amazon-search 1.1.10 → 1.1.11

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4fba2f78c6dace02da01d3179202847397bb005e
-  data.tar.gz: b24dae4f77ff25cd2f389b67e25947489007ef4f
+  metadata.gz: c35d7c9c08987d534c8430d25e91206c9ba98571
+  data.tar.gz: 7d26df4b2233e9b6bd47a521eb50be6977eb62b3
 SHA512:
-  metadata.gz: ee51c4bfcec988670566bd17d992a01dfb4d30f522796cb4faab35b7aa5dba5ba53a25dad24d93523585dd0311c97458362cf11734fd08ec66c1e4fb9deaf470
-  data.tar.gz: ef90cc57e7416fc72ffce70a8f24d19a3683cd07d75341cbab702d4b38f401384d1b1bee7a24b6511d11cca858f7b341d180b1c457a9712ba4d889cb65ac3672
+  metadata.gz: ec092cf9867ff4bda6a324936cbf557a6e0cbee74ccd2d96780ad773c092f9e130d927e341bf049c8a6f0fe4f5e996f877f6a9652554db62bad39f8a5f95d32e
+  data.tar.gz: f0396b0589455cdc8615f485f557192957ccbe9e7bb0002a22b47e2c7fffd708cd390dfd01b87ffe7a88f54de0df10da816b0069271edee1f60d34cdf110ef93

data/Readme.rdoc

@@ -10,7 +10,7 @@ This is a tool that does not require configuration of Amazon's API.  The functio
 
 == EXAMPLE
 
-    require 'amazon/search'
+    require 'amazon-search'
     
     # Search for products by keywords and store results
     Amazon::Search.find_products "ruby"

data/amazon-search.gemspec

@@ -0,0 +1,25 @@
+
+
+Gem::Specification.new do |gem|
+  gem.name        = %q{amazon-search}
+  gem.version     = '1.1.11'
+  gem.date        = '2015-09-18'
+  gem.platform = Gem::Platform::RUBY
+  gem.required_ruby_version = '>= 1.8'
+
+  gem.files = `git ls-files`.split("\n")
+  gem.test_files = `git ls-files -- test/*`.split("\n")
+
+  gem.summary     = "A simple screenscraper to search Amazon"
+  gem.description = "Simple screenscraper to search Amazon and return product titles, urls, image href, etc."
+  gem.authors     = ["John Mason"]
+  gem.email       = 'mace2345@gmail.com'
+  gem.homepage    = 'https://github.com/m8ss/amazon-search' 
+  gem.license       = 'MIT'
+
+  gem.add_runtime_dependency('mechanize',    '~> 2.7')
+
+end
+
+
+

data/amazon-search.rb

@@ -0,0 +1,93 @@
+#!/usr/bin/env ruby
+
+require 'mechanize'
+
+module Amazon
+
+class Search
+    def self.find_products(keywords)
+        #--------- submit the search form with keywords ---------------------
+        agent = Mechanize.new
+        main_page = agent.get("http://amazon.com")
+        search_form = main_page.form_with :name => "site-search" # find the search form in Amazon
+
+        search_form.field_with(:name => "field-keywords").value = keywords # sets value of search box
+        search_results = agent.submit search_form # submits form
+
+        #--------- scan each page and store the results ---------------------
+        @product_divs = []
+        page_num = 0
+        next_page = agent.get(search_results.uri) # initial search results are the first page
+
+        last_page_num = search_results.search '//*[contains(concat( " ", @class, " " ), concat( " ", "pagnDisabled", " " ))]'
+        last_page_num = last_page_num.text.to_i # change to int for upcoming iteration instructions
+
+        last_page_num.times do # cycle all pages and stop on last page
+            page_num += 1
+            page = agent.get(next_page.uri) # load the next page
+            
+            @product_divs << page.search('//li[starts-with(@id, "result")]') # store the div of each product
+           
+            next_page_link = page.link_with text: /Next Page/ # find the next page link   
+            next_page = next_page_link.click unless page_num == last_page_num # click to next page unless on last page
+        end # ends pagination loop
+
+        puts "\n\n(end of search results)"
+    end
+
+
+    def self.display_results
+        # nokogiri syntax is needed when iterating...not mechanize!
+        product_divs.each do |product|
+            
+            #--------- nokogiri select html sections from css ---------------------
+            title = product.at_css(".s-access-title")
+            seller = product.at_css(".a-row > .a-spacing-none") #".a-spacing-small .a-spacing-none"
+            price = product.at_css(".s-price")
+            stars = product.at_css(".a-icon-star")
+            reviews = product.at_css("span+ .a-text-normal") # ".a-span-last .a-spacing-mini > span+ .a-text-normal"
+            image = product.at_css(".s-access-image")
+            url = product.at_css(".a-row > a") 
+            
+            #--------- avoid the related items gotchas ---------------------  
+            if title == nil # if it's nil it's prob an ad 
+                break
+            else
+                title = title.text
+                
+                if seller == nil # if seller is nil put unknown
+                    seller = "Unknown"
+                else
+                    seller = seller.text
+                    if price == nil # no price? prob not worthy item
+                    break
+                    
+                    else
+                        price = price.text
+                        if stars == nil
+                            break
+                            
+                        else
+                            stars = stars.text
+                            reviews = reviews.text
+                            image = image['src'] 
+                            url = url['href']
+
+                            # errors properly avoided, now puts the results   
+                            STDOUT.puts "--"*50
+                            STDOUT.puts "title: \t\t#{title}"
+                            STDOUT.puts "seller: \t#{seller}"
+                            STDOUT.puts "price: \t\t#{price}"
+                            STDOUT.puts "stars: \t\t#{stars}"
+                            STDOUT.puts "reviews: \t#{reviews}"
+                            STDOUT.puts "image url: \t#{image}"
+                            STDOUT.puts "product url: \t#{url}"
+                            
+                        end # ends nil price
+                    end # ends nil stars
+                end # ends nil seller 
+            end # ends nil product 
+        end # ends each product div iteration (page is finished)    
+    end # ends display_results
+end # ends Search Class
+end # ends Amazon Module

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: amazon-search
 version: !ruby/object:Gem::Version
-  version: 1.1.10
+  version: 1.1.11
 platform: ruby
 authors:
 - John Mason
@@ -32,6 +32,8 @@ extensions: []
 extra_rdoc_files: []
 files:
 - Readme.rdoc
+- amazon-search.gemspec
+- amazon-search.rb
 homepage: https://github.com/m8ss/amazon-search
 licenses:
 - MIT