open_events 0.0.1

data/.gitignore

@@ -0,0 +1 @@
+html_cache/

data/README

@@ -0,0 +1,3 @@
+# Open Events
+
+Description coming

data/Rakefile

@@ -0,0 +1,7 @@
+require 'rake'
+require 'rake/testtask'
+require 'bundler'
+Bundler::GemHelper.install_tasks
+
+$LOAD_PATH.unshift File.join(File.dirname(__FILE__), 'lib')
+

data/boston/booksmith.rb

@@ -0,0 +1,39 @@
+require 'event_scraper'
+
+class Booksmith < EventScraper
+
+  def about
+    { 
+      title: 'Brookline Booksmith',
+      url: 'http://www.brooklinebooksmith.com/events/mainevent.html',
+      categories: %w(books speakers),
+      locations: %w(coolidge-corner)
+    }
+  end
+
+  def nodes
+    doc.search('strong a').
+      select {|x| 
+        x['href'] =~ %r{^http://www.brooklinebooksmith-shop.com/event/}}.
+      map {|x| 
+        x.ancestors.detect {|y| y.name == 'tr'}}
+  end
+
+  def event(n)
+    return n.inner_html
+    date = if (x = n.at('.entry-meta'))
+             x.inner_text
+           else
+             @res.last && @res.last[:date]
+           end
+    time = (x = n.at('li/text()')) && x.text.strip
+    link = n.at('.entry-title a')['href']
+    { 
+      date: date,
+      time: time,
+      title: n.at('.entry-title').inner_text,
+      link: link 
+    }
+  end
+end
+

data/boston/brattle.rb

@@ -0,0 +1,34 @@
+require 'event_scraper'
+
+class Brattle < EventScraper
+
+  def about
+    { 
+      title: 'Brattle Theater Special Events',
+      url: 'http://brattlefilm.org/category/calendar-2/special-events/',
+      categories: %w(movies),
+      locations: %w(harvard-square)
+    }
+  end
+
+  def nodes
+    doc.at('#calendarframe').xpath('./div')
+  end
+
+  def event(n)
+    date = if (x = n.at('.entry-meta'))
+             x.inner_text
+           else
+             @res.last && @res.last[:date]
+           end
+    time = (x = n.at('li/text()')) && x.text.strip
+    link = n.at('.entry-title a')['href']
+    { 
+      date: date,
+      time: time,
+      title: n.at('.entry-title').inner_text,
+      link: link 
+    }
+  end
+end
+

data/lib/cached_html.rb

@@ -0,0 +1,27 @@
+require 'open-uri'
+module CachedHtml
+
+  CACHE_DIR = 'html_cache'
+  `mkdir -p #{CACHE_DIR}`
+
+  def filename(url)
+    File.join(CACHE_DIR, munge(url))
+  end
+
+  def munge(url)
+    url.sub("http://", '').sub(/\W$/, '').gsub('/', '.').gsub(/\W/, '-')
+  end
+
+  def cached_html(url)
+    if File.size?(filename(url))
+      File.read(filename(url))
+    else
+      puts "Fetching #{url}"
+      res  = open(url)
+      html = res.read
+      File.open(filename(url), 'w') {|f| f.write html}
+      puts "Cached html to #{filename(url)}"
+      html
+    end
+  end
+end

data/lib/event_scraper.rb

@@ -0,0 +1,33 @@
+require 'cached_html'
+require 'nokogiri'
+
+class EventScraper
+  include CachedHtml
+  attr_accessor :doc
+
+  def initialize
+    @doc = Nokogiri::HTML.parse(self.html)
+  end
+
+  def html
+    @html ||= cached_html(about[:url])
+  end
+
+  def parse
+    @res = []
+    nodes.map {|n|
+      @res << event(n)
+    }
+    @res
+  end
+
+  def parse_test
+    nodes.each {|n|
+      puts n.inner_html
+      puts n.inner_text.gsub(/\s{2,}/, ' ').strip
+      puts '-' * 80
+    }
+  end
+
+end
+

data/lib/runner.rb

@@ -0,0 +1,29 @@
+$LOAD_PATH.unshift('scrapers')
+
+# TODO use opt parse to output HTML fragments for interative development
+
+
+class Runner
+  def initialize(venue)
+    require venue
+    klass = camelize(venue)
+    puts klass 
+    parser = Object.const_get(klass).new
+    puts parser.about.inspect
+    res = parser.parse
+    puts res
+  end
+
+  # from active support
+  def camelize(lower_case_and_underscored_word, first_letter_in_uppercase = true)
+    if first_letter_in_uppercase
+      lower_case_and_underscored_word.to_s.gsub(/\/(.?)/) { "::#{$1.upcase}" }.gsub(/(?:^|_)(.)/) { $1.upcase }
+    else
+      lower_case_and_underscored_word.to_s[0].chr.downcase + camelize(lower_case_and_underscored_word)[1..-1]
+    end
+  end
+end
+
+if __FILE__ == $0
+  Runner.new(ARGV.first)
+end

data/notes.txt

@@ -0,0 +1,9 @@
+
+A curated simple event scraper for select Cambridge, MA places.
+
+Plus Sinatra app.
+
+Develop most the scrapers on the command line, then develop the Sinatra
+app.
+
+

data/open_events.gemspec

@@ -0,0 +1,25 @@
+# -*- encoding: utf-8 -*-
+$:.push File.expand_path("../lib", __FILE__)
+#require "open_events/version"
+
+Gem::Specification.new do |s|
+  s.name        = "open_events"
+  s.version     = "0.0.1"
+  s.platform    = Gem::Platform::RUBY
+  s.required_ruby_version = '>= 1.9.0'
+
+  s.authors     = ["Daniel Choi"]
+  s.email       = ["dhchoi@gmail.com"]
+  #s.homepage    = "http://danielchoi.com/software/open_events.html"
+  s.summary     = %q{Events listings web scrapers and tools}
+  s.description = %q{An open-source repository of events listings web scrapers and tools}
+
+  s.rubyforge_project = "open_events"
+
+  s.files         = `git ls-files`.split("\n")
+  s.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
+  s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  s.require_paths = ["lib"]
+
+  s.add_dependency 'nokogiri'
+end

metadata

@@ -0,0 +1,76 @@
+--- !ruby/object:Gem::Specification 
+name: open_events
+version: !ruby/object:Gem::Version 
+  prerelease: 
+  version: 0.0.1
+platform: ruby
+authors: 
+- Daniel Choi
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2011-04-18 00:00:00 -04:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: nokogiri
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: "0"
+  type: :runtime
+  version_requirements: *id001
+description: An open-source repository of events listings web scrapers and tools
+email: 
+- dhchoi@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- .gitignore
+- README
+- Rakefile
+- boston/booksmith.rb
+- boston/brattle.rb
+- lib/cached_html.rb
+- lib/event_scraper.rb
+- lib/runner.rb
+- notes.txt
+- open_events.gemspec
+has_rdoc: true
+homepage: 
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: 1.9.0
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+requirements: []
+
+rubyforge_project: open_events
+rubygems_version: 1.6.1
+signing_key: 
+specification_version: 3
+summary: Events listings web scrapers and tools
+test_files: []
+