Last active
December 14, 2015 21:39
-
-
Save prwteas/5152804 to your computer and use it in GitHub Desktop.
Script to download images from www dot buttonlovers dot com
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
class Button | |
require 'mechanize' | |
def get_all_categories | |
a= Mechanize.new | |
linkdet=[] | |
page = a.get("https://www.buttonlovers.com/storefrontCommerce/categorybrowse.do?category_name=Buttons&path=Imported%2f%2f%2f%2fSlimline%E2%84%A2¤tPage=1&numResults=48") | |
cats = page.links_with(:href=>/.*subcategory.*numResults/) | |
cats.each do |cat| | |
quantity = cat.text.gsub(",","").scan(/\d+/).first | |
name = cat.text.scan(/\w+.?\w+/).first | |
href = cat.href | |
linkdet << Hash[:name=>name,:quantity=>quantity,:href=>href] | |
end | |
linkdet | |
end | |
def get_them_all | |
categories = self.get_all_categories | |
categories.map do |cate| | |
puts "\rGetting #{cate[:name]}" | |
link = cate[:href].gsub("numResults=48","numResults=#{cate[:quantity]}") | |
self.get_all_buttons("https://www.buttonlovers.com/storefrontCommerce/"+link,cate[:name]) | |
end | |
end | |
def get_all_buttons(url,dirname) | |
$stdout.sync=true | |
a=Mechanize.new | |
page = a.get(url) | |
(1...10).each do |k| | |
print "\rWe are waiting for #{k} secs before the page is prepared" | |
sleep(1) | |
end | |
products = a.get(url).links_with(:text=>/Product#/) | |
puts "\rThere are #{products.count} pics in total" | |
FileUtils.mkdir_p(dirname) | |
pr_count = products.count | |
products.each do |i| | |
at = i.attributes['onclick'].scan(/\d+/) | |
item_id = at.first | |
item_custom_id = at.last | |
img_text = i.attributes.text.partition(/\n\n/) | |
pr_name = img_text[0] | |
pr_id = img_text[2].scan(/\d+/).first | |
a=Mechanize.new | |
page = a.get("https://www.buttonlovers.com/storefrontCommerce/itemDetail.do?itm_id=#{item_id}&cust_item=#{item_custom_id}") | |
imageurl = page.search("div#browse-details-middle").css("img").first.attributes['src'].value.partition("&")[0] | |
unless File.exists?("#{dirname}/#{pr_id}.jpg") | |
download = a.get("http://www.buttonlovers.com#{imageurl}").save("#{dirname}/#{pr_id}.jpg") | |
end | |
unless File.exists?("#{dirname}/#{pr_id}.html") | |
download_text = File.open("#{dirname}/#{pr_id}.html","w"){|file| file.write(page.search("div#browse-details-left").to_html+page.search("div#product-aside").to_html);file.close} | |
end | |
print "\rThere are just #{pr_count = pr_count - 1} to go" | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment