murlsh 0.2.4 → 0.3.0

data/README.textile

@@ -16,8 +16,8 @@ Phusion Passenger Setup:
 
 <pre>
 <code>
-rake gemspec build
-gem install pkg/murlsh-x.x.x.gem
+gem sources -a http://gemcutter.org/
+gem install murlsh
 </code>
 </pre>
 

data/Rakefile

@@ -175,6 +175,7 @@ begin
     gemspec.email = 'matthewm@boedicker.org'
     gemspec.homepage = 'http://github.com/mmb/murlsh'
     gemspec.authors = ['Matthew M. Boedicker']
+    gemspec.executables = %w{murlsh}
 
     %w{
       activerecord 2.3.4

data/VERSION

@@ -1 +1 @@
-0.2.4
+0.3.0

data/bin/murlsh

@@ -6,7 +6,7 @@ FileUtils.cp_r(
   %w{.htaccess config.ru config.yaml plugins/ public/ Rakefile}.collect { |x| 
     File.join(File.dirname(__FILE__), '..', x) }, '.', :verbose => true)
 
-FileUtils.mkdir('tmp')
+FileUtils.mkdir_p('tmp', :verbose => true)
 
 puts <<eos
 Next steps:

data/config.yaml

@@ -1,5 +1,5 @@
 --- 
-auth_file: /home/mm6/murlsh_users
+auth_file: murlsh_users
 description: URLs found interesting by Matthew M. Boedicker
 db_file: murlsh.db
 feed_file: atom.xml

data/lib/murlsh/atom_feed.rb

@@ -5,8 +5,14 @@ require 'uri'
 
 module Murlsh
 
+  # ATOM feed builder.
   class AtomFeed
 
+    # root_url is the base url for the feed items.
+    #
+    # Options:
+    # * :filename - the file name of the feed (atom.xml)
+    # * :title - the feed title
     def initialize(root_url, options={})
       options = {
         :filename => 'atom.xml',
@@ -18,6 +24,7 @@ module Murlsh
       setup_id_fields
     end
 
+    # Set up fields to use for building item ids.
     def setup_id_fields
       uri_parsed = URI(@root_url)
 
@@ -28,6 +35,7 @@ module Murlsh
       @path = uri_parsed.path
     end
 
+    # Generate the feed and write it to the filesystem with locking.
     def write(entries, path)
       open(path, 'w') do |f|
         f.flock(File::LOCK_EX)
@@ -38,6 +46,8 @@ module Murlsh
       end
     end
 
+    # Build the feed using XML builder. Options are passed to
+    # Builder::XmlMarkup.new.
     def make(entries, options={})
       xm = Builder::XmlMarkup.new(options)
       xm.instruct! :xml
@@ -62,10 +72,12 @@ module Murlsh
       xm
     end
 
+    # Build the entry's id.
     def entry_id(url)
       "tag:#{@domain},#{url.time.strftime('%Y-%m-%d')}:#{@host}#{@path}#{url.id}"
     end
 
+    # Add an ATOM enclosure if the url is an image.
     def enclosure(xm, mu)
       xm.link(:rel => 'enclosure', :type => mu.content_type, :href => mu.url,
         :title => 'Full-size') if mu.is_image?

data/lib/murlsh/auth.rb

@@ -6,12 +6,21 @@ require 'digest/md5'
 
 module Murlsh
 
+  # Interface to authentication file. Format of authentication file:
+  #
+  # username,MD5 hash of email address,bcrypted password
+  #
+  # Authentication is done using password only to make adding easier and
+  # because there will be a small number of trusted users.
+  #
+  # See Rakefile for user maintenance tasks.
   class Auth
 
     def initialize(file)
       @file = file
     end
 
+    # Authenticate a user by password. Return their name and email if correct.
     def auth(password)
       CSV::Reader.parse(open(@file)) do |row|
         return { :name => row[0], :email => row[1] } if
@@ -19,6 +28,7 @@ module Murlsh
       end
     end
 
+    # Add a user to the authentication file.
     def add_user(username, email, password)
       open(@file, 'a') do |f|
         f.flock(File::LOCK_EX)

data/lib/murlsh/dispatch.rb

@@ -11,8 +11,10 @@ yaml
 
 module Murlsh
 
+  # Dispatch requests.
   class Dispatch
 
+    # Set up config hash and database connection.
     def initialize
       @config = YAML.load_file('config.yaml')
       @url_root = URI(@config.fetch('root_url')).path
@@ -25,6 +27,7 @@ module Murlsh
       @url_server = Murlsh::UrlServer.new(@config, @db)
     end
 
+    # Rack call.
     def call(env)
       dispatch = {
         ['GET', @url_root] => [@url_server, :get],
@@ -41,6 +44,7 @@ module Murlsh
       obj.send(meth, req).finish
     end
 
+    # Called if the request is not found.
     def not_found(req)
       Rack::Response.new("<p>#{req.url} not found</p>
 

data/lib/murlsh/get_content_type.rb

@@ -4,6 +4,7 @@ require 'uri'
 
 class URI::Generic
 
+  # Return the path and query string.
   def path_query
     path + (query ? "?#{query}" : '')
   end
@@ -14,6 +15,11 @@ module Murlsh
 
   module_function
 
+  # Try to get the content type of a url.
+  #
+  # Options:
+  # * :failproof - if true hide all exceptions and return empty string on failure
+  # * :headers - hash of headers to send in request
   def get_content_type(url, options={})
     options[:headers] = default_headers(url).merge(
       options.fetch(:headers, {}))
@@ -48,6 +54,10 @@ module Murlsh
     uri.is_a?(URI::HTTP) ? uri : URI(uri)
   end
 
+  # Create a Net::HTTP to a host and port.
+  #
+  # Options:
+  # * :debug - stream to write debug output to
   def make_net_http(url, options={})
     net_http = Net::HTTP.new(url.host, url.port)
     net_http.use_ssl = (url.scheme == 'https')
@@ -55,16 +65,14 @@ module Murlsh
     net_http
   end
 
-  # Get the response to HTTP HEAD. If HEAD not allowed do GET.
+  # Get the response to HTTP HEAD. If HEAD returns anything other than success
+  # try GET.
   def get_resp(http, url, headers={})
-    resp = http.request_head(url.path_query, headers)
-    if Net::HTTPMethodNotAllowed === resp
-      http.request_get(url.path_query, headers)
-    else
-      resp
-    end
+    resp = http.request_head(url.path_query, headers) === Net::HTTPSuccess ?
+      resp : http.request_get(url.path_query, headers)
   end
 
+  # Get default headers sent with the request. Can be based on url.
   def default_headers(url)
     result = {
       'User-Agent' =>

data/lib/murlsh/get_title.rb

@@ -10,6 +10,11 @@ module Murlsh
 
   module_function
 
+  # Try to get the title of a url.
+  #
+  # Options:
+  # * :failproof - if true hide all exceptions and return empty string on failure
+  # * :headers - hash of headers to send in request
   def get_title(url, options={})
     options[:headers] = default_headers(url).merge(
       options.fetch(:headers, {}))
@@ -35,6 +40,8 @@ module Murlsh
     (result and !result.empty?) ? result : url
   end
 
+  # Return true if the content type is likely to have a title that can be
+  # parsed.
   def might_have_title(content_type)
     content_type[/^text\/html/]
   end

data/lib/murlsh/markup.rb

@@ -1,7 +1,12 @@
 module Murlsh
 
+  # Helper mixin for XML builder.
   module Markup
 
+    # Javascript link builder. Takes list of script urls.
+    #
+    # Options:
+    # * :prefix - prefix to append to all script urls
     def javascript(sources, options={})
       sources.to_a.each do |src|
         script('', :type => 'text/javascript',
@@ -9,6 +14,15 @@ module Murlsh
       end
     end
 
+    # Image tag builder.
+    #
+    # Options:
+    # * :href - make the image a link to this url
+    # * :prefix - prefix to append to all image urls
+    # * :size - image size if square or [w, h]
+    # * :text - text for alt and title tag
+    #
+    # Any other options in hash will be added as attributes.
     def murlsh_img(options={})
       img_convert_prefix(options)
       img_convert_size(options)
@@ -24,10 +38,16 @@ module Murlsh
       end
     end
 
+    # ATOM feed link builder.
     def atom(href)
       link(:rel => 'alternate', :type => 'application/atom+xml', :href => href)
     end
 
+    # CSS link builder.
+    #
+    # Options:
+    # * :media - optional media attribute
+    # * :prefix - prepended to all CSS urls
     def css(hrefs, options={})
       hrefs.to_a.each do |href|
         attrs = {
@@ -40,10 +60,16 @@ module Murlsh
       end
     end
 
+    # Meta tag builder. Takes a hash of name => content.
     def metas(tags)
       tags.each { |k,v| meta(:name => k, :content => v) }
     end
 
+    # Gravatar builder. Takes MD5 hash of email address.
+    # Options:
+    # * 'd' - default Gravatar (identicon, monsterid, or wavatar)
+    # * 's' - size (0 - 512)
+    # * 'r' - rating (g, pg, r or x)
     def gravatar(email_hash, options={})
       query = options.reject do |k,v|
         not ((k == 'd' and %w{identicon monsterid wavatar}.include?(v)) or
@@ -60,6 +86,7 @@ module Murlsh
       murlsh_img(options)
     end
 
+    # Query string builder. Takes hash of query string variables.
     def build_query(h)
       h.empty? ? '' :
         '?' + h.collect { |k,v| URI.escape("#{k}=#{v}") }.join('&')

data/lib/murlsh/plugin.rb

@@ -1,13 +1,26 @@
 module Murlsh
 
+  # Superclass for plugins. How plugins are registered.
+  #
+  # Hooks:
+  # * add_pre - called before a new url is saved
+  #   run arguments (url, config hash)
+  # * add_post - called after a new url is saved
+  #   run arguments (config hash)
+  # * hostrec - called to post process the domain that shows after links
+  #   run arguments (domain, url, title)
   class Plugin
 
+    # Called when a plugin class inherits from this class (the way plugins
+    # are registered).
     def self.inherited(child)
       registered << child
     end
 
+    # Get registered plugins by hook (add_pre, add_post, etc.)
     def self.hooks(name)
-      matches = registered.select { |p| p::Hook == name }
+      matches = registered.select { |p| p::Hook == name }.
+        sort { |a,b| a.to_s <=> b.to_s }
 
       if block_given?
         matches.each { |p| yield p }

data/lib/murlsh/referrer.rb

@@ -3,6 +3,7 @@ require 'uri'
 
 module Murlsh
 
+  # For parsing query strings from referring search engines.
   class Referrer
 
     def initialize(url)
@@ -19,6 +20,7 @@ module Murlsh
       end
     end
 
+    # Get the searched for string from a search engine query string.
     def search_query(qmap=Qmap)
       if hostpath and query_string
         qmap.each_pair do |r,v|
@@ -34,6 +36,7 @@ module Murlsh
       nil
     end
 
+    # Regex host match to name of query string parameter.
     Qmap = {
       /^www\.google\.(bs|ca|com|cz|dk|es|fi|nl)(\/m)?\/search$/ => 'q',
       /^www\.bing\.com\/search$/ => 'q',

data/lib/murlsh/sqlite3_adapter.rb

@@ -2,6 +2,7 @@ require 'active_record/connection_adapters/sqlite3_adapter'
 
 class ActiveRecord::ConnectionAdapters::SQLite3Adapter
 
+  # Add MATCH function for regex matching.
   def initialize(connection, logger, config)
     super
     @connection.create_function('MATCH', 2) do |func,search_in,search_for|

data/lib/murlsh/time.rb

@@ -2,6 +2,8 @@ require 'time'
 
 class Time
 
+  # Return a string of the approximate amount of time that has passed since
+  # this time.
   def fuzzy
      days_ago = (Time.now.to_i - to_i) / 86400
 

data/lib/murlsh/url.rb

@@ -5,36 +5,35 @@ require 'uri'
 
 module Murlsh
 
+  # URL ActiveRecord.
   class Url < ActiveRecord::Base
 
+    # Get the title of this url.
     def title
       read_attribute(:title) || read_attribute(:url) || 'title missing'
     end
 
+    # Return true if this url has the same author as another url.
     def same_author?(other)
       other and other.email and other.name and
         email and name and email == other.email and name == other.name
     end
 
-    Widely_known = %w{
-      wikipedia.org
-      flickr.com
-      github.com
-      twitter.com
-      vimeo.com
-      youtube.com
-      }
-
+    # Return text showing what domain a link goes to.
     def hostrec
       begin
         domain = URI(url).host[/[a-z\d-]+\.[a-z]{2,}(\.[a-z]{2})?$/].downcase
       rescue Exception => e
         domain = nil
       end
-      yield domain unless !domain or title.downcase.index(domain) or
-        Widely_known.include?(domain)
+
+      domain = Murlsh::Plugin.hooks('hostrec').inject(domain) {
+        |result,plugin| plugin.run(result, url, title) }
+
+      yield domain if domain
     end
 
