Preserving HTML entities
Alex Dunae
HTML entities when parsing HTML.
and friends are all converted to actual characters; I need them
to be preserved. I've tried various combinations of the
XML::ParseOptions Is there any way to preserve the entities?
Thanks in advance for any help.
Aaron Patterson
When you ask for "text", it is converted to the actual characters. If
you ask for inner_html, it will return the escaped values. Here is an
example:
require 'nokogiri'
f = Nokogiri.HTML('<body>hello world</body>')
node = f.css('body')
p node.text
p node.inner_html
Hope that helps!
--
Aaron Patterson
http://tenderlovemaking.com/
Alex Dunae
both .text and .inner_html return the same thing:
=> "hello world"
I've tried this on two machines:
$ nokogiri -v
---
warnings: []
nokogiri: 1.4.3.1
ruby:
version: 1.9.1
platform: i386-darwin10
engine: ruby
libxml:
binding: extension
compiled: 2.7.6
loaded: 2.7.6
$ nokogiri -v
---
warnings: []
ruby:
engine: mri
version: 1.8.7
platform: x86_64-linux
libxml:
loaded: 2.6.32
binding: extension
compiled: 2.6.32
nokogiri: 1.4.3.1
Any insight would be greatly appreciated.
On Nov 4, 12:32 pm, Aaron Patterson <aaron.patter...@gmail.com> wrote:
Aaron Patterson
I'm really not sure what to tell you. I can't reproduce the problem.
Here is a screencast of what I did:
http://www.youtube.com/watch?v=YEY_xixuOes
Am I doing something different than you? Can you try the code that I
did in the video?
Alex Dunae
On Nov 10, 10:00 am, Aaron Patterson <aaron.patter...@gmail.com>
wrote:
Aaron Patterson
> I think we're doing the same thing: http://www.youtube.com/watch?v=ZJvLh1pcmbo
AHA! It's the ruby version. Seems different with 1.9, though I'm not sure why.
I'm investigating, but this seems like a bug.
Alex Dunae
works.
The setup that did work was MRI 1.8.7 on Mac (run under RVM):
---
warnings: []
ruby:
engine: mri
version: 1.8.7
libxml:
loaded: 2.7.6
binding: extension
compiled: 2.7.6
nokogiri: 1.4.3.1
Setups that did not work:
$ nokogiri -v
---
warnings: []
nokogiri: 1.4.3.1
ruby:
platform: x86_64-darwin10.4.0
Aaron Patterson
> I experimented a bit more and have finally found a combination that
> works.
Okay, I can explain the behavior now. Basically, the problem boils
down to encoding.
In Ruby 1.9, we examine the encoding of the string you're feeding to
Nokogiri. If the input string is "utf-8", the document is assumed to
be a UTF-8 document. When you output the document, since " " can
be represented as a UTF-8 character, it is output as that UTF-8
character.
In 1.8, since we cannot detect the encoding of the document, we assume
binary encoding and allow libxml2 to detect the encoding.
If you set the encoding of the input document to binary, it will give
you back the entities you want. Here is some code to demo:
require 'nokogiri'
html = '<body>hello world</body>'
f = Nokogiri.HTML(html)
node = f.css('body')
p node.inner_html
f = Nokogiri.HTML(html.encode('ASCII-8BIT'))
node = f.css('body')
p node.inner_html
I posted a youtube video too! :-)
http://www.youtube.com/watch?v=X2SzhXAt7V4
Alex Dunae
fantastic. Much appreciated.
On Nov 11, 11:10 am, Aaron Patterson <aaron.patter...@gmail.com>
wrote: