What would be the fastest way to do this.
I have may html documents that might (or might not) contain the word "Instructions" followed by several lines of instructions. I want to parse these pages that contain the word "Instructions" and the lines that follow.
This is not the most "correct" way, but will work mostly. Use a regular expression to find the strings:ruby regex
The regex you want is something like /instructions([^<]+)/. This assumes that you are ending with a < character.
You can start by just testing if a document matches:
if open('docname.html').read =~ /Instructions/
# Parse to remove the instructions.
end
I'd recommend using Hpricot to then extract the part you want - this will be more or less difficult depending on how your html is structured. Please post some more details about the structure if you want some more specific help.
Maybe something along this lines
require 'rubygems'
require 'nokogiri'
def find_instructions doc
doc.xpath('//body//text()').each do |text|
instructions = text.content.select do |line|
# flip-flop matches all sections starting with
# "Instructions" and ending with an empty line
true if (line =~ /Instructions/)..(line =~ /^$/)
end
return instructions unless instructions.empty?
end
return []
end
puts find_instructions(Nokogiri::HTML(DATA.read))
__END__
<html>
<head>
<title>Instructions</title>
</head>
<body>
lorem
ipsum
<p>
lorem
ipsum
<p>
lorem
ipsum
<p>
Instructions
- Browse stackoverflow
- Answer questions
- ???
- Profit
More
<p>
lorem
ipsum
</body>
</html>