+    # Return true if this url is an image.
     def is_image?
       %w{image/gif image/jpeg image/png}.include?(content_type)
     end

data/lib/murlsh/url_body.rb

@@ -1,5 +1,6 @@
 module Murlsh
 
+  # Url list page builder.
   class UrlBody < Builder::XmlMarkup
     include Murlsh::Markup
 
@@ -8,12 +9,14 @@ module Murlsh
       super(:indent => 2)
     end
 
+    # Fetch urls base on query string parameters.
     def urls
       Murlsh::Url.all(:conditions => search_conditions, :order => 'id DESC',
         :limit =>  @req.params['n'] ? @req.params['n'].to_i :
           @config.fetch('num_posts_page', 100))
     end
 
+    # Search conditions builder for ActiveRecord conditions.
     def search_conditions
       if @q
         search_cols = %w{name title url}
@@ -24,6 +27,7 @@ module Murlsh
       end
     end
 
+    # Url list page body builder.
     def each
       instruct! :xml
       declare! :DOCTYPE, :html, :PUBLIC, '-//W3C//DTD XHTML 1.1//EN',
@@ -65,10 +69,12 @@ module Murlsh
           clear
           powered_by
           js
+          div('', :id => 'bottom')
         }
       }
     end
 
+    # Head builder.
     def headd
       head {
         titlee
@@ -84,20 +90,24 @@ module Murlsh
       }
     end
 
+    # Title builder.
     def titlee
       title(@config.fetch('page_title', '') + (@q ? " /#{@q}" : ''))
     end
 
+    # Google verification link builder.
     def google_verify
       (gv = @config.fetch('google_verify')) and metas('verify-v1' => gv)
     end
 
+    # Feed icon builder.
     def feed_icon
       div(:class => 'icon') {
         a('feed', :href => @config.fetch('feed_file'), :class => 'feed')
       }
     end
 
+    # Search form builder.
     def search_form
       form(:action => '', :method => 'get') {
         value = @q
@@ -113,6 +123,7 @@ module Murlsh
       }
     end
 
+    # Url add form builder.
     def add_form
       form(:action => '', :method => 'post') {
         fieldset(:id => 'add') {
@@ -125,15 +136,18 @@ module Murlsh
       }
     end
 
+    # Url add form input builder.
     def add_form_input(label, id, size, tipe='text')
       label(label, :for => id)
       input(:type => tipe, :id => id, :name => id, :size => size)
     end
 
+    # Clear div builder.
     def clear
       div(:style => 'clear : both')
     end
 
+    # Powered by builder.
     def powered_by
       self.p {
         text! 'powered by '
@@ -141,6 +155,7 @@ module Murlsh
       }
     end
 
+    # Required javascript builder.
     def js
       javascript(%w{
         jquery-1.3.2.min.js

data/lib/murlsh/url_server.rb

@@ -6,6 +6,7 @@ rack
 
 module Murlsh
 
+  # Build responses for HTTP requests.
   class UrlServer
 
     def initialize(config, db)
@@ -16,6 +17,8 @@ module Murlsh
       Dir['plugins/*.rb'].each { |p| load p }
     end
 
+    # Respond to a GET request. Return a page of urls based on the query
+    # string parameters.
     def get(req)
       resp = Murlsh::XhtmlResponse.new
 
@@ -30,6 +33,7 @@ module Murlsh
       resp
     end
 
+    # Respond to a POST request. Add the new url and return json.
     def post(req)
       resp = Rack::Response.new
 
@@ -42,16 +46,15 @@ module Murlsh
           ActiveRecord::Base.establish_connection(:adapter => 'sqlite3',
             :database => @config.fetch('db_file'))
 
-          content_type = Murlsh.get_content_type(url)
           mu = Murlsh::Url.new do |u|
             u.time = Time.now.gmtime
             u.url = url
             u.email = user[:email]
             u.name = user[:name]
-            u.title = Murlsh.get_title(url, :content_type => content_type)
-            u.content_type = content_type
           end
 
+          Murlsh::Plugin.hooks('add_pre') { |p| p.run(mu, @config) }
+
           mu.save
 
           Murlsh::Plugin.hooks('add_post') { |p| p.run(@config) }

data/lib/murlsh/xhtml_response.rb

@@ -1,9 +1,10 @@
 module Murlsh
 
+  # Set the content type correctly based on accept header and user agent.
   class XhtmlResponse < Rack::Response
 
