Last active
December 21, 2015 14:59
-
-
Save kturney/6323837 to your computer and use it in GitHub Desktop.
A Ruby script to help make a list of all the beers at Untapped Fest Dallas 2013
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
require 'nokogiri' | |
require 'open-uri' | |
doc = Nokogiri::HTML(open('http://dallasbrewscene.com/brewlist.html')) | |
#File.open("/home/kyle/Downloads/untapped.html", "w") { |io| io.write(doc) } | |
#breweries = doc.css('div#accordion h4').map { |e| e.text } | |
#beers = doc.css('div#accordion p').map { |e| e.text.split(', ') } | |
#tapList = Hash[breweries.zip(beers)] | |
breweries = [] | |
beers = [] | |
cideries = [] | |
ciders = [] | |
atCiders = false | |
# Iterate through so that ciders can be separated | |
doc.at('//div[@id=\'accordion\']').search('//h4 | //p | //h2[text() = \'Ciders\']').each { |node| | |
case node.name | |
when 'h4' | |
if atCiders | |
cideries.push node.text | |
else | |
breweries.push node.text | |
end | |
when 'p' | |
if atCiders | |
ciders.push node.text.split(', ') | |
else | |
beers.push node.text.split(', ') | |
end | |
when 'h2' | |
atCiders = true | |
else | |
puts 'should never get here' | |
end | |
} | |
tapList = Hash[breweries.zip(beers)] | |
ciderList = Hash[cideries.zip(ciders)] | |
builder = Nokogiri::HTML::Builder.new { |doc| | |
doc.html { | |
doc.head { | |
doc.meta charset: 'UTF-8' | |
} | |
doc.body { | |
doc.div "Brewery Count: #{breweries.size}" | |
doc.div "Beer Count: #{beers.map{ |e| e.size }.reduce(:+)}" | |
doc.div "Cidery Count: #{cideries.size}" | |
doc.div "Cider Count: #{ciders.map{ |e| e.size }.reduce(:+)}" | |
# Use spaces instead of a list because it works better when pasting into Google Doc | |
doc.h2 'Beers' | |
tapList.sort.each { |brewery, beer| | |
doc.b brewery | |
beer.each { |e| doc.div " #{e}" } | |
} | |
doc.h2 'Ciders' | |
ciderList.sort.each { |cidery, cider| | |
doc.b cidery | |
cider.each { |e| doc.div " #{e}" } | |
} | |
} | |
} | |
} | |
# Use gsub! to fix Nokogiri/libxml2's auto escaping of '&' | |
File.open('taplist.html', 'w') { |file| file.write(builder.to_html.gsub!(' ', ' ')) } | |
puts 'DONE!' |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment