Twitter: rewrite tests

Also add partial support for profile banner images.

app/logical/source/extractor/twitter.rb

@@ -58,13 +58,16 @@ class Source::Extractor
     end
 
     def intent_url
-      user_id = api_response.dig(:user, :id_str)
       return nil if user_id.blank?
       "https://twitter.com/intent/user?user_id=#{user_id}"
     end
 
     def profile_urls
-      [profile_url, intent_url].compact
+      [profile_url, intent_url].compact.uniq
+    end
+
+    def user_id
+      parsed_url.user_id || parsed_referer&.user_id || api_response.dig(:user, :id_str)
     end
 
     def tag_name
@@ -72,8 +75,6 @@ class Source::Extractor
         tag_name_from_url
       elsif api_response.present?
         api_response.dig(:user, :screen_name)
-      else
-        ""
       end
     end
 

app/logical/source/url/twitter.rb

@@ -11,8 +11,6 @@
 # * https://video.twimg.com/ext_tw_video/1496554514312269828/pu/vid/480x360/amWjOw0MmLdnPMPB.mp4?tag=12
 #
 # Profile image URLs:
-#
-# * https://pbs.twimg.com/profile_banners/780804311529906176/1475001696
 # * https://pbs.twimg.com/profile_images/1493345400929112064/lF1mY1i2_normal.jpg
 #
 # Shortened URLs:
@@ -71,6 +69,8 @@ class Source::URL::Twitter < Source::URL
     # https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg
     in "twimg.com", ("media" | "tweet_video_thumb" | "ext_tw_video_thumb" | "amplify_video_thumb") => media_type, *subdirs, file
       # EBGbJe_U8AA4Ekb.jpg:small
+      @media_type = media_type
+
       @file, @file_size = file.split(":")
       @file, @file_ext = @file.split(".")
 
@@ -80,7 +80,14 @@ class Source::URL::Twitter < Source::URL
 
       # /media/EBGbJe_U8AA4Ekb.jpg
       # /ext_tw_video_thumb/1243725361986375680/pu/img/JDA7g7lcw7wK-PIv.jpg
-      @file_path = File.join(media_type, subdirs.join("/"), "#{@file}.#{@file_ext}")
+      @file_path = File.join(@media_type, subdirs.join("/"), "#{@file}.#{@file_ext}")
+
+    # https://pbs.twimg.com/profile_banners/780804311529906176/1475001696
+    # https://pbs.twimg.com/profile_banners/780804311529906176/1475001696/600x200
+    in "twimg.com", "profile_banners" => media_type, /^\d+$/ => user_id, /^\d+$/ => file_id, *dimensions
+      @user_id = user_id
+      @media_type = media_type
+      @file_path = "profile_banners/#{user_id}/#{file_id}/1500x500"
 
     # https://twitter.com/motty08111213
     # https://twitter.com/motty08111213/likes
@@ -102,7 +109,11 @@ class Source::URL::Twitter < Source::URL
   # https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg:orig
   def full_image_url
     return to_s unless @file_path.present?
-    "#{site}/#{@file_path}:orig"
+    if @media_type == "profile_banners"
+      "#{site}/#{@file_path}"
+    else
+      "#{site}/#{@file_path}:orig"
+    end
   end
 
   def page_url

test/test_helpers/source_test_helper.rb

@@ -35,7 +35,7 @@ module SourceTestHelper
     should_match_source_data(strategy, arguments)
   end
 
-  def should_download_successfully(strategy, download_size = nil)
+  def should_download_successfully(strategy, download_size = nil) # XXX perhaps refactor into should_upload_successfully in order to streamline the source tests further?
     should "download successfully" do
       file = strategy.download_file!(strategy.image_urls.first)
       if download_size.present?
@@ -56,8 +56,8 @@ module SourceTestHelper
       end
     else
       should "not incorrectly extract a profile url or artist data when there's none to be found" do
-        assert_nil(strategy.profile_url)
+        assert_nil(strategy.profile_url.presence)
-        assert_nil(strategy.artist_name)
+        assert_nil(strategy.artist_name.presence)
         assert_equal([], strategy.other_names)
       end
     end

test/unit/sources/twitter_test.rb

@@ -6,181 +6,6 @@ module Sources
       skip "Twitter credentials are not configured" if !Source::Extractor::Twitter.enabled?
     end
 