-    # set the content type to application/xhtml+xml for anything that
-    # claims to accept it, for anything else or IE use text/html
+    # Set the content type to application/xhtml+xml for anything that
+    # claims to accept it, for anything else or IE use text/html.
     def set_content_type(http_accept, http_user_agent)
       self['Content-Type'] = if http_accept and
         http_accept[/((\*|application)\/\*|application\/xhtml\+xml)/i] and

data/murlsh.gemspec

@@ -5,11 +5,11 @@
 
 Gem::Specification.new do |s|
   s.name = %q{murlsh}
-  s.version = "0.2.4"
+  s.version = "0.3.0"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.authors = ["Matthew M. Boedicker"]
-  s.date = %q{2009-10-23}
+  s.date = %q{2009-11-05}
   s.default_executable = %q{murlsh}
   s.description = %q{url sharing site framework with easy adding, title lookup, atom feed, thumbnails and embedding}
   s.email = %q{matthewm@boedicker.org}
@@ -43,6 +43,9 @@ Gem::Specification.new do |s|
      "lib/murlsh/url_server.rb",
      "lib/murlsh/xhtml_response.rb",
      "murlsh.gemspec",
+     "plugins/hostrec_redundant.rb",
+     "plugins/hostrec_skip.rb",
+     "plugins/lookup_content_type_title.rb",
      "plugins/update_feed.rb",
      "public/css/jquery.jgrowl.css",
      "public/css/phone.css",

data/plugins/hostrec_redundant.rb

@@ -0,0 +1,14 @@
+module Murlsh
+
+  # skip showing host record if domain is contained in title
+  class HostrecRedundant < Plugin
+
+    Hook = 'hostrec'
+
+    def self.run(domain, url, title)
+      domain unless (title and domain and title.downcase.index(domain))
+    end
+
+  end
+
+end

data/plugins/hostrec_skip.rb

@@ -0,0 +1,23 @@
+module Murlsh
+
+  # skip showing host record for some domains
+  class HostrecSkip < Plugin
+
+    Hook = 'hostrec'
+
+    def self.run(domain, url, title)
+      domain unless Skips.include?(domain)
+    end
+
+    Skips = %w{
+      wikipedia.org
+      flickr.com
+      github.com
+      twitter.com
+      vimeo.com
+      youtube.com
+      }
+
+  end
+
+end

data/plugins/lookup_content_type_title.rb

@@ -0,0 +1,17 @@
+require 'murlsh'
+
+module Murlsh
+
+  # try to fetch the content type and title of a url
+  class LookupContentTypeTitle < Plugin
+
+    Hook = 'add_pre'
+
+    def self.run(url, config)
+      url.content_type = Murlsh.get_content_type(url.url)
+      url.title = Murlsh.get_title(url.url, :content_type => url.content_type)
+    end
+
+  end
+
+end

data/public/css/phone.css

@@ -1,3 +1,7 @@
+a.feed {
+  display : none;
+}
+
 body {
   margin : 0;
   padding : 0;
@@ -9,6 +13,14 @@ body {
   width : 290px;
 }
 
+#urls li {
+  overflow : hidden;
+}
+
 input#q {
   width : 130px;
 }
+
+input#url {
+  width : 160px;
+}

data/public/js/js.js

@@ -205,7 +205,12 @@ Murlsh.orientation_changed = function() {
 window.onorientationchange = Murlsh.orientation_changed;
 
 $(document).ready(function() {
-  Murlsh.orientation_changed();
+  if (Murlsh.is_iphone()) {
+    Murlsh.orientation_changed();
+    $('#urls li:first').prepend($('<a />').attr('href', '#bottom').text(
+      'bottom'));
+    $('#urls li:last').append($('<a />').attr('href', '#urls').text('top'));
+  }
   $('a').map(Murlsh.add_extra);
   $('#urls li:even').addClass('even');
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: murlsh
 version: !ruby/object:Gem::Version 
-  version: 0.2.4
+  version: 0.3.0
 platform: ruby
 authors: 
 - Matthew M. Boedicker
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-10-23 00:00:00 -04:00
+date: 2009-11-05 00:00:00 -05:00
 default_executable: murlsh
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -116,6 +116,9 @@ files:
 - lib/murlsh/url_server.rb
 - lib/murlsh/xhtml_response.rb
 - murlsh.gemspec
+- plugins/hostrec_redundant.rb
+- plugins/hostrec_skip.rb
+- plugins/lookup_content_type_title.rb
 - plugins/update_feed.rb
 - public/css/jquery.jgrowl.css
 - public/css/phone.css