liveblog-indexer 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 56937c6f8eb2b40f417e73a5512fa60593647517
+  data.tar.gz: a87152912b799beab92e305565065ee391364b32
+SHA512:
+  metadata.gz: 9e99db1533e3279c4d4b70ddb58609fb5e234e8e9124d401bff3970d5e8d2ffa5671556d1ec0e6706c0212868bde0c469f5e07e5a85609838aa5494df7bf9426
+  data.tar.gz: 9d58190002e8cb8b490c61879f47901fe07c0ec374546dd52fae46d0c5c95ec8a6dce1c92eff850e957c0d776fe99f59577eb5c32f1bf46c2512a516140f385e

checksums.yaml.gz.sig

@@ -0,0 +1,5 @@
+5>Y�9��&��Ʈfh��/Ec;�	�V����
+���f�d�ߟ��Q��z��,�$��
+9�%ħ
+�U�����������.��=��o!,pʅ�\wφ��|�E�W����/A�FV^=��t��/���e:�Z9	��2�t�|�z_,����J�a�Ǹ}b���f���F��������+����亰�v��j'K�ѳ��(��^>g���l��
+���N�DEN٘C}��?HњCc��>��`%?j

data/lib/liveblog-indexer.rb

@@ -0,0 +1,76 @@
+#!/usr/bin/env ruby
+
+# file: liveblog-indexer.rb
+
+require 'xws'
+require 'json'
+require 'polyrex'
+
+
+class LiveBlogIndexer
+
+  def initialize(filepath=nil)
+
+    @master = if filepath and File.exists? filepath then
+      JSON.parse(File.read(filepath))
+    else
+      {}
+    end
+
+    @xws = XWS.new
+
+  end
+
+  def add_index(src)
+
+    doc = Rexle.new(RXFHelper.read(src).first )
+    sections = doc.root.xpath 'records/section'
+
+    sections.each do |section|
+
+      url = "%s/#%s" % [src[/^https?:\/\/[^\/]+(.*)(?=\/formatted.xml$)/,1], \
+                                                       section.attributes[:id]]
+
+      h = @xws.scan section.element('details')
+      
+      h.each do |k, v|
+
+        word, count = k, v
+
+        keyword = @master[word]
+
+        if keyword then
+
+          keyword[url] = count
+          
+        else
+
+          @master[word] = {}
+          @master[word][url] = count
+
+        end # /keyword
+      end # /h
+    end # /section
+  end # /add_index
+
+
+  def save(filepath=nil)
+
+    File.write filepath, @master.to_json
+    puts 'saved ' + File.basename(filepath)
+
+    px = Polyrex.new 'words/key[word]/location[url, wordcount]'
+
+    @master.each do |k,rows|
+      px.create.key(word: k) do |create|
+        rows.each do |k, v|
+          create.location url: k, wordcount: v
+        end
+      end
+    end
+    pxfilepath = File.join(File.dirname(filepath), 'search-index.xml')
+    px.save pxfilepath, options: {pretty: true}
+    puts 'saved ' + File.basename(pxfilepath)
+  end
+
+end

metadata

@@ -0,0 +1,108 @@
+--- !ruby/object:Gem::Specification
+name: liveblog-indexer
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- James Robertson
+autorequire: 
+bindir: bin
+cert_chain:
+- |
+  -----BEGIN CERTIFICATE-----
+  MIIDljCCAn6gAwIBAgIBATANBgkqhkiG9w0BAQUFADBIMRIwEAYDVQQDDAlnZW1t
+  YXN0ZXIxHjAcBgoJkiaJk/IsZAEZFg5qYW1lc3JvYmVydHNvbjESMBAGCgmSJomT
+  8ixkARkWAmV1MB4XDTE1MTExMjIxMTkyMVoXDTE2MTExMTIxMTkyMVowSDESMBAG
+  A1UEAwwJZ2VtbWFzdGVyMR4wHAYKCZImiZPyLGQBGRYOamFtZXNyb2JlcnRzb24x
+  EjAQBgoJkiaJk/IsZAEZFgJldTCCASIwDQYJKoZIhvcNAQEBBQADggEPADCCAQoC
+  ggEBAMjQ3f/YG2mPM5EMmrUlk2/+K3hOdZCCNIsuVYmfDaSVJO7T2A1hGDpp6qw2
+  NDCOLKXJ8B8MkIuSTZlsqIcM52RtZDLtjuFQvxlDysHjQdS89ab9NXSH+FtjEsnl
+  MAw+HB1na9+esmUfORTq3Qvf8/NVanJ0jpO78zZhHoIcH6/ZNxLGXEXhxCA3SSL/
+  RUwnF2ds4EmYjpYAF4DrtT7uRP80sOa5JDO9HWRIsJH70uZu/YNkzgD4qw9mssUW
+  yCDOOPPBXu1ZyCTYlkb8REMoZHCbIDASY/fJHmB+lfwI7WR10NauCNdru+d5L6db
+  lZ2+imBz1/JaibgbzwtUTkXbw8ECAwEAAaOBijCBhzAJBgNVHRMEAjAAMAsGA1Ud
+  DwQEAwIEsDAdBgNVHQ4EFgQUDiC6KdE35bCvxp4kHsAfxpJz4dEwJgYDVR0RBB8w
+  HYEbZ2VtbWFzdGVyQGphbWVzcm9iZXJ0c29uLmV1MCYGA1UdEgQfMB2BG2dlbW1h
+  c3RlckBqYW1lc3JvYmVydHNvbi5ldTANBgkqhkiG9w0BAQUFAAOCAQEAh7gf6A0R
+  5MN6ar8lV6/5ONlT9aljXGddCasCyC0pM94TIjH/2QblBWRodcb/oTGcnf28s9Zs
+  zBbBfl+YNUAS/y2a3Qj8/xgCC003tXgSuvXzrMlQ/o+2gDuLuGEkId8ceUXfUuXf
+  F4uMWpNiNk1Wt41dDGTzn9oCPt7Ssau3MqfiI5Lyl4lFQNqI+NlnsSkuJ3Fj7yOy
+  f1v1HUEOKUabrwhEuev8u1qZh+K4wUcsEkcvI06xLgObh2dbV2lW1Z0cOF7X8hWQ
+  mOnsi2V1CXpq2biJtSgD7mBx4cO9FXgbK3Xnsv45ygAPo6jj4Eb34udqz+0v88Ys
+  KoTqNQOniHAW2w==
+  -----END CERTIFICATE-----
+date: 2015-11-12 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: xws
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.1'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.1.1
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.1'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.1.1
+- !ruby/object:Gem::Dependency
+  name: polyrex
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.0.11
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.0.11
+description: 
+email: james@r0bertson.co.uk
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/liveblog-indexer.rb
+homepage: https://github.com/jrobertson/liveblog-indexer
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.4.8
+signing_key: 
+specification_version: 4
+summary: This gem is under development. Generates a Liveblog indexed file in JSON
+  format as well as Polyrex format.
+test_files: []

metadata.gz.sig

@@ -0,0 +1 @@
+|�U;\�@�M�hN��3��,��lV�0?�\�if*��+Z'5oX�1�Æ�Ԗ���[�g�*�TE`0��aVh�!��C�6�eA�ڒEO-M����Y����2�w�+Q��p��Z��)ɂZV�h2j���b���k�G�CV��\	-������˥���`�����[z&�k���Q�����M�p��<��F�q���U�����ػ��3�O��ۙhJ�������?P&����H�MD��{��/