Implement Unified Embeds for StackExchange and StackOverflow URLs (#16415)

* complete implementation and add specs * improvements * more robust condition * update API version; re-record VCR cassettes * PR review comments
2022-02-07 11:21:11 -05:00 · 2022-02-07 11:21:11 -05:00 · 63c957b12d
commit 63c957b12d
parent 0559a808c4
5 changed files with 108 additions and 78 deletions
--- a/app/liquid_tags/stackexchange_tag.rb
+++ b/app/liquid_tags/stackexchange_tag.rb
@ -1,7 +1,11 @@
 class StackexchangeTag < LiquidTagBase
  PARTIAL = "liquids/stackexchange".freeze
-  # update API version here and in stackexchange_tag_spec when a new version is out
-  API_URL = "https://api.stackexchange.com/2.2/".freeze
+  REGISTRY_REGEXP = %r{https://(?:(?<subdomain>\w+)\.)?(?:stackexchange\.com|stackoverflow\.com)/(?<post_type>q|a|questions)/(?<id>\d{1,20})}
+  ID_REGEXP = /\A(?<id>\d{1,20})\Z/
+  SITE_REGEXP = /(?<subdomain>\b[a-zA-Z]+\b)/
+  REGEXP_OPTIONS = [REGISTRY_REGEXP, ID_REGEXP, SITE_REGEXP].freeze
+  STACKOVERFLOW_REGEXP = %r{https://stackoverflow\.com/(q|a|questions)/\d{1,20}(?:/[\w\-]+)?}
+  API_URL = "https://api.stackexchange.com/2.3/".freeze
  # Filter codes come from the example tools in the docs. For example: https://api.stackexchange.com/docs/posts-by-ids
  FILTERS = {
    "post" => "!3tz1WbZW5JxrG-f99",
@ -9,16 +13,16 @@ class StackexchangeTag < LiquidTagBase
    "question" => "!*1SgQGDOL9bPBHULz9sKS.y6qv7V9fYNszvdhDuv5",
    "site" => "!mWxO_PNa4i"
  }.freeze
-  ID_REGEXP = /\A\d{1,20}\z/

  attr_reader :site, :post_type

  def initialize(_tag_name, input, _parse_context)
    super

-    @site = parse_site(input.strip)
-    @post_type = "question"
-    @json_content = get_data(input.strip)
+    stripped_input  = strip_tags(input)
+    unescaped_input = CGI.unescape_html(stripped_input)
+    @site = parse_site(unescaped_input)
+    @json_content, @post_type = get_data(unescaped_input)
  end

  def render(_context)
@ -42,43 +46,29 @@ class StackexchangeTag < LiquidTagBase

  private

-  def valid_site?(site)
-    (site =~ /[a-z.]+/i)&.zero?
-  end
-
  def parse_site(input)
-    return "stackoverflow" if tag_name == "stackoverflow"
+    return "stackoverflow" if tag_name == "stackoverflow" || input.match?(STACKOVERFLOW_REGEXP)

-    site = input.match(/[a-z.]+/i)[0]
-    unless valid_site?(site)
-      raise StandardError, I18n.t("liquid_tags.stackexchange_tag.invalid_site", tag: tag_name, input: input)
-    end
+    match = pattern_match_for(input, REGEXP_OPTIONS)
+    # rubocop:disable Layout/LineLength
+    raise StandardError, I18n.t("liquid_tags.stackexchange_tag.invalid_site", tag: tag_name, input: input) unless match && match_group_present?(match, "subdomain")
+    # rubocop:enable Layout/LineLength

-    site
+    match[:subdomain].downcase
  end

-  def valid_input?(input)
-    return false if input.nil?
-
-    ID_REGEXP.match?(input.split.first)
-  end
-
-  def handle_response_error(response, input)
-    raise StandardError, "Calling StackExchange API failed: #{response&.error_message}" if response.code != 200
-
-    return unless response["items"].length.zero?
-
-    raise StandardError, I18n.t("liquid_tags.stackexchange_tag.post_not_found", tag: tag_name, input: input)
+  def match_group_present?(match, group_name)
+    match.names.include?(group_name)
  end

  def get_data(input)
-    unless valid_input?(input)
+    id = input.split.first
+    match = pattern_match_for(id, REGEXP_OPTIONS)
+    unless match && match_group_present?(match, "id")
      raise StandardError, I18n.t("liquid_tags.stackexchange_tag.invalid_id", tag: tag_name, input: input)
    end

