onebox 2.2.9 → 2.2.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 541807b758c73492e2868271a63b8f05ef4d94aad316be86fb5c42c3c3f0e2ce
-  data.tar.gz: 10f460d3d6a1b09a6087a38ad0692c9f652e027beb65b22845536480055b4a48
+  metadata.gz: f942eef9390a961ec26aa8d4301bf83eedabe577d067639ab6e0036ca20e02b9
+  data.tar.gz: b05c99bd27f025edc02475e38cc28b900bfa2bf64e1c5f9cd4be74cd5ab927a4
 SHA512:
-  metadata.gz: d13068e9dbb437388ee689581bc64bb08bf4c5ae898755680c3c978a00fc4a4ee13004f0f2a3071f40d226d15e84f799a949d0d07572c6537ddd92950416d56b
-  data.tar.gz: 1efacb58ff5d24daff63cdd855a32ae80826f3585eee0305ba032269ee4431f1541848582bd5587aacffdcd6cb9c5bb35c650706c6028ed5e115e17daf362621
+  metadata.gz: '081ccc4c533884804d094d40b71175409077002595b10b9c3fd56d8f6c11bcacdc99502fdd78f082adb1e59d83090182550436579ec8bc2361714bb659f7a693'
+  data.tar.gz: d3dfcb8f455e7d9189415cfdf9e00f7f1d938b8ce41d8d307ff2ff4bbb6fcf09449a4a881b0bfec34f8b275556691601f2e6d070351ea2b51d5626e5a561b2fd

data/lib/onebox/engine/amazon_onebox.rb

@@ -14,17 +14,20 @@ module Onebox
       matches_regexp(/^https?:\/\/(?:www\.)?(?:smile\.)?(amazon|amzn)\.(?<tld>com|ca|de|it|es|fr|co\.jp|co\.uk|cn|in|com\.br|com\.mx|nl|pl|sa|sg|se|com\.tr|ae)\//)
 
       def url
-        # Have we cached the HTML body of the requested URL?
-        # If so, try to grab the canonical URL from that document,
+        # If possible, fetch the cached HTML body immediately so we can
+        # try to grab the canonical URL from that document,
         # rather than guess at the best URL structure to use
-        if @body_cacher && @body_cacher.respond_to?('cache_response_body?')
-          if @body_cacher.cached_response_body_exists?(uri.to_s)
-            @raw ||= Onebox::Helpers.fetch_html_doc(@url, http_params, @body_cacher)
-            canonical_link = @raw.at('//link[@rel="canonical"]/@href')
-            return canonical_link.to_s if canonical_link
+        if body_cacher&.respond_to?('cache_response_body?')
+          if body_cacher.cache_response_body?(uri.to_s) && body_cacher.cached_response_body_exists?(uri.to_s)
+            @raw ||= Onebox::Helpers.fetch_html_doc(@url, http_params, body_cacher)
           end
         end
 
+        if @raw
+          canonical_link = @raw.at('//link[@rel="canonical"]/@href')
+          return canonical_link.to_s if canonical_link
+        end
+
         if match && match[:id]
           return "https://www.amazon.#{tld}/dp/#{Onebox::Helpers.uri_encode(match[:id])}"
         end
@@ -45,7 +48,7 @@ module Onebox
       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 image
@@ -60,6 +63,10 @@ module Onebox
         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 +117,7 @@ module Onebox
           end
 
           result = {
-            link: link,
+            link: url,
             title: title,
             by_info: authors,
             image: og.image || image,
@@ -141,7 +148,7 @@ module Onebox
           end
 
           result = {
-            link: link,
+            link: url,
             title: title,
             by_info: authors,
             image: og.image || image,
@@ -157,7 +164,7 @@ module Onebox
         else
           title = og.title || CGI.unescapeHTML(raw.css("title").inner_text)
           result = {
-            link: link,
+            link: url,
             title: title,
             image: og.image || image,
             price: price
@@ -167,7 +174,10 @@ module Onebox
           result[:by_info] = Onebox::Helpers.clean(result[:by_info].inner_html) if result[:by_info]
 
           summary = raw.at("#productDescription")
-          result[:description] = og.description || (summary && summary.inner_text) || CGI.unescapeHTML(Onebox::Helpers.truncate(raw.css("meta[name=description]").first["content"], 250))
+
+          description = og.description || summary&.inner_text
+          description ||= raw.css("meta[name=description]").first&.[]("content")
+          result[:description] = CGI.unescapeHTML(Onebox::Helpers.truncate(description, 250)) if description
         end
 
         result[:price] = nil if result[:price].start_with?("$0") || result[:price] == 0

data/lib/onebox/engine/html.rb

@@ -11,10 +11,13 @@ module Onebox
       end
 
       def raw
-        body_cacher = self.options[:body_cacher] if self.options
         @raw ||= Onebox::Helpers.fetch_html_doc(url, http_params, body_cacher)
       end
 
+      def body_cacher
+        self.options&.[](:body_cacher)
+      end
+
       def html?
         raw.respond_to(:css)
       end

data/lib/onebox/version.rb

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

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: onebox
 version: !ruby/object:Gem::Version
-  version: 2.2.9
+  version: 2.2.10
 platform: ruby
 authors:
 - Joanna Zeta
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-03-31 00:00:00.000000000 Z
+date: 2021-04-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: addressable