eat 0.0.7 → 0.1.0

data/.gitignore

@@ -2,3 +2,4 @@ pkg/*
 *.gem
 .bundle
 Gemfile.lock
+rdoc/

data/CHANGELOG

@@ -0,0 +1,5 @@
+0.1.0
+* no longer tries to open file with sudo if it gets a permission error
+* remove Eat::Config.timeout global option
+* add :timeout option example: eat(x, :timeout => y)
+* add :limit option example: eat(x, :limit => y)

data/README.rdoc

@@ -1,25 +1,33 @@
 =eat
 
-Instead of
+Problems with the standard <tt>open-uri</tt> library
 
-    require 'open-uri'
-    open('http://169.254.169.254/latest/meta-data/security-groups').gets
-    open('/home/seamus/foo.txt').gets
+* sometimes it returns <tt>String</tt> and sometimes <tt>StringIO</tt> (check out <tt>OpenURI::Buffer::StringMax</tt>, usually 10,240 bytes)
+* sometimes you have to call <tt>gets</tt> (for example <tt>open('http://yahoo.com').gets</tt>)
+* it overrides <tt>#open</tt> everywhere, which may be confusing.
 
-Try
+Try <tt>#eat</tt>, which ALWAYS returns a <tt>String</tt>:
 
     require 'eat'
-    eat('http://169.254.169.254/latest/meta-data/security-groups')
-    eat('/home/seamus/foo.txt')
+    eat('http://yahoo.com')                 #=> String
+    eat('/home/seamus/foo.txt')             #=> String
+    eat('file:///home/seamus/foo.txt')      #=> String
 
-==Global configuration
+==Options
 
-    ::Eat.config.remote_timeout = 10 #seconds
+    eat('http://yahoo.com', :timeout => 10)  # timeout after 10 seconds
+    eat('http://yahoo.com', :limit => 1024)  # only read the first 1024 chars
+
+==Warning: doesn't verify SSL certs
+
+If you need to check SSL certificates, please don't use this gem. It always sets
+
+    http.verify_mode = ::OpenSSL::SSL::VERIFY_NONE
 
 ==Supported schemas
 
-* local filesystem (it will try to use <tt>sudo</tt> if it can't read the file)
-* http (the timeout is 2 seconds)
+* local filesystem
+* http
 * https (it won't check the SSL certificate... if you want security, don't use this!)
 
 Copyright 2011 Seamus Abshere

data/Rakefile

@@ -8,3 +8,14 @@ Rake::TestTask.new(:test) do |test|
   test.pattern = 'test/**/test_*.rb'
   test.verbose = true
 end
+
+require 'rake/rdoctask'
+Rake::RDocTask.new do |rdoc|
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = 'eat'
+  rdoc.options << '--line-numbers' << '--inline-source'
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
+end
+
+task :default => :test

data/lib/eat.rb

@@ -1,42 +1,68 @@
 require 'uri'
 require 'singleton'
 
+# http://weblog.jamisbuck.org/2007/2/7/infinity
+unless defined?(::Infinity)
+  ::Infinity = 1.0/0
+end
+
 module Eat
-  def self.config
-    ::Eat::Config.instance
-  end
-  
-  class Config
-    include ::Singleton
-    attr_writer :remote_timeout
-    def remote_timeout
-      @remote_timeout || 2 #seconds
-    end
-  end
-  
   module ObjectExtensions
-    def eat(filesystem_path_or_uri)
-      uri = ::URI.parse filesystem_path_or_uri.to_s
+    # <tt>url</tt> can be filesystem or http/https
+    #
+    # Options:
+    # * <tt>:timeout</tt> in seconds
+    # * <tt>:limit</tt> is characters (bytes in Ruby 1.8)
+    #
+    # Example:
+    #    eat('http://brighterplanet.com')                 #=> '...'
+    #    eat('http://brighterplanet.com', :timeout => 10) #=> '...'
+    #    eat('http://brighterplanet.com', :limit => 1)    #=> '.'
+    def eat(url, options = {})
+      timeout = options[:timeout] || options['timeout'] || 2
+      limit = options[:limit] || options['limit'] || ::Infinity
+      uri = ::URI.parse url.to_s
+
+      body = []
+      read_so_far = 0
+
       case uri.scheme
+
       when 'file', nil
-        if ::File.readable? uri.path
-          ::File.read uri.path
-        else
-          `sudo /bin/cat #{uri.path}`
+        chunk_size = limit < 1_048_576 ? limit : 1_048_576
+        ::File.open(uri.path, 'r') do |f|
+          while chunk = f.read(chunk_size)
+            break if read_so_far > limit
+            read_so_far += chunk_size
+            body << chunk
+          end
         end
+
       when 'http', 'https'
         require 'net/http'
         require 'net/https' if uri.scheme == 'https'
-        (defined?(::SystemTimer) ? ::SystemTimer : ::Timeout).timeout(::Eat.config.remote_timeout) do
+        (defined?(::SystemTimer) ? ::SystemTimer : ::Timeout).timeout(timeout) do
           http = ::Net::HTTP.new uri.host, uri.port
           if uri.scheme == 'https'
             http.use_ssl = true
             # if you were trying to be real safe, you wouldn't use this library
             http.verify_mode = ::OpenSSL::SSL::VERIFY_NONE
           end
-          http.start { |session| session.get uri.request_uri }
-        end.body
+          http.start do |session|
+            catch :stop do
+              session.get(uri.request_uri, 'Accept-Encoding' => '') do |chunk|
+                throw :stop if read_so_far > limit
+                read_so_far += chunk.length
+                body << chunk
+              end
+              session.finish
+            end
+          end
+        end
+
       end
+
+      limit == ::Infinity ? body.join : body.join[0...limit]
     end
   end
 end