-    id = input.split.first
-
-    url = "#{API_URL}posts/#{id}?site=#{@site}&filter=#{FILTERS['post']}" \
+    url = "#{API_URL}posts/#{match[:id]}?site=#{@site}&filter=#{FILTERS['post']}" \
          "&key=#{ApplicationConfig['STACK_EXCHANGE_APP_KEY']}"
    post_response = HTTParty.get(url)

@ -86,15 +76,24 @@ class StackexchangeTag < LiquidTagBase

    @post_type = post_response["items"][0]["post_type"]

-    url = "#{API_URL}#{@post_type.pluralize}/#{id}?site=#{@site}" \
+    url = "#{API_URL}#{@post_type.pluralize}/#{match[:id]}?site=#{@site}" \
          "&filter=#{FILTERS[@post_type]}&key=#{ApplicationConfig['STACK_EXCHANGE_APP_KEY']}"
    final_response = HTTParty.get(url)

    handle_response_error(final_response, input)

-    final_response["items"][0]
+    [final_response["items"][0], @post_type]
+  end
+
+  def handle_response_error(response, input)
+    raise StandardError, "Calling StackExchange API failed: #{response&.error_message}" unless response.ok?
+
+    return unless response["items"].length.zero?
+
+    raise StandardError, I18n.t("liquid_tags.stackexchange_tag.post_not_found", tag: tag_name, input: input)
  end
 end

 Liquid::Template.register_tag("stackoverflow", StackexchangeTag)
 Liquid::Template.register_tag("stackexchange", StackexchangeTag)
+UnifiedEmbed.register(StackexchangeTag, regexp: StackexchangeTag::REGISTRY_REGEXP)
--- a/spec/liquid_tags/unified_embed/registry_spec.rb
+++ b/spec/liquid_tags/unified_embed/registry_spec.rb
@ -68,6 +68,15 @@ RSpec.describe UnifiedEmbed::Registry do
      "https://stackblitz.com/edit/web-platform-3tqbd4?embed=1&file=index.html&theme=light",
    ]

+    valid_stackexchange_stackoverflow_url_formats = [
+      "https://travel.stackexchange.com/questions/172014/is-it-okay-to-mix-in-local-language-when-i-know-it-poorly",
+      "https://diy.stackexchange.com/q/244088",
+      "https://academia.stackexchange.com/a/181893",
+      "https://stackoverflow.com/q/70974409/9091371",
+      "https://stackoverflow.com/questions/70976451/changing-h1-element-based-on-input-element-content-vanilla-js",
+      "https://stackoverflow.com/a/70976251/9091371",
+    ]
+
    valid_twitch_url_formats = [
      "https://clips.twitch.tv/embed?clip=SpeedyVivaciousDolphinKappaRoss-IQl5YslMAGKbMOGM&parent=www.example.com",
      "https://player.twitch.tv/?video=1222841752&parent=www.example.com",
@ -226,6 +235,13 @@ RSpec.describe UnifiedEmbed::Registry do
        .to eq(StackeryTag)
    end

