transportscrapper 0.0.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 962ca5b21538ea50a440cb1091fc46177c4a7d4d2fe8a374ff484ea77c2f5b07
+  data.tar.gz: a52bfa7eaa60ccdf53f878b419c0ad674d055cf120873da5968bc10af9f395a0
+SHA512:
+  metadata.gz: ee7fdeb9062f4e7bea0e03c3548427d20b1d1e40ac11b32950b5bc6ca7bc24894ebd493fb4615aa2ebb9cd91a82a99a1d1ff7848ba12ad2c4470dbfaf9899976
+  data.tar.gz: 33424b4dc0dceb300c9887265c35c1eeebb9d3fcace9b674510eebb8f9d699f0e50bd2cb876bf02684d7ba3faefa0d7ee6b79f5f6e6bf1ac1022763862ca9a27

data/lib/TransportScrapper.rb

@@ -0,0 +1,148 @@
+require 'nokogiri'
+require 'watir'
+
+class TransportScrapper
+
+args = %w[--disable-infobars --headless window-size=1600,1200 --no-sandbox --disable-gpu --disable-dev-shm-usage]
+options = {
+       binary: ENV['GOOGLE_CHROME_BIN'],
+       prefs: { password_manager_enable: false, credentials_enable_service: false },
+       args:  args
+     }
+
+@browser = Watir::Browser.new(:chrome, options: options)
+
+@outwardArr = []
+@returnArr = []
+
+def self.scrapewithtime(value)
+@browser.goto value
+doc = Nokogiri::HTML.parse(@browser.html)
+#taking all the hotels list in HTML
+onwardtransport_snip = doc.search('div#JourneyRowsOut')
+onwardtransport_snip_Full= onwardtransport_snip.css('div.fullTable')
+onwardtransport_snip_array = onwardtransport_snip_Full.css('div.tableDataRow.Divtrue')
+
+
+#iterating over each hotel
+onwardtransport_snip_array.each do |transport_elemets|
+	depart=	transport_elemets.css('div.tableDataTime.d-T').text.strip!
+	arrival=transport_elemets.css('div.tableDataTime').text.strip!
+	
+	route_element=transport_elemets.css('div.tableRowSelector')
+	route=route_element.css('div.tableData').text.strip!
+	price_element=	transport_elemets.css('div.tableDataFare')
+	price=	price_element.css('span.tablePrice').text
+	
+	if arrival
+	arrivalArr=arrival.split(':')
+	val1 = arrivalArr[1].split(//).last(2).join
+	val2=	arrivalArr[2].split(//).first(2).join
+	arrival ="#{val1}:#{val2}"
+	end
+	
+	
+	if price
+	price.slice! "\u20AC"
+	end
+	
+	if route
+	route=route.gsub(" ","")
+	route=route.gsub("\n","")
+	end
+		
+	#new Hash is created and all the values are pushed into the hash map
+	output = Hash.new
+	output.store("depart", depart)
+	output.store("arrival", arrival)
+	output.store("price", price)
+	output.store("id", "#{depart}_#{arrival}_#{route}")
+	
+
+#pushing to array
+if(!@outwardArr.include?(output))
+	@outwardArr.push(output)
+end
+   end
+
+
+
+
+#taking all the hotels list in HTML
+returntransport_snip = doc.search('div#JourneyRowsRet')
+returntransport_snip_Full= returntransport_snip.css('div.fullTable')
+returntransport_snip_array = returntransport_snip_Full.css('div.tableDataRow.Divtrue')
+
+
+#iterating over each hotel
+returntransport_snip_array.each do |transport_elemets|
+	depart=	transport_elemets.css('div.tableDataTime.d-T').text.strip!
+	arrival=	transport_elemets.css('div.tableDataTime').text.strip!
+	route_element=transport_elemets.css('div.tableRowSelector')
+	route=route_element.css('div.tableData').text.strip!
+	price_element=	transport_elemets.css('div.tableDataFare')
+	price=	price_element.css('span.tablePrice').text
+	
+	if arrival
+	arrivalArr=arrival.split(':')
+	val1 = arrivalArr[1].split(//).last(2).join
+	val2=	arrivalArr[2].split(//).first(2).join
+	arrival ="#{val1}:#{val2}"
+	end
+	
+	if price
+	price.slice! "\u20AC"
+	end
+	
+	if route
+	route=route.gsub(" ","")
+	route=route.gsub("\n","")
+	end
+		
+	#new Hash is created and all the values are pushed into the hash map
+	output = Hash.new
+	output.store("depart", depart)
+	output.store("arrival", arrival)
+	output.store("price", price)
+	output.store("id", "#{depart}_#{arrival}_#{route}")
+
+#pushing to array
+if(!@returnArr.include?(output))
+	@returnArr.push(output)
+end
+   end
+
+puts '----------------------------------------------------------------------------'
+puts @outwardArr
+puts @returnArr
+puts '----------------------------------------------------------------------------'
+
+end
+
+def self.scrape(value)
+
+timeArr = ["0000","0400","0800","1200","1600","2000"]
+
+for time in timeArr do
+	temp = value 
+  joinstr ="&returntime=#{time}&departtime=#{time}"
+  temp = value+joinstr
+ 
+  self.scrapewithtime(temp)
+end
+
+  #returning array
+  result = Hash.new
+  result.store("outward" ,@outwardArr)
+  result.store("return" ,@returnArr)
+  
+return result
+
+end
+
+value ='https://national.buseireann.ie/?originStop=13500&destinationstop=55504&ticketType=2&departdate=27-nov-2019&returndate=27-nov-2019&adult=1&child=0&student=0&family=0'
+
+res = TransportScrapper.scrape(value)
+puts res
+end
+

metadata

@@ -0,0 +1,70 @@
+--- !ruby/object:Gem::Specification
+name: transportscrapper
+version: !ruby/object:Gem::Version
+  version: 0.0.2
+platform: ruby
+authors:
+- Niranjan Karunanithi
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2019-11-27 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: watir
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description: Scraps the transport data
+email: niranjankarunanidhi@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/TransportScrapper.rb
+homepage: http://rubygems.org/gems/transportscrapper
+licenses: []
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.0.3
+signing_key: 
+specification_version: 4
+summary: Scraps the transport data
+test_files: []