Last active
July 28, 2017 10:54
-
-
Save LitvinenkoD89/d6fda8d33ba331d54cf89bfe332fac2c to your computer and use it in GitHub Desktop.
six02
This file contains hidden or 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
# SIX:02 {source_name: 'six02', batch_number: 6, request_id: 6000000074, request_name: 'SIX:02'} | |
scraper_service.scrape do |browser, scraper, init_vars| | |
easy_seeder = Library.lib('EasySeeder') | |
easy_extractor = Library.lib('EasyExtractor') | |
fetcher_agent = Library.lib('FetcherAgent') | |
easy_seeder.seed(source_name: init_vars[:source_name]) do | |
loop_postal_codes "us", radius: 10 do |zip, lat, long| | |
queue_url "http://www.six02.com/stores" | |
end | |
end | |
easy_extractor.extract( | |
source_name: init_vars[:source_name], | |
scraper: scraper, | |
batch_number: init_vars[:batch_number], | |
request_id: init_vars[:request_id], | |
request_name: init_vars[:request_name], | |
) do | |
find_pages page_format: :html do |url, parser_page, page| | |
page.search('.sep_locations_store').each do |store| | |
id = store.attr('data-store-id') | |
address_container = store.at('address').inner_html.gsub("\t", '').gsub("\r", '').gsub("\n", '') | |
tmp = address_container.split('<br>') | |
tel = tmp.pop | |
address_container_html = tmp | |
adr = tmp.pop | |
street1,street2 = tmp | |
city, state_zip = adr.split(', ') | |
state, zip_code = state_zip.split(' ') | |
href = store.at('.sep_location_store_details').attr('href') | |
location_url = URI.join(url, href).to_s | |
doc_id = store_doc({ | |
store_id: id, | |
# brand: brand, | |
type: "SIX:02", | |
property_id: id, | |
name: '', # required | |
address_1: street1, | |
address_2: street2, | |
city: city, | |
state: state, | |
zipcode: zip_code, | |
country: '', | |
# lat: lat, | |
# long: long, | |
address_container_html: address_container_html, | |
# map_link: store.parent.parent.at('.googlemap').attr('name'), | |
location_url: location_url, | |
flags: { | |
} | |
}) | |
seeder.queue_url location_url, { | |
page_type: 'profile', | |
doc_id: doc_id | |
} | |
end | |
end | |
# Find profile pages | |
find_pages page_format: :html, page_type: 'profile' do |url, parser_page, page| | |
doc = find_location(parser_page[:doc_id]) | |
if page.at('[itemprop="map"]').nil? | |
lat, long = [nil, nil] | |
else | |
map_link = page.at('[itemprop="map"]').attr('content') | |
lat, long = map_link.split('/@')[1].split(',') | |
end | |
brand = page.at('[itemprop="brand"]').nil? ? nil : page.at('[itemprop="brand"]').attr('content') | |
doc[:lat] = lat | |
doc[:long] = long | |
doc[:brand] = brand | |
store_doc doc | |
end | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment