3

Once your Pull request is opened, most probably someone will suggest 1 or 2 changes.  hers is how to do them:

1. switch to your feature branch

   	$ git checkout my_branch
   

2. make changes & commit
3. squash changes into one commit. rule is one feature == one commit
4. force push

To close, let me welcome you the world of open source.  where code you write gets used by thousands of developers & affects lives of millions of people around the world. Its pretty amazing!

Related posts:

1. Enable gzip compression for Rails 3.2 on heroku Cedar
2. Understanding Differences Between Symbols & Strings in Ruby
3. Ruby vs Java: Why do I like ruby more

Cloud computing is platform which provides  free space where we can keep our data securely.example gmail,Dropbox

Cloud  provides you software on rent

Cloud is the new type of hosting. Earlier we had shared,dedicated & now cloud hosting

Cloud allows you to rent computer(resources), so you can do work that requires i7 CPU with much slower intel Atom processor.

Are these people correct? this is what cloud computing means? Well…yes & no. Confused?

What is Cloud?

Cloud is different things to different people. The perspectives are very different; depending upon whom you are you asking this question.  such as,

• to consumers/non-technical people: Services like Dropbox,Gmail are cloud computing to them.
• to developers: Cloud is On Demand,elastic & Pro-rata billed, virtualized resource for running applications. Cloud allows us developers to stay developers & not have to become system administrator.
• to Sysadmins/DevOps: Cloud is platform that virtulizes the entire IT infrastructure, including servers, storage and networks. A
  cloud combines these resources into a simple and uniform set of computing resources in the virtual environment. Due to its easy provisioning & elastic nature cloud could be used to solve hard scaling problems like Scaling apps from 1 request/minute to 1,000requests/second & then back to 1 request/minute.This  kind of brust-scaling which was really hard to with bare-metal resources. But its much easier wth cloud computing
• to CIOs/CEO: Cloud is new way using which we can manage IT infrastructural resources as a shared utility, and can dynamically provision these resources efficiently to different business units and projects, without worrying about the underlying hardware differences and limitations
• to Journalists: Cloud is new Buzz word that everybody keeps is talking about & has a lot of activity, hence lot of articles/editorial needs to written about it.

In My opinion, what is cloud computing according?

To me, when someone says cloud; depending upon the context it could represent any of the following:
Cloud can be Application hosting platform(Google App engine, Microsoft Azure), Cloud can be Consumer Internet service like Gmail Dropbox, Cloud can be On-demand virtual machines,Cloud can be way of managing I.T infrastructure(servers,storage, networks etc) & so on…

Hence, I like to think Cloud computing more as marketing term which doesn’t mean much, but at same time its a term that most people across spectrum can understand enough get hint on what it it could it be related to.  This marketing term also  helps to push latest advances in Virtualization Technology, IT resource managent, Deployment & provisioning tools,  Standardization & Inter-Operability efforts. . so the term might be vague & may have different meaning depending upon the context which it is an useful term explaining all these areas.

So, that how I think about The cloud. what comes to your mind when you think about cloud?

What is AppFog?

AppFog is a Cloud Hosting platform which is categorized as Platform as a Service(PaaS). PaaS is the thing that lets developers stay developers but still empower them to deploy applications that can reach millions.  Basically, PaaS saves you from having to do system administration & focus only on development of your web application. think of all the tasks you have to do when a new box is commissioned into production; they are silly things like installing web server,download & compile PHP/ruby, installing security update, firewall & security configuration & countless other small configurations & optimizations. With PaaS like Appfog,  you get that application environment ready out-of-the-box  & with lots of other features like scaling, cloning etc. Appfog supports multiple runtimes,  PHP(v5.3),Ruby MRI(v1.9), Java(v7), Node.js(v0.4 & v0.6), Python(v2.7) & Erlang(R14).

Why should I care about AppFog?