-    context "An extended tweet" do
-      should "extract the correct image url" do
-        @site = Source::Extractor.find("https://twitter.com/onsen_musume_jp/status/865534101918330881")
-        assert_equal(["https://pbs.twimg.com/media/DAL-ntWV0AEbhes.jpg:orig"], @site.image_urls)
-      end
-
-      should "extract all the image urls" do
-        @site = Source::Extractor.find("https://twitter.com/baalbuddy/status/1455330043828264963")
-
-        urls = %w[
-          https://pbs.twimg.com/media/FDJekEfX0AQZ-Mx.png:orig
-          https://pbs.twimg.com/media/FDJekEkWYAICHzF.png:orig
-          https://pbs.twimg.com/media/FDJekEiWEAEsPky.png:orig
-          https://pbs.twimg.com/media/FDJekEkWEAo4sMQ.png:orig
-        ]
-
-        assert_equal(urls, @site.image_urls)
-      end
-    end
-
-    context "A video" do
-      should "get the correct urls" do
-        @site = Source::Extractor.find("https://twitter.com/CincinnatiZoo/status/859073537713328129")
-        assert_equal(["https://video.twimg.com/ext_tw_video/859073467769126913/pu/vid/1280x720/cPGgVROXHy3yrK6u.mp4"], @site.image_urls)
-        assert_equal("https://twitter.com/CincinnatiZoo/status/859073537713328129", @site.page_url)
-      end
-
-      should "work when given a video thumbnail" do
-        # https://twitter.com/Kekeflipnote/status/1241038667898118144
-        @site = Source::Extractor.find("https://pbs.twimg.com/tweet_video_thumb/ETkN_L3X0AMy1aT.jpg:small")
-        assert_equal(["https://pbs.twimg.com/tweet_video_thumb/ETkN_L3X0AMy1aT.jpg:orig"], @site.image_urls)
-      end
-
-      should "work when given an external video thumbnail" do
-        # https://twitter.com/chivedips/status/1243850897056133121
-        @site = Source::Extractor.find("https://pbs.twimg.com/ext_tw_video_thumb/1243725361986375680/pu/img/JDA7g7lcw7wK-PIv.jpg:small")
-        assert_equal(["https://pbs.twimg.com/ext_tw_video_thumb/1243725361986375680/pu/img/JDA7g7lcw7wK-PIv.jpg:orig"], @site.image_urls)
-      end
-
-      should "work when given an amplify video thumbnail" do
-        # https://twitter.com/UNITED_CINEMAS/status/1223138847417978881
-        @site = Source::Extractor.find("https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg:small")
-        assert_equal(["https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg:orig"], @site.image_urls)
-      end
-    end
-
-    context "An animated gif" do
-      setup do
-        @site = Source::Extractor.find("https://twitter.com/i/web/status/1252517866059907073")
-      end
-
-      should "get the image url" do
-        assert_equal(["https://video.twimg.com/tweet_video/EWHWVrmVcAAp4Vw.mp4"], @site.image_urls)
-      end
-    end
-
-    context "A twitter summary card from twitter with a :large image" do
-      setup do
-        @site = Source::Extractor.find("https://twitter.com/aranobu/status/817736083567820800")
-      end
-
-      should "get the image url" do
-        assert_equal(["https://pbs.twimg.com/media/C1kt72yVEAEGpOv.jpg:orig"], @site.image_urls)
-      end
-
-      should "get the page url" do
-        assert_equal("https://twitter.com/aranobu/status/817736083567820800", @site.page_url)
-      end
-    end
-
-    context "The source site for a restricted twitter" do
-      setup do
-        @site = Source::Extractor.find("https://mobile.twitter.com/Strangestone/status/556440271961858051")
-      end
-
-      should "get the urls" do
-        assert_equal(["https://pbs.twimg.com/media/B7jfc1JCcAEyeJh.png:orig"], @site.image_urls)
-        assert_equal("https://twitter.com/Strangestone/status/556440271961858051", @site.page_url)
-      end
-    end
-
-    context "A tweet without any images" do
-      should "not fail" do
-        @site = Source::Extractor.find("https://twitter.com/teruyo/status/1058452066060853248")
-
-        assert_equal([], @site.image_urls)
-        assert_nothing_raised { @site.to_h }
-      end
-    end
-
-    context "The source site for twitter" do
