DEV: Small Amazon fixes

lib/onebox/engine/amazon_onebox.rb

@@ -45,7 +45,11 @@ def http_params
       private
 
       def match
-        @match ||= @url.match(/(?:d|g)p\/(?:product\/|video\/detail\/)?(?<id>[^\/]+)(?:\/|$)/mi)
+        @match ||= @url.match(/(?:d|g)p\/(?:product\/|video\/detail\/)?(?<id>[A-Z0-9]+)(?:\/|\?|$)/mi)
+      end
+
+      def link
+        @link ||= ::Onebox::Helpers.uri_encode(url)
       end
 
       def image
@@ -60,6 +64,10 @@ def image
         end
 
         if (landing_image = raw.css("#landingImage")) && landing_image.any?
+          attributes = landing_image.first.attributes
+
+          return attributes["data-old-hires"].to_s if attributes["data-old-hires"]
+
           landing_image.first["src"].to_s
         end
 
@@ -110,7 +118,7 @@ def data
           end
 
           result = {
-            link: link,
+            link: url,
             title: title,
             by_info: authors,
             image: og.image || image,
@@ -141,7 +149,7 @@ def data
           end
 
           result = {
-            link: link,
+            link: url,
             title: title,
             by_info: authors,
             image: og.image || image,
@@ -157,7 +165,7 @@ def data
         else
           title = og.title || CGI.unescapeHTML(raw.css("title").inner_text)
           result = {
-            link: link,
+            link: url,
             title: title,
             image: og.image || image,
             price: price

lib/onebox/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Onebox
-  VERSION = "2.2.9"
+  VERSION = "2.2.10"
 end

spec/lib/onebox/engine/amazon_onebox_spec.rb

@@ -73,6 +73,13 @@ def check_link(tdl, link)
         expect(described_class.new("http://www.amazon.fr/gp/product/B01BYD0TZM").url)
           .to eq("https://www.amazon.fr/dp/B01BYD0TZM")
       end
+
+      let(:long_url) { "https://www.amazon.ca/gp/product/B087Z3N428?pf_rd_r=SXABADD0ZZ3NF9Q5F8TW&pf_rd_p=05378fd5-c43e-4948-99b1-a65b129fdd73&pd_rd_r=0237fb28-7f47-49f4-986a-be0c78e52863&pd_rd_w=FfIoI&pd_rd_wg=Hw4qq&ref_=pd_gw_unk" }
+
+      it "removes parameters from the URL" do
+        expect(described_class.new(long_url).url)
+          .not_to include("?pf_rd_r")
+      end
     end
 
     describe "#to_html" do
@@ -91,6 +98,10 @@ def check_link(tdl, link)
       it "includes title" do
         expect(html).to include("Seven Languages in Seven Weeks: A Pragmatic Guide to Learning Programming Languages (Pragmatic Programmers)")
       end
+
+      it "includes canonical link" do
+        expect(html).to include(@uri)
+      end
     end
   end