richurls 0.4.0 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 42cff335ffa60073310cdec6c1b16f2a1f1a4cfce0bde60835a76538a0f138aa
-  data.tar.gz: cdded0800498b21f24b5c89910da73fd5150492c4f9c7ccedcb5499d538b1423
+  metadata.gz: f524681ea31f85291124dcf29aeff080bba7e2a6afb94cf376888fd8449d8e9a
+  data.tar.gz: 3d271748e1894aedc12c7579308f9fa6520c26ed01defd313160a8c67abb6ca0
 SHA512:
-  metadata.gz: a5633b6852996d4938f002e2e1e4a90e1b25037ace20ba5da9303abc4dc073efa9ba5819134080d6c467dbe905733226cd66554e99edcd9ab57fbfd49bdc4a57
-  data.tar.gz: '08bce91b2a1c2a301b3eb288883621f7f7d607e56e36765ee00d5d09a84348f281cff0e48f61e074c9c394d5478b9c7140ca9a4f02eb37778e129c3ec8ed6277'
+  metadata.gz: 599e985858d4bafca9865f9880b9e89b0ec4a690da1bda7a07eafa0f685d5b8c0478e838268e6a839690d777588e84ffde9d2d90b1f505c3ebe9c69d246c66f3
+  data.tar.gz: ed33ba734fdf4636c2e996f04eb6e9ab1f2df3a7b7cfebf2ff8b0f564c38186f71fb3842f7b38cf3b10d6ca7c9336c78fdd76940c8b1eb90a0e876e9667f66df

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    richurls (0.4.0)
+    richurls (0.5.0)
       oj (~> 3)
       ox (~> 2)
       patron (~> 0.13)

data/README.md

@@ -9,6 +9,8 @@ gem install richurls
 
 **Usage:**
 
+Default usage:
+
 ```ruby
 require 'richurls'
 
@@ -25,6 +27,19 @@ RichUrls.enrich('https://wetransfer.com')
 # }
 ```
 
+Partial attributes:
+
+```ruby
+require 'richurls'
+
+RichUrls.enrich('https://wetransfer.com', filter: %w[title])
+
+# Returns:
+# {
+#   "title"=>"WeTransfer"
+# }
+```
+
 **Caching:**
 
 By default caching is turned off. Caching can be enabled by writing a cache wrapper as such:
@@ -39,12 +54,14 @@ class CustomCache < RichUrls::Cache::Wrapper
     # Callback for fetching a cache entry
   end
 
-  def set(key, value)
-    # Callback for setting a value in a cache to a certain key
+  def set(key, value, time)
+    # Callback for setting a value in a cache to a certain key for a certain
+    # `time`*.
   end
 
-  def extend(key)
-    # Callback for extending a cached value by key
+  def extend(key, time)
+    # Callback for extending a cached value for a certain key for a certain
+    # `time`*.
   end
 end
 ```
@@ -52,5 +69,19 @@ end
 Finally you can enable the `CustomCache` by adding:
 