-      setup do
-        @site = Source::Extractor.find("https://mobile.twitter.com/nounproject/status/540944400767922176")
-      end
-
-      should "get the main profile url" do
-        assert_equal("https://twitter.com/nounproject", @site.profile_url)
-      end
-
-      should "get the profile urls" do
-        assert_includes(@site.profile_urls, "https://twitter.com/nounproject")
-        assert_includes(@site.profile_urls, "https://twitter.com/intent/user?user_id=88996186")
-      end
-
-      should "get the artist name" do
-        assert_equal("Noun Project", @site.artist_name)
-      end
-
-      should "get the image urls" do
-        assert_equal(["https://pbs.twimg.com/media/B4HSEP5CUAA4xyu.png:orig"], @site.image_urls)
-      end
-
-      should "get the page url" do
-        assert_equal("https://twitter.com/nounproject/status/540944400767922176", @site.page_url)
-      end
-
-      should "get the tags" do
-        assert_equal([], @site.tags)
-      end
-
-      should "get the artist commentary" do
-        assert_not_nil(@site.artist_commentary_desc)
-      end
-
-      should "convert a page into a json representation" do
-        assert_nothing_raised do
-          @site.to_json
-        end
-      end
-    end
-
-    context "The source site for a direct image and a referer" do
-      setup do
-        @site = Source::Extractor.find("https://pbs.twimg.com/media/B4HSEP5CUAA4xyu.png:large", "https://twitter.com/nounproject/status/540944400767922176")
-      end
-
-      should "get the source data" do
-        assert_equal("nounproject", @site.tag_name)
-        assert_equal("Noun Project", @site.artist_name)
-        assert_equal(["https://pbs.twimg.com/media/B4HSEP5CUAA4xyu.png:orig"], @site.image_urls)
-      end
-    end
-
-    context "The source site for a direct image url (pbs.twimg.com/media/*.jpg) without a referer url" do
-      setup do
-        @site = Source::Extractor.find("https://pbs.twimg.com/media/B4HSEP5CUAA4xyu.png:large")
-      end
-
-      should "work" do
-        assert_equal(["https://pbs.twimg.com/media/B4HSEP5CUAA4xyu.png:orig"], @site.image_urls)
-        assert(@site.artist_name.blank?)
-        assert(@site.profile_url.blank?)
-        assert(@site.artists.empty?)
-        assert(@site.tags.empty?)
-        assert(@site.artist_commentary_desc.blank?)
-        assert(@site.dtext_artist_commentary_desc.blank?)
-        assert_nothing_raised { @site.to_h }
-      end
-    end
-
-    context "The source site for a direct image url (pbs.twimg.com/media/*?format=jpg&name=*) without a referer url" do
-      setup do
-        @site = Source::Extractor.find("https://pbs.twimg.com/media/EBGp2YdUYAA19Uj?format=jpg&name=small")
-      end
-
-      should "work" do
-        assert_equal(["https://pbs.twimg.com/media/EBGp2YdUYAA19Uj.jpg:orig"], @site.image_urls)
-      end
-
-      should "work for filenames containing dashes" do
-        @site = Source::Extractor.find("https://pbs.twimg.com/media/EAjc-OWVAAAxAgQ.jpg", "https://twitter.com/asteroid_ill/status/1155420330128625664")
-        assert_equal(["https://pbs.twimg.com/media/EAjc-OWVAAAxAgQ.jpg:orig"], @site.image_urls)
-      end
-    end
-
     context "A https://twitter.com/:username/status/:id url" do
       strategy_should_work(
         "https://twitter.com/motty08111213/status/943446161586733056",
@@ -188,12 +13,13 @@ module Sources
         image_urls: [
           "https://pbs.twimg.com/media/DRfKHmgV4AAycFB.jpg:orig",
           "https://pbs.twimg.com/media/DRfKHioVoAALRlK.jpg:orig",
-          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig"
+          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig",
         ],
+        download_size: 275_713,
         profile_url: "https://twitter.com/motty08111213",
         artist_name: "えのぐマネージャー 丸茂",
         tag_name: "motty08111213",
-        tags: ["岩本町芸能社", "女優部"],
+        tags: ["岩本町芸能社", "女優部"]
       )
     end
 
