truncato 0.7.11 → 0.7.13

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 65813e94f63e796eff6633a5b52c0b8628cc93235dcc31e12e4d9127b1933a9f
-  data.tar.gz: dffa08e0d89e45bf95f739f7f31db30954d08fd1b0f8be4b476f9461ad368e31
+  metadata.gz: 98abd765f6cdb6de053cfe7322ab5df0de64f2a4c21c200fda656bdedb52cce3
+  data.tar.gz: ee299ddc104eb523cba8251d1da989a41bf3d25fa705612a1c9b50df7261816d
 SHA512:
-  metadata.gz: c9eff1f2e23b4fc7288d6463b35f65e64379ff55049b45c28893ec3312726da4470c461163c202d5f07cf463bbb3061af4d5d7714aa1ead07afc0171e0abf71e
-  data.tar.gz: 72cf0afab39b6abd5f279ec4395955cfe69f3fba5980dc15df675224fe26f6aea793fa90ee62a4526fa80f74305e7951987973a38c9efaa4fcf22e8c02ee94b7
+  metadata.gz: 82303b0a35c3e4465af5ab47f1bd1f5084ca22b8f98d52d36e1b7d0230a6b61f445ea02be3cb7e33d13c9d65acc1b36a990fdcc4705138f2c84b12102a7e1f37
+  data.tar.gz: 0ecb7576a3a1d9adeaf231f81c9f1385fcc05c020c4b37c13e5996a421fa7107372cbd068db862a664619ea91e983336aedd4ad1706c180b0ee643b1ac1cde3a

data/README.md

@@ -68,5 +68,3 @@ The results comparing truncato with other libs:
 ```ruby
 rake spec
 ```
-
-

data/Rakefile

@@ -4,29 +4,9 @@ rescue LoadError
   puts 'You must `gem install bundler` and `bundle install` to run rake tasks'
 end
 
-require 'rdoc/task'
-
-RDoc::Task.new(:rdoc) do |rdoc|
-  rdoc.rdoc_dir = 'rdoc'
-  rdoc.title    = 'MailgunRails'
-  rdoc.options << '--line-numbers'
-  rdoc.rdoc_files.include('README.rdoc')
-  rdoc.rdoc_files.include('lib/**/*.rb')
-end
-
-
-
-
-Bundler::GemHelper.install_tasks
-
-require 'rake/testtask'
-
-Rake::TestTask.new(:test) do |t|
-  t.libs << 'lib'
-  t.libs << 'test'
-  t.pattern = 'test/**/*_test.rb'
-  t.verbose = false
-end
+require "rake"
+require "rspec/core/rake_task"
 
+RSpec::Core::RakeTask.new(:spec)
 
-task default: :test
+task default: :spec

data/lib/truncato/truncated_sax_document.rb

@@ -6,7 +6,7 @@ class TruncatedSaxDocument < Nokogiri::XML::SAX::Document
 
   SINGLE_TAGS = %w{br img}
 
-  attr_reader :truncated_string, :max_length, :max_length_reached, :tail,
+  attr_reader :max_length, :max_length_reached, :tail,
               :count_tags, :filtered_attributes, :filtered_tags, :ignored_levels
 
   def initialize(options)
@@ -54,6 +54,10 @@ class TruncatedSaxDocument < Nokogiri::XML::SAX::Document
     close_truncated_document if max_length_reached
   end
 
+  def truncated_string
+    @truncated_buffer.join
+  end
+
   private
 
   def capture_options(options)
@@ -78,7 +82,7 @@ class TruncatedSaxDocument < Nokogiri::XML::SAX::Document
   end
 
   def init_parsing_state
-    @truncated_string = ""
+    @truncated_buffer = []
     @closing_tags = []
     @estimated_length = @count_tail ? tail_length : 0
     @max_length_reached = false
@@ -94,7 +98,7 @@ class TruncatedSaxDocument < Nokogiri::XML::SAX::Document
   end
 
   def append_to_truncated_string string, overriden_length=nil
-    @truncated_string << string
+    @truncated_buffer << string
     increase_estimated_length(overriden_length || string.length)
   end
 

data/lib/truncato/truncato.rb

@@ -6,7 +6,7 @@ module Truncato
       filtered_attributes: []
   }
 
-  ARTIFICIAL_ROOT_NAME = '__truncato_root__'
+  ARTIFICIAL_ROOT_NAME = 'truncato-artificial-root'
 
   # Truncates the source XML string and returns the truncated XML. It will keep a valid XML structure
   # and insert a _tail_ text indicating the position where content were removed (...).
@@ -26,12 +26,23 @@ module Truncato
   private
 
   def self.truncate_html source, options
+    source = unicode_normalize(source)
     self.do_truncate_html(source, options) ? self.do_truncate_html(with_artificial_root(source), options) : nil
   end
 
+  def self.unicode_normalize(string)
+    string.unicode_normalize
+  rescue Encoding::CompatibilityError
+    # By relying on rescue we don't have to maintain a list of compatible encodings.
+    string
+  end
+
   def self.do_truncate_html source, options
     truncated_sax_document = TruncatedSaxDocument.new(options)
-    parser = Nokogiri::HTML::SAX::Parser.new(truncated_sax_document)
+
+    # Only nokogiri >= 1.17 accept Encoding object, older needs a String as encoding
+    parser = Nokogiri::HTML::SAX::Parser.new(truncated_sax_document, source.encoding.to_s)
+
     parser.parse(source) { |context| context.replace_entities = false }
     truncated_string = truncated_sax_document.truncated_string
     truncated_string.empty? ? nil : truncated_string

data/lib/truncato/version.rb

@@ -1,3 +1,3 @@
 module Truncato
-  VERSION='0.7.11'
+  VERSION='0.7.13'
 end

metadata

@@ -1,11 +1,10 @@
 --- !ruby/object:Gem::Specification
 name: truncato
 version: !ruby/object:Gem::Version
-  version: 0.7.11
+  version: 0.7.13
 platform: ruby
 authors:
 - Jorge Manrubia
-autorequire: 
 bindir: bin
 cert_chain: []
 date: 2013-09-10 00:00:00.000000000 Z
@@ -48,30 +47,30 @@ dependencies:
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.14.1
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.14.1
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 10.1.1
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 10.1.1
+        version: '0'
 description: Ruby tool for truncating HTML strings keeping a valid HTML markup
 email: jorge.manrubia@gmail.com
 executables: []
@@ -90,8 +89,8 @@ files:
 homepage: https://github.com/jorgemanrubia/truncato
 licenses:
 - MIT
-metadata: {}
-post_install_message: 
+metadata:
+  allowed_push_host: https://rubygems.org
 rdoc_options: []
 require_paths:
 - lib
@@ -106,9 +105,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.7.6
-signing_key: 
+rubygems_version: 3.6.2
 specification_version: 4
 summary: A tool for truncating HTML strings efficiently
 test_files: []