sources: don't escape Unicode characters in tag search URLs.

Fix it so that Unicode characters aren't unnecessarily percent-encoded when generating tag search
URLs. For example, generate URLs like this:

* https://www.pixiv.net/tags/オリジナル/artworks

Not like this:

* https://www.pixiv.net/tags/%E3%82%AA%E3%83%AA%E3%82%B8%E3%83%8A%E3%83%AB/artworks

app/logical/danbooru/url.rb

@@ -68,6 +68,17 @@ module Danbooru
       nil
     end
 
+    # Escape a string for use in an URL path or query parameter. Like `CGI.escape`, but leaves Unicode characters as Unicode.
+    #
+    # @example
+    #   Danbooru::URL.escape("fate/stay_night") # => "fate%2Fstay_night"
+    #   Danbooru::URL.escape("大丈夫?おっぱい揉む?") # => "大丈夫%3Fおっぱい揉む%3F"
+    #
+    # @return [String] The escaped string
+    def self.escape(string)
+      Addressable::URI.encode_component(string, /[\/?#&+%]/).force_encoding("UTF-8")
+    end
+
     # @return [String] the URL in unnormalized form
     def to_s
       original_url

app/logical/source/extractor/art_station.rb

@@ -48,7 +48,7 @@ class Source::Extractor
 
     def tags
       api_response[:tags].to_a.map do |tag|
-        [tag, "https://www.artstation.com/search?q=#{CGI.escape(tag)}"]
+        [tag, "https://www.artstation.com/search?q=#{Danbooru::URL.escape(tag)}"]
       end
     end
 

app/logical/source/extractor/gelbooru.rb

@@ -39,7 +39,7 @@ module Source
 
         tags = api_response[:tags].split + ["rating:#{api_response[:rating]}"]
         tags.map do |tag|
-          [tag, "https://#{domain}/index.php?page=post&s=list&tags=#{CGI.escape(tag)}"]
+          [tag, "https://#{domain}/index.php?page=post&s=list&tags=#{Danbooru::URL.escape(tag)}"]
         end
       end
 

app/logical/source/extractor/hentai_foundry.rb

@@ -39,7 +39,7 @@ module Source
         tags = page&.search(".boxbody [rel='tag']").to_a.map(&:text)
 
         tags.map do |tag|
-          [tag, "https://www.hentai-foundry.com/pictures/tagged/#{CGI.escape(tag)}"]
+          [tag, "https://www.hentai-foundry.com/pictures/tagged/#{Danbooru::URL.escape(tag)}"]
         end
       end
 

app/logical/source/extractor/moebooru.rb

@@ -23,7 +23,7 @@ module Source
 
       def tags
         api_response[:tags].to_s.split.map do |tag|
-          [tag, "https://#{domain}/post?tags=#{CGI.escape(tag)}"]
+          [tag, "https://#{domain}/post?tags=#{Danbooru::URL.escape(tag)}"]
         end
       end
 

app/logical/source/extractor/nico_seiga.rb

@@ -77,12 +77,8 @@ module Source
       def tags
         return [] if api_client.blank?
 
-        base_url = "https://seiga.nicovideo.jp/"
-        base_url += "manga/" if manga_id.present?
-        base_url += "tag/"
-
         api_client.tags.map do |name|
-          [name, base_url + CGI.escape(name)]
+          [name, "https://seiga.nicovideo.jp/#{"manga/" if manga_id}tag/#{Danbooru::URL.escape(name)}"]
         end
       end
 

app/logical/source/extractor/pixiv.rb

@@ -105,7 +105,7 @@ module Source
 
       def tags
         tags = api_illust.dig(:tags, :tags).to_a.map do |item|
-          [item[:tag], "https://www.pixiv.net/tags/#{CGI.escape(item[:tag])}/artworks"]
+          [item[:tag], "https://www.pixiv.net/tags/#{Danbooru::URL.escape(item[:tag])}/artworks"]
         end
 
         if api_illust["aiType"] == 2

app/logical/source/extractor/rule34_dot_us.rb

@@ -28,7 +28,7 @@ module Source
 
       def tags
         page&.css("meta[name='keywords']")&.attr("content")&.value.to_s.split(/, /).compact.map do |tag|
-          [tag.tr(" ", "_"), "https://rule34.us/index.php?r=posts/index&q=#{CGI.escape(tag)}"]
+          [tag.tr(" ", "_"), "https://rule34.us/index.php?r=posts/index&q=#{Danbooru::URL.escape(tag)}"]
         end
       end
 

app/logical/source/extractor/tinami.rb

@@ -47,7 +47,7 @@ module Source
 
       def tags
         page&.css("#view .tag a[href^='/search/list']").to_a.map do |tag|
-          [tag.text, "https://www.tinami.com/search/list?keyword=#{CGI.escape(tag.text)}"]
+          [tag.text, "https://www.tinami.com/search/list?keyword=#{Danbooru::URL.escape(tag.text)}"]
         end
       end
 