Agreed, there are a lot of other PaaS solutions like Google App Engine, Redhat’s OpenShift, Salesforce’s Heroku, Microsoft’s Azure and many others but I am going to give 2 things that according to me are special about Appfog:

1. Ability to Choose Underlying Cloud Platform & Locations: Appfog lets you pick almost any cloud(current supported are AWS, Rackspace, Azure & HP’s Openstack) with choice of location where your app would be hosted(current locations are in US,Europe & Asia). This is in sharp contract with PaaS solutions like heroku where you are forced to host in US even half of your traffic comes from other parts of the world resulting increased latency. With Appfog, you can seamlessly deploy your application to multiple clouds at different locations. Thereby reducing SPOF
2. Generous Free Quota: Free plan gives you total 2 GB RAM(per account) which scalable over multiple instances, 1GB Database(per account), 50 GB data transfer limit(per account) & 100mb diskspace(per application). Except the 100mb diskspace, all others are pretty generous.  There is a comparison of appfog with other PaaS solutions on what you have to pay for 4 instances with 512MB each:
   

   Pass Provider	Price/mo
   dotcloud	$276
   Heroku	$107
   AppFog	$0

The Great WordPress Migration

creating new applications on appfog is straightforward & trivial thanks to nicely designed web console. Migrating existing apps involves few extra steps.In this part, I will describe how i got this blog running on a VPS migrated to Appfog. feel Skip this part, if you do not plan to run any PHP. However, if you plan to move your wordpress blog here, this is something you should read:

1. Download file backup & the Database dump(.SQL file) of your blog from your existing host.
2. Install ruby: we are going to the af command line tool to interact with appfog(reason in next section).  af is a rubygem which needs ruby to run. Unless, you have ruby programming language already installed like me(hello there, fellow rubyist!) you need to:
   • On Windows: Download & install railsinstaller
   • On Linux or Mac: Use rvm.  just run the following command

     curl -L get.rvm.io | bash -s stable

     for further details, see check out the RVM installation page or just watch the video below Once you have installed ruby, Verify ruby is correctly installed & is above version 1.9.2

     $ ruby --version
     ruby 1.9.3p194 (2012-04-20 revision 35410) [i686-linux]
     

3. Install & setup af tool: Now, time that you open your command prompt/terminal and start typing following commands. Install af tool

   $ gem install af
   Successfully installed af-0.3.16.5
   1 gem installed
   Installing ri documentation for af-0.3.16.5...
   Installing RDoc documentation for af-0.3.16.5...
    

   login with appfog account

   $ af login
   Attempting login to [https://api.appfog.com]
   Email: user@example.com
   Password: ***
   

4. Configure Database details:In Appfog, the database details are set as environment variables, so we need to edit wp-config.php and tell wordpress. basically, you need to replace staic values:

   define('DB_NAME', 'db_name');
   define('DB_USER', 'username');
   define('DB_PASSWORD', 'password');
   define('DB_HOST', 'localhost');
   define('DB_PORT', 3306);
   

   with the environment variables

   $services_json = json_decode(getenv('VCAP_SERVICES'),true);
   $mysql_config = $services_json['mysql-5.1'][0]['credentials'];
   define('DB_NAME', $mysql_config['name']);
   define('DB_USER', $mysql_config['user']);
   define('DB_PASSWORD', $mysql_config['password']);
   define('DB_HOST', $mysql_config['hostname']);
   define('DB_PORT', $mysql_config['port']);
   

5. Create the app: now, we are going to create the application & upload it on the cloud. for that first cd into the directory where you blog files are stored. for example, I have my files stored at /home/gaurish/wordpress so navigate to that director. remember, to cd into the directory & not outside the directory.

   $ cd /home/gaurish/wordpress/

   next, lets create the application. the commands are self-exploratory

   gaurish@DEN:~/wordpress$ af push
   Would you like to deploy from the current directory? [Yn]: Y
   Application Name: gaurishblog
   Detected a PHP Application, is this correct? [Yn]: Y
   Application Deployed URL [gaurishblog.aws.af.cm]: gaurishblog.aws.af.cm
   Memory reservation (128M, 256M, 512M, 1G, 2G) [128M]: 512M
   How many instances? [1]: 1
   Bind existing services to 'gaurishblog'? [yN]: n
   Create services to bind to 'gaurishblog'? [yN]: Y
   1: mongodb
   2: mysql
   What kind of service?: 2
   Specify the name of the service [mysql-7149]: mysql-gaurishblog
   Create another? [yN]: n
   Would you like to save this configuration? [yN]: Y
   Manifest written to manifest.yml.
   Creating Application: OK
   Creating Service [mysql-gaurishblog]: OK
   Binding Service [mysql-gaurishblog]: OK
   Uploading Application:
     Checking for available resources: OK
     Packing application: OK
     Uploading (0K): OK
   Push Status: OK
   Staging Application 'gaurishblog': OK
   Starting Application 'gaurishblog': OK

   If you see output similar to above, you have successfully created an application. if it fails to try checking the logs

   af logs gaurishblog --all

6. Migrate database:appfog doesn’t offer with direct access to database. so I created a tunnel to reach mysql using caldecott gem & import SQL dump file into mysql. Please note the tunnelling support is not available in all locations(ex not available in Asia, Europe). 

   $ gem install caldecott
   $ af tunnel
   1: mysql-gaurishblog
   Which service to tunnel to?: 1
   Binding Service [mysql-gaurishblog]: OK
   Stopping Application 'caldecott': OK
   Staging Application 'caldecott': OK
   Starting Application 'caldecott': OK
   Getting tunnel connection info: OK
   
   Service connection info:
     username : uXopxxxxmeI7
     password : plwWxxxxxekQPg
     name     : db25966b4xxxxxxxxxxx7dc515acf2c4
   
   Starting tunnel to mysql-gaurishblog on port 10000.
   1: none
   2: mysql
   3: mysqldump
   Which client would you like to start?: 1
   Open another shell to run command-line clients or
   use a UI tool to connect using the displayed information.
   Press Ctrl-C to exit...
   

   at this point the remote mysql database is been tunneled on your localhost.   you can use the given the Service connection info given above to import data using Mysql CLI client or install GUI client like Mysql Workbench.for mysql CLI client, open a new command prompt or terminal and enter

   $ mysql --protocol=TCP -P 10000 -h localhost -u uXopxxxxmeI7 -p'plwWxxxxxekQPg' db25966b4xxxxxxxxxxx7dc515acf2c4 < /path/to/mysqldump.sql
   

   For Mysql Workbench, use the following parameters:

7. Custom Domains: Now, my site was running on gaurishblog.aws.af.cm but I need to map it gaurishsharma.com. Good thing is Appfog allows to set up custom domains even on free plan! to make it work, I had to do two things. first, add gaurishsharma.com to Domain name list in web console. Second, create a new CNAME record for www.gaurishsharma.com pointing to cname01.us01.aws.af.cm. This worked but I still need to resolve gaurishsharma.com. However, as noted in the Web Console -> Domains, currently root-level domains are not supported. so As a workaround, I setup a URL forwarder using cloudflare’s page rules to send all traffic from root domain(gaurishsharma.com) to www.gaurishsharma.com. Worked Perfectly!

Initial Impressions

• No Vendor Lock-in, Multiple Clouds: As said earlier, Appfog allows you to choose which between different cloud platform along with server locations which is cool, if you don’t want to rely on single vendor & establish multi-cloud solution. And incase you are worried about lock-in with Appfog itself, you might be pleased to know that Appfog is based on VMware’s opensource Cloudfoundry platform.
• Free Plan which gives you generous resources which is compelling reason to try Appfog.Also, Appfog seems like excellent platform to run your personnel blog on for FREE
• Command Line tool is swiss army knife: the af CLI tool is handy and works well. Af gem can manage everything related to your appfog account with ease.
• Web console is a dud:, works sometimes but sometimes starts spliting out errors for some reason. Example, I tried creating a new php app at AWS Singapore from their web control panel & nothing was created. app didn’t show up, tried multiple times; didn’t work. sometimes the app would show up in the app list but open it in browser but then nginx would throws 404 – page not found error at you.  This left a bad first impression & bitter taste as Appfog is not beta service but they want to be used as platform to host production applications! But to credit of appfog, they reached out to me on Hacker new & offered support on twitter(even though I was free user!). In morning the same app which was throwing 404, started working just fine. I was greeted with WordPress 3-step install screen.
• There are Rough edges:  There are some of rough edges.When I uploaded my blog files to appfog using af update. the process worked flawlessly(no errors) & blog did start to appear. But few minutes after, mischievously nginx gave 404s. Later, I checked logs & found that I was exceeding disk-space limit of 100mb.      My fault but the thing is at no-point I was informed that I had exceeded diskspace nor there were any errors  during af update or af push.  This should have been handled better.
• PHP Platform: As per phpinfo()They seems to be using standard packages provided Ubuntu with which support all common php extensions. further, the storage is not writable which the same with all mordern PaaS. Nothing wrong but something you have to keep in mind specially for PHP apps like WordPress.

Conclusion

Inspite of all its current bugs, AppFog is a seems a promising PaaS platform because it is cheaper than other PaaS specially if you are running CPU intensive applications, their pricing model is simple, supports multiple runtimes,clouds & service locations. And not to mention they seem to have good support. I personally, would use Appfog to host my blog & host my rails application under development.

What about you? How is your experience with Appfog

image credits

HAML is a markup language that gets transformed into HTML. you may ask, why not write write my code in HTML, What’s wrong with HTML? And how does HAML solve does the problem?

Problem with HTML

HTML is verbose & becomes hard to maintain specially, if you have a large site. To see why HTML’s verbosity is issue, consider this example:

HTML

</pre>
<div id="profile">
<div class="left column">
<div id="date"><strong>4th June 2012</strong></div>
<div id="address"><em>Jaipur, India</em></div>
</div>
<div class="right column">
<div id="email"><a href="www.example.com">example.com </a></div>
<div id="bio">Rails Developer</div>
</div>
</div>
<pre>

As you may notice, there are couple of issues with html:

1. Every starting tag has to have a ending tag
2. The number of words & lines are more
3. plain HTML is pain to write & read

Now, compare this to HAML which basically does the same thing but in lot more human readable form

#profile
.left.column
#date
       %strong 4th June 2012
#address
       %em Jaipur, India
.right.column
#email
       %a{:href="www.example.com"}
            example.com
#bio   Rails Developer

So in short,

For more, kindly refer to HAML official site

Now, let’s go back to the reason of this blog post. Why do I like ruby more:
well, there are lot of reasons but for me, the main reason is the philosophy behind the langauge

Ruby is designed to make programmers happy

- Matz, Creator of Ruby

To explain this, consider this ruby code:

5.times {print "My Name is Gaurish "}

so even if you do not know ruby programming or heck do not know programming at all, I am pretty sure you can know what this code is doing does just by reading aloud words by word. Why, because it reads like “5 times print my name is gaurish”.

Lets take another example, this time a bit more complex , to read a file & then print it.  here is the code to do that in Java

class FileOperations{
static void readfile(String filename) throws FileNotFoundException {
  File input = new File(filename);
  String line;
  Scanner reader = null;
  try {
    reader = new Scanner(input);
    while(reader.hasNextLine()) {
      System.out.println(reader.nextLine());
    }
  } finally { reader.close(); }
}
public static void main(String args[]){
 readfile("hello.txt");
}

}

now, compare this to ruby code which does the same thing without using classes which is mandatory in java.  Another thing is, the Java version needs to wrap the creation of the Scanner in a try block so it can be guaranteed to be closed.Now, lets compare this to ruby version of same program

def readfile(filename)
  File.open(filename, "r") do |f|
    f.each_line {|line| print line }
  end
end
readfile "hello.txt"

Because of the existence of blocks, it is possible to abstract away the need to close the File in a single location, minimizing programmer error and reducing duplication. And this is one of the reason, I like ruby more than Java. There are plenty of other reasons & maybe I will cover those in future posts, if anyone is interested, then leave a comment below

PS: I have used print, instead of puts because to highlight the readability of ruby. But if you write ruby more, you will find yourself using puts as it auto inserts a new line which has to be manually inserted when using print.

Related posts:

1. New Ruby 1.9 hash syntax
2. Java is finally Free and Open! GPL Compatible.
3. Learning Java Programming Language – Day 1

hash = {:symbol => "value"}

can be written as

hash = {symbol: "value"}

now, what if both key & values are symbols, in ruby 1.8 & before you would write

:pick => :any

but now, this could be written in short form as

pick: :any

And this both new & old syntax for hash are supported in ruby 1.9, you can use whatever syntax you are comfortable the most but it helps to know both forms of syntax specially when reading the code that other might have written. personally, I like the new one — fewer characters to type

PS: this new has syntax is applicable only if the key is a :symbol. In case when key is not a symbol. you have to still stick with old array syntax(=>)

Related posts:

1. Understanding Differences Between Symbols & Strings in Ruby
2. Ruby vs Java: Why do I like ruby more
3. Contributing to Ruby on Rails – A Guide

In Cedar, the HTTP requests terminates directly at your app server & no longer goes through a proxy server(Nginx), hence there can’t be automatic gzip compression. More information about this can be found in HTTP routing dev center article So we have to manage gzip compression on our own. fortunately, this is trivial as adding just one line to your config. basically, you will need to use Rack::Deflater & make sure that it get loaded before ActionDispatch::Static. The simplest way to enable gzip compression in Rails 3.2 is by adding “use Rack::Deflater” in config.ru file.

After the change, this how your config.ru file should look like:

# This file is used by Rack-based servers to start the application.

require ::File.expand_path('../config/environment',  __FILE__)
use Rack::Deflater
run Improvingoutcomes::Application

I spending sometime gather information about this. So Hope this post help you save some time.