@@ -204,12 +30,13 @@ module Sources
         image_urls: [
           "https://pbs.twimg.com/media/DRfKHmgV4AAycFB.jpg:orig",
           "https://pbs.twimg.com/media/DRfKHioVoAALRlK.jpg:orig",
-          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig"
+          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig",
         ],
+        download_size: 275_713,
         profile_url: "https://twitter.com/motty08111213",
         artist_name: "えのぐマネージャー 丸茂",
         tag_name: "motty08111213",
-        tags: ["岩本町芸能社", "女優部"],
+        tags: ["岩本町芸能社", "女優部"]
       )
     end
 
@@ -220,107 +47,165 @@ module Sources
         image_urls: [
           "https://pbs.twimg.com/media/DRfKHmgV4AAycFB.jpg:orig",
           "https://pbs.twimg.com/media/DRfKHioVoAALRlK.jpg:orig",
-          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig"
+          "https://pbs.twimg.com/media/DRfKHgHU8AE7alV.jpg:orig",
         ],
+        download_size: 275_713,
         profile_url: "https://twitter.com/motty08111213",
         artist_name: "えのぐマネージャー 丸茂",
         tag_name: "motty08111213",
-        tags: ["岩本町芸能社", "女優部"],
+        tags: ["岩本町芸能社", "女優部"]
+      )
+    end
+
+    context "A video tweet" do
+      strategy_should_work(
+        "https://twitter.com/CincinnatiZoo/status/859073537713328129",
+        image_urls: ["https://video.twimg.com/ext_tw_video/859073467769126913/pu/vid/1280x720/cPGgVROXHy3yrK6u.mp4"],
+        page_url: "https://twitter.com/CincinnatiZoo/status/859073537713328129",
+        download_size: 8_602_983
+      )
+    end
+
+    context "A video thumbnail" do
+      # https://twitter.com/Kekeflipnote/status/1241038667898118144
+      strategy_should_work(
+        "https://pbs.twimg.com/tweet_video_thumb/ETkN_L3X0AMy1aT.jpg:small",
+        image_urls: ["https://pbs.twimg.com/tweet_video_thumb/ETkN_L3X0AMy1aT.jpg:orig"],
+        download_size: 18_058
+      )
+    end
+
+    context "An external video thumbnail" do
+      strategy_should_work(
+        "https://pbs.twimg.com/ext_tw_video_thumb/1578376127801761793/pu/img/oGcUqPnwRYYhk-gi.jpg:small",
+        image_urls: ["https://pbs.twimg.com/ext_tw_video_thumb/1578376127801761793/pu/img/oGcUqPnwRYYhk-gi.jpg:orig"],
+        download_size: 243_227
+      )
+    end
+
+    context "An amplify video thumbnail" do
+      # https://twitter.com/UNITED_CINEMAS/status/1223138847417978881
+      strategy_should_work(
+        "https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg:small",
+        image_urls: ["https://pbs.twimg.com/amplify_video_thumb/1215590775364259840/img/lolCkEEioFZTb5dl.jpg:orig"],
+        download_size: 106_942
+      )
+    end
+
+    context "A tweet with an animated gif" do
+      strategy_should_work(
+        "https://twitter.com/i/web/status/1252517866059907073",
+        image_urls: ["https://video.twimg.com/tweet_video/EWHWVrmVcAAp4Vw.mp4"],
+        download_size: 542_833
+      )
+    end
+
+    context "A restricted tweet" do
+      strategy_should_work(
+        "https://mobile.twitter.com/Strangestone/status/556440271961858051",
+        image_urls: ["https://pbs.twimg.com/media/B7jfc1JCcAEyeJh.png:orig"],
+        page_url: "https://twitter.com/Strangestone/status/556440271961858051",
+        profile_url: "https://twitter.com/Strangestone",
+        profile_urls: ["https://twitter.com/Strangestone", "https://twitter.com/intent/user?user_id=93332575"],
+        tag_name: "Strangestone",
+        artist_name: "比村奇石",
+        dtext_artist_commentary_desc: "ブレザーが描きたかったのでJK鈴谷"
+      )
+    end
+
+    context "A tweet without any images" do
+      strategy_should_work(
+        "https://twitter.com/teruyo/status/1058452066060853248",
+        profile_url: "https://twitter.com/teruyo",
+        image_urls: []
+      )
+    end
+
+    context "A direct image url" do
+      strategy_should_work(
+        "https://pbs.twimg.com/media/EBGp2YdUYAA19Uj?format=jpg&name=small",
+        image_urls: ["https://pbs.twimg.com/media/EBGp2YdUYAA19Uj.jpg:orig"],
+        download_size: 229_661,
+        profile_url: nil
+      )
+    end
+
+    context "A direct image url with dashes" do
+      strategy_should_work(
+        "https://pbs.twimg.com/media/EAjc-OWVAAAxAgQ.jpg",
+        image_urls: ["https://pbs.twimg.com/media/EAjc-OWVAAAxAgQ.jpg:orig"],
+        download_size: 842_373,
+        profile_url: nil
       )
     end
 
     context "A deleted tweet" do
-      should "still find the artist name" do
+      strategy_should_work(
-        @site = Source::Extractor.find("https://twitter.com/masayasuf/status/870734961778630656")
+        "https://twitter.com/masayasuf/status/870734961778630656",
-        @artist = FactoryBot.create(:artist, name: "masayasuf", url_string: @site.url)
+        deleted: true,
-
+        tag_name: "masayasuf",
-        assert_equal("masayasuf", @site.tag_name)
+        profile_url: "https://twitter.com/masayasuf"
-        assert_equal("https://twitter.com/masayasuf", @site.profile_url)
+      )
-        assert_equal([@artist], @site.artists)
-      end
-    end
-
-    context "A tweet" do
-      setup do
-        @site = Source::Extractor.find("https://twitter.com/noizave/status/875768175136317440")
-      end
-
-      should "convert urls, hashtags, and mentions to dtext" do
-        desc = 'test "#foo":[https://twitter.com/hashtag/foo] "#ホワイトデー":[https://twitter.com/hashtag/ホワイトデー] "@noizave":[https://twitter.com/noizave]\'s blah http://www.example.com <>& 😀'
-        assert_equal(desc, @site.dtext_artist_commentary_desc)
-      end
-
-      should "get the tags" do
-        tags = [
-          %w[foo https://twitter.com/hashtag/foo],
-          %w[ホワイトデー https://twitter.com/hashtag/ホワイトデー]
-        ]
-
-        assert_equal(tags, @site.tags)
-      end
-    end
-
-    context "A profile banner image" do
-      should "work" do
-        @site = Source::Extractor.find("https://pbs.twimg.com/profile_banners/1225702850002468864/1588597370/1500x500")
-        assert_equal([@site.url], @site.image_urls)
-        assert_nothing_raised { @site.to_h }
-      end
-    end
-
-    context "A tweet containing non-normalized Unicode text" do
-      should "be normalized to nfkc" do
-        site = Source::Extractor.find("https://twitter.com/aprilarcus/status/367557195186970624")
-        desc1 = "𝖸𝗈 𝐔𝐧𝐢𝐜𝐨𝐝𝐞 𝗅 𝗁𝖾𝗋𝖽 𝕌 𝗅𝗂𝗄𝖾 𝑡𝑦𝑝𝑒𝑓𝑎𝑐𝑒𝑠 𝗌𝗈 𝗐𝖾 𝗉𝗎𝗍 𝗌𝗈𝗆𝖾 𝚌𝚘𝚍𝚎𝚙𝚘𝚒𝚗𝚝𝚜 𝗂𝗇 𝗒𝗈𝗎𝗋 𝔖𝔲𝔭𝔭𝔩𝔢𝔪𝔢𝔫𝔱𝔞𝔯𝔶 𝔚𝔲𝔩𝔱𝔦𝔩𝔦𝔫𝔤𝔳𝔞𝔩 𝔓𝔩𝔞𝔫𝔢 𝗌𝗈 𝗒𝗈𝗎 𝖼𝖺𝗇 𝓮𝓷𝓬𝓸𝓭𝓮 𝕗𝕠𝕟𝕥𝕤 𝗂𝗇 𝗒𝗈𝗎𝗋 𝒇𝒐𝒏𝒕𝒔."
-        desc2 = "Yo Unicode l herd U like typefaces so we put some codepoints in your Supplementary Wultilingval Plane so you can encode fonts in your fonts."
-
-        assert_equal(desc1, site.artist_commentary_desc)
-        assert_equal(desc2, site.dtext_artist_commentary_desc)
-      end
-
-      should "normalize full-width hashtags" do
-        site = Source::Extractor.find("https://twitter.com/corpsmanWelt/status/1037724260075069441")
-        desc1 = %{新しいおともだち\n＃けものフレンズ https://t.co/sEAuu16yAQ}
-        desc2 = %{新しいおともだち\n"#けものフレンズ":[https://twitter.com/hashtag/けものフレンズ]}
-
-        assert_equal(desc1, site.artist_commentary_desc)
-        assert_equal(desc2, site.dtext_artist_commentary_desc)
-      end
-    end
-
-    context "A twitter post with a pixiv referer" do
-      should "use the twitter strategy" do
-        site = Source::Extractor.find("https://twitter.com/Mityubi/status/849630665603665920", "https://www.pixiv.net/member_illust.php?mode=medium&illust_id=56735489")
-
-        assert_equal(site.site_name, "Twitter")
-        assert_equal(["https://pbs.twimg.com/media/C8p-gPhVoAMZupS.png:orig"], site.image_urls)
-      end
     end
 
     context "A tweet from a suspended user" do
-      should "not fail" do
+      strategy_should_work(
-        site = Source::Extractor.find("https://twitter.com/tanso_panz/status/1192429800717029377")
+        "https://twitter.com/tanso_panz/status/1192429800717029377",
-
+        tag_name: "tanso_panz",
-        assert_equal(site.site_name, "Twitter")
+        profile_url: "https://twitter.com/tanso_panz",
-        assert_equal("tanso_panz", site.tag_name)
+        image_urls: []
-        assert_equal("https://twitter.com/tanso_panz", site.profile_url)
+      )
-        assert_equal([], site.image_urls)
-      end
     end
 
-    context "A tweet with hashtags" do
+    context "A profile banner image" do
-      should "ignore common suffixes when translating hashtags" do
+      strategy_should_work(
-        as(create(:user)) do
+        "https://pbs.twimg.com/profile_banners/780804311529906176/1475001696",
-          create(:tag, name: "nishizumi_miho", post_count: 1)
+        image_urls: ["https://pbs.twimg.com/profile_banners/780804311529906176/1475001696/1500x500"],
-          create(:wiki_page, title: "nishizumi_miho", other_names: "西住みほ")
+        download_size: 128_141,
-        end
+        profile_url: nil
+        # profile_url: "https://twitter.com/intent/user?user_id=780804311529906176"
+        # XXX we COULD fully support these by setting the page_url to https://twitter.com/Kekeflipnote/header_photo, but it's a lot of work for a niche case
+      )
+    end
 
-        site = Source::Extractor.find("https://twitter.com/kasaishin100/status/1186658635226607616")
+    context "A profile banner image sample" do
+      strategy_should_work(
+        "https://pbs.twimg.com/profile_banners/780804311529906176/1475001696/600x200",
+        image_urls: ["https://pbs.twimg.com/profile_banners/780804311529906176/1475001696/1500x500"],
+        download_size: 128_141,
+        profile_url: nil
+      )
+    end
 
-        assert_includes(site.tags.map(&:first), "西住みほ生誕祭2019")
+    context "A tweet with hashtags with normalizable prefixes" do
-        assert_includes(site.normalized_tags, "西住みほ")
+      strategy_should_work(
-        assert_includes(site.translated_tags.map(&:name), "nishizumi_miho")
+        "https://twitter.com/kasaishin100/status/1186658635226607616",
-      end
+        tags: ["西住みほ生誕祭2019"],
+        normalized_tags: ["西住みほ"]
+      )
+    end
+
+    context "A tweet with mentions that can be converted to dtext" do
+      strategy_should_work(
+        "https://twitter.com/noizave/status/875768175136317440",
+        dtext_artist_commentary_desc: 'test "#foo":[https://twitter.com/hashtag/foo] "#ホワイトデー":[https://twitter.com/hashtag/ホワイトデー] "@noizave":[https://twitter.com/noizave]\'s blah http://www.example.com <>& 😀'
+      )
+    end
+
+    context "A tweet with normalizable unicode text" do
+      strategy_should_work(
+        "https://twitter.com/aprilarcus/status/367557195186970624",
+        artist_commentary_desc: "𝖸𝗈 𝐔𝐧𝐢𝐜𝐨𝐝𝐞 𝗅 𝗁𝖾𝗋𝖽 𝕌 𝗅𝗂𝗄𝖾 𝑡𝑦𝑝𝑒𝑓𝑎𝑐𝑒𝑠 𝗌𝗈 𝗐𝖾 𝗉𝗎𝗍 𝗌𝗈𝗆𝖾 𝚌𝚘𝚍𝚎𝚙𝚘𝚒𝚗𝚝𝚜 𝗂𝗇 𝗒𝗈𝗎𝗋 𝔖𝔲𝔭𝔭𝔩𝔢𝔪𝔢𝔫𝔱𝔞𝔯𝔶 𝔚𝔲𝔩𝔱𝔦𝔩𝔦𝔫𝔤𝔳𝔞𝔩 𝔓𝔩𝔞𝔫𝔢 𝗌𝗈 𝗒𝗈𝗎 𝖼𝖺𝗇 𝓮𝓷𝓬𝓸𝓭𝓮 𝕗𝕠𝕟𝕥𝕤 𝗂𝗇 𝗒𝗈𝗎𝗋 𝒇𝒐𝒏𝒕𝒔.",
+        dtext_artist_commentary_desc: "Yo Unicode l herd U like typefaces so we put some codepoints in your Supplementary Wultilingval Plane so you can encode fonts in your fonts."
+      )
+    end
+
+    context "A tweet with normalizable full-width hashtags" do
+      strategy_should_work(
+        "https://twitter.com/corpsmanWelt/status/1037724260075069441",
+        artist_commentary_desc: %{新しいおともだち\n＃けものフレンズ https://t.co/sEAuu16yAQ},
+        dtext_artist_commentary_desc: %{新しいおともだち\n"#けものフレンズ":[https://twitter.com/hashtag/けものフレンズ]}
+      )
     end
 
     should "Parse Twitter URLs correctly" do
@@ -343,7 +228,7 @@ module Sources
       assert(Source::URL.profile_url?("https://twitter.com/intent/user?screen_name=ryuudog_NFT"))
       assert(Source::URL.profile_url?("https://twitter.com/i/user/889592953"))
 
-      refute(Source::URL.profile_url?("https://twitter.com/home"))
+      assert_not(Source::URL.profile_url?("https://twitter.com/home"))
 
       assert_nil(Source::URL.parse("https://twitter.com/i/status/1261877313349640194").username)
       assert_nil(Source::URL.parse("https://twitter.com/i/web/status/1261877313349640194").username)