data/lib/eat/version.rb

@@ -1,3 +1,3 @@
 module Eat
-  VERSION = "0.0.7"
+  VERSION = "0.1.0"
 end

data/test/test_eat.rb

@@ -1,3 +1,4 @@
+# -*- encoding: utf-8 -*-
 require 'helper'
 
 require 'tempfile'
@@ -9,10 +10,6 @@ class Tempfile
 end
 
 class TestEat < Test::Unit::TestCase
-  def setup
-    ::Eat.config.remote_timeout = 10
-  end
-  
   def test_filesystem
     assert eat(__FILE__).include?('class TestEat < Test::Unit::TestCase')
   end
@@ -22,27 +19,15 @@ class TestEat < Test::Unit::TestCase
   end
   
   def test_uri
-    assert eat(::URI.parse('http://brighterplanet.com/robots.txt')).include?('User-agent')
+    assert eat(::URI.parse('http://brighterplanet.com/robots.txt'), :timeout => 10).include?('User-agent')
   end
   
   def test_http
-    assert eat('http://brighterplanet.com/robots.txt').include?('User-agent')
+    assert eat('http://brighterplanet.com/robots.txt', :timeout => 10).include?('User-agent')
   end
   
   def test_https
-    assert eat('https://brighterplanet.com/robots.txt').include?('User-agent')
-  end
-  
-  def test_sudo_filesystem
-    f = File.open('test_sudo_filesystem.txt', 'w')
-    f.write "hello world"
-    f.close
-    `sudo chown root #{f.path}`
-    `sudo chmod go-rwx #{f.path}`
-    assert !File.readable?(f.path)
-    assert eat(f.path).include?('hello world')
-  ensure
-    `sudo rm -f #{f.path}`
+    assert eat('https://brighterplanet.com/robots.txt', :timeout => 10).include?('User-agent')
   end
   
   def test_openuri_uses_tempfile
@@ -54,7 +39,17 @@ class TestEat < Test::Unit::TestCase
   
   def test_eat_doesnt_use_tempfile
     assert_nothing_raised do
-      eat 'http://do1ircpq72156.cloudfront.net/0.2.47/javascripts/prototype.rails-3.0.3.js'
+      eat 'http://do1ircpq72156.cloudfront.net/0.2.47/javascripts/prototype.rails-3.0.3.js', :timeout => 10
     end
   end
+
+  def test_limit_on_local_files
+    assert_equal '# -', eat(__FILE__, :limit => 3)
+    assert_equal '# -*-', eat(__FILE__, :limit => 5)
+  end
+    
+  def test_limit_on_remote_files
+    assert_equal 'Use', eat(::URI.parse('http://brighterplanet.com/robots.txt'), :timeout => 10, :limit => 3)
+    assert_equal 'User-', eat(::URI.parse('http://brighterplanet.com/robots.txt'), :timeout => 10, :limit => 5)
+  end
 end

metadata

@@ -1,48 +1,37 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: eat
-version: !ruby/object:Gem::Version 
-  hash: 17
+version: !ruby/object:Gem::Version
+  version: 0.1.0
   prerelease: 
-  segments: 
-  - 0
-  - 0
-  - 7
-  version: 0.0.7
 platform: ruby
-authors: 
+authors:
 - Seamus Abshere
 autorequire: 
 bindir: bin
 cert_chain: []
-
-date: 2011-04-13 00:00:00 -05:00
-default_executable: 
-dependencies: 
-- !ruby/object:Gem::Dependency 
+date: 2011-06-27 00:00:00.000000000Z
+dependencies:
+- !ruby/object:Gem::Dependency
   name: test-unit
-  prerelease: false
-  requirement: &id001 !ruby/object:Gem::Requirement 
+  requirement: &2153835220 !ruby/object:Gem::Requirement
     none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id001
-description: Lets you open local and remote files by immediately returning their contents as a string.
-email: 
+  prerelease: false
+  version_requirements: *2153835220
+description: Lets you open local and remote files by immediately returning their contents
+  as a string.
+email:
 - seamus@abshere.net
 executables: []
-
 extensions: []
-
 extra_rdoc_files: []
-
-files: 
+files:
 - .gitignore
+- CHANGELOG
 - Gemfile
 - README.rdoc
 - Rakefile
@@ -51,40 +40,30 @@ files:
 - lib/eat/version.rb
 - test/helper.rb
 - test/test_eat.rb
-has_rdoc: true
 homepage: https://github.com/seamusabshere/eat
 licenses: []
-
 post_install_message: 
 rdoc_options: []
-
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
+required_ruby_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
   none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-
 rubyforge_project: eat
-rubygems_version: 1.6.2
+rubygems_version: 1.8.5
 signing_key: 
 specification_version: 3
 summary: A more trustworthy open-uri for use with RSS feeds, config scripts, etc.
-test_files: 
+test_files:
 - test/helper.rb
 - test/test_eat.rb