app/logical/source/extractor/tumblr.rb

@@ -75,7 +75,7 @@ class Source::Extractor
 
     def tags
       post[:tags].to_a.map do |tag|
-        [tag, "https://tumblr.com/tagged/#{CGI.escape(tag)}"]
+        [tag, "https://tumblr.com/tagged/#{Danbooru::URL.escape(tag)}"]
       end.uniq
     end
 

test/unit/sources/nico_seiga_test.rb

@@ -8,12 +8,12 @@ module Sources
 
     context "A nicoseiga post url" do
       tags = [
-        ["アニメ", "https://seiga.nicovideo.jp/tag/%E3%82%A2%E3%83%8B%E3%83%A1"],
-        ["コジコジ", "https://seiga.nicovideo.jp/tag/%E3%82%B3%E3%82%B8%E3%82%B3%E3%82%B8"],
-        ["さくらももこ", "https://seiga.nicovideo.jp/tag/%E3%81%95%E3%81%8F%E3%82%89%E3%82%82%E3%82%82%E3%81%93"],
-        ["ドット絵", "https://seiga.nicovideo.jp/tag/%E3%83%89%E3%83%83%E3%83%88%E7%B5%B5"],
-        ["ニコニコ大百科", "https://seiga.nicovideo.jp/tag/%E3%83%8B%E3%82%B3%E3%83%8B%E3%82%B3%E5%A4%A7%E7%99%BE%E7%A7%91"],
-        ["お絵カキコ", "https://seiga.nicovideo.jp/tag/%E3%81%8A%E7%B5%B5%E3%82%AB%E3%82%AD%E3%82%B3"],
+        ["アニメ", "https://seiga.nicovideo.jp/tag/アニメ"],
+        ["コジコジ", "https://seiga.nicovideo.jp/tag/コジコジ"],
+        ["さくらももこ", "https://seiga.nicovideo.jp/tag/さくらももこ"],
+        ["ドット絵", "https://seiga.nicovideo.jp/tag/ドット絵"],
+        ["ニコニコ大百科", "https://seiga.nicovideo.jp/tag/ニコニコ大百科"],
+        ["お絵カキコ", "https://seiga.nicovideo.jp/tag/お絵カキコ"],
       ]
       strategy_should_work(
         "http://seiga.nicovideo.jp/seiga/im4937663",
@@ -32,12 +32,12 @@ module Sources
 
     context "A nicoseiga image url" do
       tags = [
-        ["アニメ", "https://seiga.nicovideo.jp/tag/%E3%82%A2%E3%83%8B%E3%83%A1"],
-        ["コジコジ", "https://seiga.nicovideo.jp/tag/%E3%82%B3%E3%82%B8%E3%82%B3%E3%82%B8"],
-        ["さくらももこ", "https://seiga.nicovideo.jp/tag/%E3%81%95%E3%81%8F%E3%82%89%E3%82%82%E3%82%82%E3%81%93"],
-        ["ドット絵", "https://seiga.nicovideo.jp/tag/%E3%83%89%E3%83%83%E3%83%88%E7%B5%B5"],
-        ["ニコニコ大百科", "https://seiga.nicovideo.jp/tag/%E3%83%8B%E3%82%B3%E3%83%8B%E3%82%B3%E5%A4%A7%E7%99%BE%E7%A7%91"],
-        ["お絵カキコ", "https://seiga.nicovideo.jp/tag/%E3%81%8A%E7%B5%B5%E3%82%AB%E3%82%AD%E3%82%B3"],
+        ["アニメ", "https://seiga.nicovideo.jp/tag/アニメ"],
+        ["コジコジ", "https://seiga.nicovideo.jp/tag/コジコジ"],
+        ["さくらももこ", "https://seiga.nicovideo.jp/tag/さくらももこ"],
+        ["ドット絵", "https://seiga.nicovideo.jp/tag/ドット絵"],
+        ["ニコニコ大百科", "https://seiga.nicovideo.jp/tag/ニコニコ大百科"],
+        ["お絵カキコ", "https://seiga.nicovideo.jp/tag/お絵カキコ"],
       ]
       strategy_should_work(
         "http://lohas.nicoseiga.jp/o/910aecf08e542285862954017f8a33a8c32a8aec/1433298801/4937663",
@@ -186,7 +186,7 @@ module Sources
         "https://seiga.nicovideo.jp/watch/mg302561",
         image_urls: image_urls,
         page_url: "https://seiga.nicovideo.jp/watch/mg302561",
-        tags: [["ロリ", "https://seiga.nicovideo.jp/manga/tag/%E3%83%AD%E3%83%AA"]],
+        tags: [["ロリ", "https://seiga.nicovideo.jp/manga/tag/ロリ"]],
         artist_name: "とろてい",
         other_names: ["とろてい"],
         tag_name: "nicoseiga_1848060"

test/unit/sources/pixiv_test.rb

@@ -94,10 +94,8 @@ module Sources
 
         should "get the tags" do
           pixiv_tags  = @site.tags.map(&:first)
-          pixiv_links = @site.tags.map(&:last)
 
           assert_equal(%w[漫画 test], pixiv_tags)
-          assert_contains(pixiv_links, /search\.php/)
         end
 
         should "get the artist commentary" do