Created
May 10, 2012 05:42
-
-
Save topperge/2651266 to your computer and use it in GitHub Desktop.
Quick and Dirty Script to Grab all the Breweries and Beers for Savor and match them to Beer Advocate Profiles
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://www.savorcraftbeer.com/beer-food/savor-breweries/")) | |
@beerUrls = doc.xpath('//div[@class="entry"]/p/a[@href]').map { |link| link['href'] } | |
puts 'Title,Brewery,Table,Style,Food,Beer Advocate Url,Rating,ABV,IBU' | |
@beerUrls.each do |beerUrl| | |
beerTitle = '' | |
beerBrewery = '' | |
beerTable = '' | |
beerStyle = '' | |
beerFood = '' | |
getTheBeerBeerAdvocateUrl = '' | |
beerAdvocateUrl = '' | |
beerRating = '' | |
beerABV = '' | |
beerIBU = '' | |
beer = Nokogiri::HTML(open(beerUrl)) | |
beerTitle = beer.xpath('//article/h1').text | |
tags = beer.xpath("//span[@class='tags']/a") | |
if tags.length == 3 | |
beerBrewery = tags[0].text.gsub('LLC','').gsub('Co.','').gsub('Beer','').gsub('Brewing','').gsub('Company','').gsub('Brewery','') | |
beerTable = tags[1].text.split[2] | |
beerStyle = tags[2].text | |
elsif tags.length == 2 | |
beerBrewery = tags[0].text.gsub('LLC','').gsub('Co.','').gsub('Beer','').gsub('Brewing','').gsub('Company','').gsub('Brewery','') | |
beerTable = 'Supporters Circle' | |
beerStyle = tags[1].text | |
elsif tags.length == 1 | |
beerBrewery = tags[0].text.gsub('LLC','').gsub('Co.','').gsub('Beer','').gsub('Brewing','').gsub('Company','').gsub('Brewery','') | |
end | |
beerFood = beer.xpath("//div/p[@class='pairings']").text | |
#Get Beer Advocate Rating | |
beerAdvocateUrl = URI.escape('http://beeradvocate.com/search?qt=beer&q='+beerTitle+' '+beerBrewery) | |
beerAdvocate = Nokogiri::HTML(open(beerAdvocateUrl)) | |
@beerAdvocateList = beerAdvocate.xpath("//td[@id='mainContent']/div/ul/li/a[@href]").map { |link| link['href'] } | |
if @beerAdvocateList.length > 0 | |
getTheBeerBeerAdvocateUrl = 'http://beeradvocate.com' + @beerAdvocateList[0] | |
getTheBeerBeerAdvocate = Nokogiri::HTML(open(getTheBeerBeerAdvocateUrl)) | |
beerRating = getTheBeerBeerAdvocate.xpath("//span[@class='BAscore_big']").first.text if !getTheBeerBeerAdvocate.xpath("//span[@class='BAscore_big']").first.nil? | |
firstABVPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.index('ABV') | |
secondABVPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.index('ABV', firstABVPosition+1) if !firstABVPosition.nil? | |
percentABVPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.index('%', firstABVPosition) if !firstABVPosition.nil? | |
lastSpacePosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.rindex(' ', percentABVPosition) if !firstABVPosition.nil? & !percentABVPosition.nil? | |
beerABV = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.slice(lastSpacePosition, percentABVPosition - lastSpacePosition + 1) if !firstABVPosition.nil? & !percentABVPosition.nil? & !lastSpacePosition.nil? | |
firstIBUPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.index('IBU') | |
spaceBeforeIBUPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.rindex("\r", firstIBUPosition) if !firstIBUPosition.nil? | |
spaceBeforeIBUPosition = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.rindex(" ", firstIBUPosition - 1) if spaceBeforeIBUPosition.nil? | |
beerIBU = getTheBeerBeerAdvocate.xpath("//table/tr/td/table/tr[2]/td").text.slice(spaceBeforeIBUPosition+1, 2) if !firstIBUPosition.nil? & !spaceBeforeIBUPosition.nil? | |
end | |
beerTitle = '' if beerTitle.nil? | |
beerBrewery = '' if beerBrewery.nil? | |
beerTable = '' if beerTable.nil? | |
beerStyle = '' if beerStyle.nil? | |
beerFood = '' if beerFood.nil? | |
getTheBeerBeerAdvocateUrl = '' if getTheBeerBeerAdvocateUrl.nil? | |
beerAdvocateUrl = '' if beerAdvocateUrl.nil? | |
beerRating = '' if beerRating.nil? | |
beerABV = '' if beerABV.nil? | |
beerIBU = '' if beerIBU.nil? | |
puts beerBrewery.strip+','+beerTitle.strip+','+beerTable.strip+','+beerStyle.strip+','+beerFood.strip+','+getTheBeerBeerAdvocateUrl.strip+','+beerRating.strip+','+beerABV.strip+','+beerIBU.strip | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment