Crystagiri Versions Save

News

• New parsing engine #12 (thanks to @giuseongit )
• Extends Crystagiri::Tag API #13 (thanks to @giuseongit )

News:

• Add possibilty to follow redirect link #11 (thanks to @giuseongit )

• Make block return #5

Initial release

In this first version you can only parse Html file represented by Crystagiri::HTML. You can instanciate it from :

doc = Crystagiri::HTML.new "<h1>Crystagiri is awesome!!</h1>"
doc = Crystagiri::HTML.from_file "README.md"
doc = Crystagiri::HTML.from_url "http://example.com/"

Then you can search all XML::Node by:

• HTML tag

doc.tag("h2") { |node| puts node }
# => <h2> .. </h2>
# => <h2> .. </h2>
# => <h2> .. </h2>

• HTML id

puts doc.at_id("main-content") 
# => <div id="main-content"> ... </div>

• HTML class attribute

doc.class("summary") { |node| puts node }
# => <div class="summary"> .. </div>
# => <div class="summary"> .. </div>
# => <div class="summary"> .. </div>

new feature

• Implement css query

doc.css("#main-content ol.steps") {|node| puts node}
# => <ol class="steps"> .. </ol>
doc.css("#body>quote.introduction") {|node| puts node}
# => <quote class="introduction"> .. </quote>

Corrections

• Rename Crystagiri::HTML.class method as Crystagiri::HTML.where_class.
• To be more logic, I renamed also Crystagiri::HTML.tag into Crystagiri::HTML.where_tag
• Correct a bug when user search only a class with Crystagiri::HTML.css

feature

• Add anew Crystagiri::Tag class who include a XML::Node variable and add more usefull method
• add a benchmark between Crystagiri & Nokogiri

Correction

• correct impossibility to found a node who contains a number in its classname

Correction

• correct impossibility to found a node who contains a number in its classname

Correction

• add a bulletproof CSS to XPath converter