Need help parsing HTML with Hpricot

ruby-talk
1

I'm having trouble understanding Hpricot (thanks to an abominable lack
of documentation). I'm trying to parse HTML of the following nature:

This is one line of text<br />
This is another line of text<br />
It keeps going on like this<br />
<br />
Until a new paragraph is started<br />
Otherwise, it's just more of the same<br />

    I know, it looks simple but, frankly, I have no clue how to parse this
with Hpricot. Particularly, I don't know how to single out the lines of
text in between the "br" tags. This is important 'cause I need to know
where the line breaks are in the text, as well as the new paragraphs.
    Does anyone know how to do this with Hpricot?
    Thank you...

2

You can try each_child.

I will use each_child_with_index to show you what I mean:

Put your raw HTML text into @text

@parsed_html = Hpricot(@text)
@parsed_html.each_child_with_index do |c,i|
  puts "Line #{i}: #{c.to_s.strip}"
end

Produces:

Line 0: This is one line of text
Line 1: <br />
Line 2: This is another line of text
Line 3: <br />
Line 4: It keeps going on like this
Line 5: <br />
Line 6:
Line 7: <br />
Line 8: Until a new paragraph is started
Line 9: <br />
Line 10: Otherwise, it's just more of the same
Line 11: <br />
Line 12:

Hope that helps.

Mikel

···

On 10/25/07, Just Another Victim of the Ambient Morality <ihatespam@hotmail.com> wrote:

    I'm having trouble understanding Hpricot (thanks to an abominable lack
of documentation). I'm trying to parse HTML of the following nature:

This is one line of text<br />
This is another line of text<br />
It keeps going on like this<br />
<br />
Until a new paragraph is started<br />
Otherwise, it's just more of the same<br />

    I know, it looks simple but, frankly, I have no clue how to parse this
with Hpricot. Particularly, I don't know how to single out the lines of
text in between the "br" tags. This is important 'cause I need to know
where the line breaks are in the text, as well as the new paragraphs.
    Does anyone know how to do this with Hpricot?
    Thank you...

3

Try http://code.whytheluckystiff.net/hpricot/wiki/AnHpricotShowcase
for examples and some better documentation. It helped me a lot to
solve my problems.

···

2007/10/25, Just Another Victim of the Ambient Morality <ihatespam@hotmail.com>:

    I'm having trouble understanding Hpricot (thanks to an abominable lack
of documentation). I'm trying to parse HTML of the following nature:

4

Of course... you could also do:

require 'rubygems'
require 'hpricot'

text =<<HERE
This is one line of text<br />
This is another line of text<br />
It keeps going on like this<br />
<br />
Until a new paragraph is started<br />
Otherwise, it's just more of the same<br />
HERE

class String
  def not_needed?
    self.strip == "<br />" ? true : false
  end
end

@parsed_html = Hpricot(text)
@paragraphs = Array.new
@parsed_html.each_child_with_index do |c,i|
  line = c.to_s.strip
  if line == ""
    puts "<p>#{@paragraphs}</p>"
    @paragraphs.clear
  else
    @paragraphs << "#{line} " unless line.not_needed?
  end
end

Which produces:

<p>This is one line of text This is another line of text It keeps
going on like this </p>
<p>Until a new paragraph is started Otherwise, it's just more of the same </p>

Now... don't pick on my favorite HTML parser again! :smiley: Just ask nicely :slight_smile:

Mikel