Joy of Think

Find in ruby

Ruby has a module find which implement function as commnad "find".

Find.prune is used to stop recursive visit in following directory.

  require 'find'

  total_size = 0

  Find.find(ENV["HOME"]) do |path|
    if FileTest.directory?(path)
      if File.basename(path)[0] == ?.
        Find.prune       # Don't look any further into this directory.
      else
        next
      end
    else
      total_size += FileTest.size(path)
    end
  end

Cooperating with module such like File, using method like basename, extname, or Module FileTest, user can easy to implement find like functions.

Walk a directory/Recursively in Wikipedia

Wikipedia has a great page about work a directory recursively.

Hpricot

Hpricot is a fast ruby html parser.

Sanitize

Sanitize is a ruby lib which can filter html tag based on a whitelist.

Official Rake document

Rake Tutorial

Link: http://blog.aizatto.com/2007/05/21/activerecord-without-rails/

The basic idea is to using following method to connect database.

require "active_record"
ActiveRecord::Base.establish_connection(params)


The params must be a hash or object.

Based on this method, user can use yaml to load configure just like rails do:


dbconfig = YAML::load(File.open('database.yml'))
ActiveRecord::Base.establish_connection(dbconfig)


SQLite sample

development:
  adapter: sqlite3
  database: db/development.db
  pool: 5
  timeout: 5000

Migration


ActiveRecord::Migrator.up('db/migrate')
class CreateLogTables < ActiveRecord::Migration
def self.up
create_table :accounts do |t|
t.integer :account_id
end

end
end


Using ActiveRecord


class Account < ActiveRecord::Base
has_many :sessions
end

class Session < ActiveRecord::Base
belongs_to :user
has_many :actions
end

class Action < ActiveRecord::Base
belongs_to :session
end


Logger


ActiveRecord::Base.logger = Logger.new(File.open('log/database.log', 'a'))
ActiveRecord::Base.colorize_logging = false


Link: http://www.ruby-forum.com/topic/73148

Here is a very useful document about net::http performance. It introduces a idea that small buffer used in rbuf_fill causes interior performance of net/http. And he also introduce a workaround. The general idea is override the default behavior. Here are code:

class OverrideInternetMessageIO < Net::InternetMessageIO
  def rbuf_fill
    timeout(@read_timeout) {
      @rbuf << @socket.sysread(65536)
    }
  end
end

class NewHTTP < Net::HTTP
  def NewHTTP.socket_type
     OverrideInternetMessageIO
  end
end