Consider the following code:
require 'nokogiri' # v1.5.2
doc = Nokogiri.XML('<body><a name="foo">ick</a></body>')
puts doc.to_html
#=> <body><a name="foo">ick</a></body>
puts doc.to_xml
#=> <?xml version="1.0"?>
#=> <body>
#=> <a name="foo">ick</a>
#=> </body>
puts doc.to_xhtml
#=> <body>
#=> <a name="foo" id="foo">ick</a>
#=> </body>
Notice the new id attribute that has been created.
- Who is responsible for this, Nokogiri or libxml2?
- Why does this occur? (Is this enforcing a standard?)
The closest I can find is this spec describing how you may put both anidandnameattribute with the same value. - Is there any way to avoid this, given the desire to use the
to_xhtmlmethod on input that may have<a name="foo">?
This problem arises because I have some input I am parsing with an id attribute on one element and a separate element with a name attribute that happens to conflict.
No comments:
Post a Comment