 ```ruby
-RichUrls.cache = CustomCache.new(time: 2.hours)
+RichUrls.cache = CustomCache.new(time: 7200)
+```
+
+**\* About custom cache time:**
+
+If you have caching enabled and would like to deviate from the default cache time
+per URL you enrich, it's possible to do so. You'd have to pass a `cache_time`
+parameter to the URL enricher as such:
+
+```ruby
+RichUrls.enrich('https://wetransfer.com', cache_time: 3600)
 ```
+
+This `cache_time` will be accessible through the `time` parameters in the `set`
+and `extend` methods on the `Cache::Wrapper`-instance and can be used as you
+please.

data/lib/body_decorator.rb

@@ -20,15 +20,16 @@ module RichUrls
       'embed' => Parsers::EmbedParser
     }.freeze
 
-    def self.decorate(url, body)
-      new(url, body).decorate
+    def self.decorate(url, body, filter = [])
+      new(url, body, filter).decorate
     end
 
     private_class_method :new
 
-    def initialize(url, body)
+    def initialize(url, body, filter)
       @url = url
-      @xml = XMLHandler.new
+      @filter = filter
+      @xml = XMLHandler.new(filter)
 
       Ox.sax_html(@xml, StringIO.new(body))
 
@@ -40,9 +41,17 @@ module RichUrls
     end
 
     def decorate
-      PARSERS.each_with_object({}) do |(key, parser), object|
+      parsers.each_with_object({}) do |(key, parser), object|
         object[key] = parser.call(@xml.properties[key], @url)
       end
     end
+
+    private
+
+    def parsers
+      return PARSERS if @filter.empty?
+
+      PARSERS.slice(*@filter)
+    end
   end
 end

data/lib/cache.rb

@@ -5,11 +5,11 @@ module RichUrls
         raise NotImplementedError, 'wrapper needs `get` method'
       end
 
-      def set(_key, _value)
+      def set(_key, _value, _time)
         raise NotImplementedError, 'wrapper needs `set` method'
       end
 
-      def extend(_key)
+      def extend(_key, _time)
         raise NotImplementedError, 'wrapper needs `extend` method'
       end
     end
@@ -17,9 +17,9 @@ module RichUrls
     class None < Wrapper
       def get(_); end
 
-      def set(_, _); end
+      def set(_, _, _); end
 
-      def extend(_); end
+      def extend(_, _); end
     end
   end
 end

data/lib/richurls.rb

@@ -22,11 +22,11 @@ module RichUrls
     @cache ||= wrapper
   end
 
-  def self.enrich(url)
+  def self.enrich(url, filter: [], cache_time: nil)
     unless URI::DEFAULT_PARSER.make_regexp.match?(url)
       raise MalformedURLError, "this url is malformed: #{url}"
     end
 
-    UrlFetcher.fetch(url)
+    UrlFetcher.fetch(url, filter, cache_time)
   end
 end

data/lib/url_fetcher.rb

@@ -6,21 +6,23 @@ module RichUrls
 
     class UrlFetcherError < StandardError; end
 
-    def self.fetch(url)
-      new(url).fetch
+    def self.fetch(url, attributes = [], cache_time = nil)
+      new(url, attributes, cache_time).fetch
     end
 
     private_class_method :new
 
-    def initialize(url)
+    def initialize(url, attributes, cache_time)
       @url = url
+      @attributes = attributes
+      @cache_time = cache_time
     end
 
     def fetch
       cached = RichUrls.cache.get(digest)
 
       if cached
-        RichUrls.cache.extend(digest)
+        RichUrls.cache.extend(digest, @cache_time)
         Oj.load(cached)
       else
         patron_call
@@ -30,7 +32,7 @@ module RichUrls
     private
 
     def digest
-      @digest ||= Digest::MD5.hexdigest(@url)
+      @digest ||= Digest::MD5.hexdigest(@url + @attributes.sort.join('-'))
     end
 
     def patron_call
@@ -38,8 +40,10 @@ module RichUrls
       response = session.get(@url)
 
       if response.status < 400
-        decorated = BodyDecorator.decorate(response.url, response.body)
-        RichUrls.cache.set(digest, Oj.dump(decorated))
+        decorated = BodyDecorator.decorate(
+          response.url, response.body, @attributes
+        )
+        RichUrls.cache.set(digest, Oj.dump(decorated), @cache_time)
         decorated
       else
         raise UrlFetcherError, 'url cannot be found'

data/lib/xml_handler.rb

@@ -43,15 +43,11 @@ module RichUrls
 
     attr_reader :elements, :properties
 
-    def initialize
+    def initialize(filter = [])
+      @filter = filter
       @elements = []
       @counts = Set.new
-      @properties = {
-        'title' => nil,
-        'description' => nil,
-        'image' => nil,
-        'favicon' => nil
-      }
+      @properties = filtered_properties(filter)
     end
 
     def find(tag, attrs = {})
@@ -115,5 +111,14 @@ module RichUrls
 
       !find(:meta, property: FALLBACK_ELEMENTS.fetch(tag))
     end
+
+    # Turns a set of filtered properties into a Hash where
+    # the default value is `nil`
+    def filtered_properties(filter)
+      keys = FINDERS.values.uniq
+      keys &= filter if filter.any?
+
+      Hash[keys.zip([])]
+    end
   end
 end

data/richurls.gemspec

@@ -3,7 +3,7 @@ $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
 
 Gem::Specification.new do |spec|
   spec.name = 'richurls'
-  spec.version = '0.4.0'
+  spec.version = '0.5.0'
   spec.authors = ['grdw']
   spec.email = ['gerard@wetransfer.com']
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: richurls
 version: !ruby/object:Gem::Version
-  version: 0.4.0
+  version: 0.5.0
 platform: ruby
 authors:
 - grdw
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-03-23 00:00:00.000000000 Z
+date: 2020-03-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: oj