govspeak 5.2.2 → 5.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 0b89deebe8f9edfbab700411aad9fe2a35920ea1
-  data.tar.gz: 0c2bdf232a565f530966c9cf5799e6d856690e35
+  metadata.gz: 117026a2b98a5909e89263c2d9f05091b492e8b2
+  data.tar.gz: b6f5a662a795899e4beb399d87ade54181f40e81
 SHA512:
-  metadata.gz: e70fdccd35a61cc375508ae191b2264ed294ad5d9c578d1776fdae7be705be5c21ed2dd8dc05fb28e97e2599faf655aaf164d24baab3cd986e8557fe1954e19a
-  data.tar.gz: 4d3bbd83d9a8b487bac5fc59bc1bb9ad55c6f93ea1fd489f9b44d41a68efd803a326da21ecfb3a2b16915b62031e381359440c2fab751144ade636c137dc53bb
+  metadata.gz: da21f89b2844d3174d1820ecd0aa991ea6a2aac40a58a5967549cdc128f71a46bfd3776e7667f7a28f0add49173c0c2ad0f50f8c7dfb4a9292f87ea2b7698c20
+  data.tar.gz: ae41a3cdb7664e6694f23c96d9f67f24304b813371eea173523627dd695f7e44ce78b00602791c0b8baf106b9e999336da6efe39c23a7334fc5875ed25a112f4

data/CHANGELOG.md

@@ -1,3 +1,6 @@
+## 5.3.0
+* Add a link extraction class for finding links in documents [#120](https://github.com/alphagov/govspeak/pull/120)
+
 ## 5.2.2
 * Fix rendering buttons with inconsistent linebreaks seen in publishing [#118](https://github.com/alphagov/govspeak/pull/118)
 

data/lib/govspeak.rb

@@ -11,6 +11,7 @@ require 'govspeak/html_sanitizer'
 require 'govspeak/kramdown_overrides'
 require 'govspeak/blockquote_extra_quote_remover'
 require 'govspeak/post_processor'
+require 'govspeak/link_extractor'
 require 'govspeak/presenters/attachment_presenter'
 require 'govspeak/presenters/contact_presenter'
 require 'govspeak/presenters/h_card_presenter'
@@ -94,6 +95,10 @@ module Govspeak
       Govspeak::StructuredHeaderExtractor.new(self).call
     end
 
+    def extracted_links
+      Govspeak::LinkExtractor.new(self).call
+    end
+
     def preprocess(source)
       source = Govspeak::BlockquoteExtraQuoteRemover.remove(source)
       @@extensions.each do |title,regexp,block|

data/lib/govspeak/link_extractor.rb

@@ -0,0 +1,28 @@
+module Govspeak
+  class LinkExtractor
+    def initialize(document)
+      @document = document
+    end
+
+    def call
+      @links ||= extract_links
+    end
+
+  private
+
+    def extract_links
+      document_anchors.map { |link| link['href'] }
+    end
+
+    def document_anchors
+      processed_govspeak.css('a:not([href^="mailto"])').css('a:not([href^="#"])')
+    end
+
+    def processed_govspeak
+      doc = Nokogiri::HTML::Document.new
+      doc.encoding = "UTF-8"
+
+      doc.fragment(@document.to_html)
+    end
+  end
+end

data/lib/govspeak/version.rb

@@ -1,3 +1,3 @@
 module Govspeak
-  VERSION = "5.2.2".freeze
+  VERSION = "5.3.0".freeze
 end

data/test/govspeak_link_extractor_test.rb

@@ -0,0 +1,42 @@
+require "test_helper"
+
+class GovspeakLinkExtractorTest < Minitest::Test
+  def document_body
+    %{
+## Heading
+
+[link](http://www.example.com)
+
+[link_two](http://www.gov.com)
+
+[not_a_link](#somepage)
+
+[mailto:](mailto:someone@www.example.com)
+    }
+  end
+
+  def doc
+    @doc ||= Govspeak::Document.new(document_body)
+  end
+
+  def links
+    doc.extracted_links
+  end
+
+  test "Links are extracted from the body" do
+    expected_links = ["http://www.example.com", "http://www.gov.com"]
+    assert_equal expected_links, links
+  end
+
+  test "Other content is not extracted from the body" do
+    refute_includes ["Heading"], links
+  end
+
+  test "Links are not extracted if they begin with #" do
+    refute_includes ["#somepage"], links
+  end
+
+  test "Links are not extracted if they begin with mailto:" do
+    refute_includes ["mailto:someone@www.example.com"], links
+  end
+end

metadata

@@ -1,15 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: govspeak
 version: !ruby/object:Gem::Version
-  version: 5.2.2
+  version: 5.3.0
 platform: ruby
 authors:
-- Ben Griffiths
-- James Stewart
+- GOV.UK Dev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-11-21 00:00:00.000000000 Z
+date: 2018-01-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: kramdown
@@ -251,8 +250,7 @@ description: |-
   A set of extensions to markdown layered on top of the kramdown
   library for use in the UK Government Single Domain project
 email:
-- ben@alphagov.co.uk
-- james.stewart@digital.cabinet-office.gov.uk
+- govuk-dev@digital.cabinet-office.gov.uk
 executables:
 - govspeak
 extensions: []
@@ -270,6 +268,7 @@ files:
 - lib/govspeak/html_sanitizer.rb
 - lib/govspeak/html_validator.rb
 - lib/govspeak/kramdown_overrides.rb
+- lib/govspeak/link_extractor.rb
 - lib/govspeak/post_processor.rb
 - lib/govspeak/presenters/attachment_presenter.rb
 - lib/govspeak/presenters/contact_presenter.rb
@@ -287,6 +286,7 @@ files:
 - test/govspeak_attachments_test.rb
 - test/govspeak_button_test.rb
 - test/govspeak_contacts_test.rb
+- test/govspeak_link_extractor_test.rb
 - test/govspeak_link_test.rb
 - test/govspeak_structured_headers_test.rb
 - test/govspeak_test.rb
@@ -320,18 +320,19 @@ signing_key:
 specification_version: 4
 summary: Markup language for single domain
 test_files:
-- test/blockquote_extra_quote_remover_test.rb
-- test/govspeak_test_helper.rb
-- test/govspeak_structured_headers_test.rb
-- test/govspeak_attachments_image_test.rb
 - test/govspeak_attachments_test.rb
-- test/test_helper.rb
 - test/govspeak_attachments_inline_test.rb
-- test/html_sanitizer_test.rb
-- test/govspeak_button_test.rb
-- test/with_deep_merge_test.rb
-- test/html_validator_test.rb
-- test/govspeak_test.rb
 - test/govspeak_link_test.rb
+- test/govspeak_link_extractor_test.rb
 - test/govspeak_contacts_test.rb
+- test/test_helper.rb
+- test/govspeak_test.rb
+- test/govspeak_attachments_image_test.rb
+- test/with_deep_merge_test.rb
+- test/html_validator_test.rb
+- test/govspeak_structured_headers_test.rb
+- test/blockquote_extra_quote_remover_test.rb
 - test/presenters/h_card_presenter_test.rb
+- test/govspeak_button_test.rb
+- test/govspeak_test_helper.rb
+- test/html_sanitizer_test.rb