nanoc-search 0.0.1

data/README.md

@@ -0,0 +1,44 @@
+# Search
+
+Index an item in an indextank compatible API.
+
+It use [nokogiri](http://nokogiri.org/) and indextank.
+
+## Install
+
+    gem install nanoc-search
+
+If you use bundler, add it to your Gemfile:
+
+    gem "nanoc-search", "~> 0.0.1"
+
+## Usage
+
+Add to *lib/default.rb*:
+
+```ruby
+require 'nanoc-search'
+```
+
+Add to *config.yaml*:
+
+    indextank:
+      index: index_name
+      api_url: http://:password@api
+
+Add a filter at the compile time to use it:
+
+```ruby
+compile '*' do
+  filter :search
+  filter :erb
+  layout 'default'
+end
+```
+
+## License
+
+(c) 2011 Pascal Widdershoven (https://github.com/PascalW/jekyll_indextank)
+(c) 2012 Stormz
+
+This code is free to use under the terms of the MIT license

data/lib/nanoc-search.rb

@@ -0,0 +1,73 @@
+# Derived from https://github.com/PascalW/jekyll_indextank
+require 'indextank'
+require 'nokogiri'
+
+class SearchFilter < Nanoc::Filter
+  identifier :search
+  type :text
+
+  def initialize(hash = {})
+    super
+
+    raise ArgumentError.new 'Missing indextank:api_url.' unless @config[:indextank][:api_url]
+    raise ArgumentError.new 'Missing indextank:index.' unless @config[:indextank][:index]
+
+    @last_indexed_file = '.nanoc_indextank'
+
+    load_last_timestamp
+
+    api = IndexTank::Client.new(@config[:indextank][:api_url])
+    @index = api.indexes(@config[:indextank][:index])
+  end
+
+  # Index all pages except pages matching any value in config[:indextank][:excludes]
+  # The main content from each page is extracted and indexed at indextank.com
+  # The doc_id of each indextank document will be the absolute url to the resource without domain name
+  def run(content, params={})
+    # only process item that are changed since last regeneration
+    if (!@last_indexed.nil? && @last_indexed > item.mtime)
+      return content
+    end
+
+    puts "Indexing page #{@item.identifier}"
+
+    while not @index.running?
+      # wait for the indextank index to get ready
+      sleep 0.5
+    end
+
+    page_text = extract_text(content)
+
+    @index.document(@item.identifier).add({
+          :text => page_text,
+          :title => @item[:title] || item.identifier
+      })
+    puts 'Indexed ' << item.identifier
+
+    @last_indexed = Time.now
+    write_last_indexed
+
+    content
+  end
+
+  def extract_text(content)
+    doc = Nokogiri::HTML(content)
+    doc.xpath('//*/text()').to_a.join(" ").gsub("\r"," ").gsub("\n"," ")
+  end
+
+  def write_last_indexed
+    begin
+      File.open(@last_indexed_file, 'w') {|f| Marshal.dump(@last_indexed, f)}
+    rescue
+      puts 'WARNING: cannot write indexed timestamps file.'
+    end
+  end
+
+  def load_last_timestamp
+    begin
+      @last_indexed = File.open(@last_indexed_file, "rb") {|f| Marshal.load(f)}
+    rescue
+      @last_indexed = nil
+    end
+  end
+end

metadata

@@ -0,0 +1,69 @@
+--- !ruby/object:Gem::Specification
+name: nanoc-search
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+  prerelease: 
+platform: ruby
+authors:
+- François de Metz
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2012-05-31 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: &9072360 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: *9072360
+- !ruby/object:Gem::Dependency
+  name: indextank
+  requirement: &9071560 !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: *9071560
+description: 
+email: francois@2metz.fr
+executables: []
+extensions: []
+extra_rdoc_files:
+- README.md
+files:
+- README.md
+- lib/nanoc-search.rb
+homepage: https://github.com/stormz/nanoc-plugins
+licenses: []
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.10
+signing_key: 
+specification_version: 3
+summary: Index items from nanoc site
+test_files: []