From: "F. Senault" <fred@...>
Date: 2009-03-13T17:18:04+09:00
Subject: Re: read html then output one line of html if found?

Le 12 mars � 21:36, Mmcolli00 Mom a �crit :

> Here is my snippet.
> 
> require "fileutils"
> 
> Dir["C:/Respecs/*.html"].each do |htmlfile|
>     readhtml = File.read(htmlfile)
>     if readhtml.include?("seconds") == true
>      htmlbase = File.basename(htmlfile)
>      puts htmlbase 
> end

For the easy way, try readlines and grep : 

>> h = File.readlines('f1.txt')
=> ["<h1>hhhhhh</h1>\n", "<h2>20 seconds</h2>\n", "<p>Blah.</p>\n",
"\n"]
>> h.grep(/seconds/)
=> ["<h2>20 seconds</h2>\n"]

For a more sophisticated (and time-consuming) approach, try an HTML
parser like Hpricot :

>> require "hpricot"
=> true
>> doc = Hpricot(File.read('f1.txt'))
=> #<Hpricot::Doc {elem <h1> "hhhhhh" </h1>} "\n" {elem <h2> "20
seconds" </h2>} "\n" {elem <p> "Blah." </p>} "\n\n">
>> doc.children.select { |e| e.inner_html =~ /seconds/ }
=> [{elem h2 "20 seconds" h2}]

HTH.

Fred
-- 
Everyone is bad in their own way. Finding out each person's unique way
of being bad is most of the fun of getting to know them.
                                                            (Lee Wilson)