+    it "returns StackexchangeTag for a valid stackexchange or stackoverflow url", :aggregate_failures do
+      valid_stackexchange_stackoverflow_url_formats.each do |url|
+        expect(described_class.find_liquid_tag_for(link: url))
+          .to eq(StackexchangeTag)
+      end
+    end
+
    it "returns TweetTag for a tweet url" do
      expect(described_class.find_liquid_tag_for(link: "https://twitter.com/aritdeveloper/status/1483614684884484099"))
        .to eq(TweetTag)
--- a/spec/support/fixtures/vcr_cassettes/stackexchange_tag_deleted_question.yml
+++ b/spec/support/fixtures/vcr_cassettes/stackexchange_tag_deleted_question.yml
@ -2,7 +2,7 @@
 http_interactions:
 - request:
    method: get
-    uri: https://api.stackexchange.com/2.2/posts/59679915?filter=!3tz1WbZW5JxrG-f99&key=&site=askubuntu
+    uri: https://api.stackexchange.com/2.3/posts/59679915?filter=!3tz1WbZW5JxrG-f99&key=&site=askubuntu
    body:
      encoding: US-ASCII
      string: ''
@ -18,12 +18,14 @@ http_interactions:
      code: 200
      message: OK
    headers:
+      Cache-Control:
+      - private
      Content-Length:
      - '87'
      Content-Type:
      - application/json; charset=utf-8
      Strict-Transport-Security:
-      - max-age=2592000
+      - max-age=15552000
      Access-Control-Allow-Origin:
      - "*"
      Access-Control-Allow-Methods:
@ -33,12 +35,13 @@ http_interactions:
      X-Content-Type-Options:
      - nosniff
      X-Request-Guid:
-      - 356f2fd4-3560-40c4-a494-6a4f0f361e62
+      - 39492868-d51e-49b4-9c87-d5d963306c83
+      Content-Security-Policy:
+      - upgrade-insecure-requests; frame-ancestors 'self' https://stackexchange.com
      Date:
-      - Mon, 13 Jan 2020 12:23:24 GMT
+      - Fri, 04 Feb 2022 14:00:22 GMT
    body:
      encoding: ASCII-8BIT
-      string: '{"items":[],"has_more":false,"quota_max":300,"quota_remaining":287}'
-    http_version: 
-  recorded_at: Mon, 13 Jan 2020 12:23:25 GMT
-recorded_with: VCR 5.0.0
+      string: '{"items":[],"has_more":false,"quota_max":300,"quota_remaining":192}'
+  recorded_at: Fri, 04 Feb 2022 14:00:22 GMT
+recorded_with: VCR 6.0.0
--- a/spec/support/fixtures/vcr_cassettes/stackexchange_tag_stackexchange.yml
+++ b/spec/support/fixtures/vcr_cassettes/stackexchange_tag_stackexchange.yml
@ -2,7 +2,7 @@
 http_interactions:
 - request:
    method: get
-    uri: https://api.stackexchange.com/2.2/posts/1163633?filter=!3tz1WbZW5JxrG-f99&key=&site=askubuntu
+    uri: https://api.stackexchange.com/2.3/posts/1163633?filter=!3tz1WbZW5JxrG-f99&key=&site=askubuntu
    body:
      encoding: US-ASCII
      string: ''
@ -18,12 +18,14 @@ http_interactions:
      code: 200
      message: OK
    headers:
+      Cache-Control:
+      - private
      Content-Length:
-      - '138'
+      - '139'
      Content-Type:
      - application/json; charset=utf-8
      Strict-Transport-Security:
-      - max-age=2592000
+      - max-age=15552000
      Access-Control-Allow-Origin:
      - "*"
      Access-Control-Allow-Methods:
@ -33,17 +35,18 @@ http_interactions:
      X-Content-Type-Options:
      - nosniff
      X-Request-Guid:
-      - a8df412f-cc88-496c-ad89-9e40ad387650
+      - 1f73ab59-8040-47d7-9562-c3ffbff9be16
+      Content-Security-Policy:
+      - upgrade-insecure-requests; frame-ancestors 'self' https://stackexchange.com
      Date:
-      - Mon, 13 Jan 2020 12:22:38 GMT
+      - Fri, 04 Feb 2022 14:00:21 GMT
    body:
      encoding: ASCII-8BIT
-      string: '{"items":[{"last_activity_date":1565109531,"post_type":"answer","post_id":1163633}],"has_more":false,"quota_max":300,"quota_remaining":289}'
-    http_version: 
-  recorded_at: Mon, 13 Jan 2020 12:22:39 GMT
+      string: '{"items":[{"last_activity_date":1565109531,"post_type":"answer","post_id":1163633}],"has_more":false,"quota_max":300,"quota_remaining":192}'
+  recorded_at: Fri, 04 Feb 2022 14:00:21 GMT
 - request:
    method: get
-    uri: https://api.stackexchange.com/2.2/answers/1163633?filter=!.Fjr38AQkcvWfJTF-2exSL50At_pT&key=&site=askubuntu
+    uri: https://api.stackexchange.com/2.3/answers/1163633?filter=!.Fjr38AQkcvWfJTF-2exSL50At_pT&key=&site=askubuntu
    body:
      encoding: US-ASCII
      string: ''
@ -59,12 +62,14 @@ http_interactions:
      code: 200
      message: OK
    headers:
+      Cache-Control:
+      - private
      Content-Length:
-      - '608'
+      - '607'
      Content-Type:
      - application/json; charset=utf-8
      Strict-Transport-Security:
-      - max-age=2592000
+      - max-age=15552000
      Access-Control-Allow-Origin:
      - "*"
      Access-Control-Allow-Methods:
@ -74,12 +79,14 @@ http_interactions:
      X-Content-Type-Options:
      - nosniff
      X-Request-Guid:
-      - 9e23222f-b1d1-4841-9349-623e806a0321
+      - 5b3286b4-7d9b-44d9-9ed1-a2b0abda5ae4
+      Content-Security-Policy:
+      - upgrade-insecure-requests; frame-ancestors 'self' https://stackexchange.com
      Date:
-      - Mon, 13 Jan 2020 12:22:39 GMT
+      - Fri, 04 Feb 2022 14:00:21 GMT
    body:
      encoding: ASCII-8BIT
-      string: '{"items":[{"tags":[],"comment_count":3,"is_accepted":false,"score":1,"last_activity_date":1565109531,"last_edit_date":1565109531,"creation_date":1565033637,"answer_id":1163633,"question_id":885432,"link":"https://askubuntu.com/questions/885432/help-setting-up-sshfs/1163633#1163633","title":"Help
+      string: '{"items":[{"tags":[],"comment_count":3,"is_accepted":false,"score":3,"last_activity_date":1565109531,"last_edit_date":1565109531,"creation_date":1565033637,"answer_id":1163633,"question_id":885432,"link":"https://askubuntu.com/questions/885432/help-setting-up-sshfs/1163633#1163633","title":"Help
        setting up sshfs","body":"<p>As you are getting the following error message:</p>\n\n<blockquote>\n  <p>gpasswd:
        group''fuse'' does not exist in /etc/group</p>\n</blockquote>\n\n<p>This indicates
        that there is no such group, yet!<br>\nPlease test if there is and subsequently
@ -88,7 +95,6 @@ http_interactions:
        there is it would look like:  </p>\n\n<pre><code>fuse:x:1001:  \n</code></pre></li>\n<li><p>If
        there isn''t, add one like this:  </p>\n\n<pre><code>sudo groupadd fuse  \n</code></pre></li>\n<li><p>At
        this point you can add yourself to the group with:  </p>\n\n<pre><code>sudo
-        usermod -a -G fuse your_username \n</code></pre></li>\n</ol>\n"}],"has_more":false,"quota_max":300,"quota_remaining":289}'
-    http_version: 
-  recorded_at: Mon, 13 Jan 2020 12:22:39 GMT
-recorded_with: VCR 5.0.0
+        usermod -a -G fuse your_username \n</code></pre></li>\n</ol>\n"}],"has_more":false,"quota_max":300,"quota_remaining":192}'
+  recorded_at: Fri, 04 Feb 2022 14:00:21 GMT
+recorded_with: VCR 6.0.0
--- a/spec/support/fixtures/vcr_cassettes/stackexchange_tag_stackoverflow.yml
+++ b/spec/support/fixtures/vcr_cassettes/stackexchange_tag_stackoverflow.yml
@ -2,7 +2,7 @@
 http_interactions:
 - request:
    method: get
-    uri: https://api.stackexchange.com/2.2/posts/57496168?filter=!3tz1WbZW5JxrG-f99&key=&site=stackoverflow
+    uri: https://api.stackexchange.com/2.3/posts/57496168?filter=!3tz1WbZW5JxrG-f99&key=&site=stackoverflow
    body:
      encoding: US-ASCII
      string: ''
@ -18,12 +18,14 @@ http_interactions:
      code: 200
      message: OK
    headers:
+      Cache-Control:
+      - private
      Content-Length:
-      - '141'
+      - '143'
      Content-Type:
      - application/json; charset=utf-8
      Strict-Transport-Security:
-      - max-age=2592000
+      - max-age=15552000
      Access-Control-Allow-Origin:
      - "*"
      Access-Control-Allow-Methods:
@ -33,17 +35,18 @@ http_interactions:
      X-Content-Type-Options:
      - nosniff
      X-Request-Guid:
-      - d25e24d9-a659-49ff-ab38-aaaef36d5bbf
+      - 11201cf1-0f43-4e4c-b839-6f9b4e8aa3c3
+      Content-Security-Policy:
+      - upgrade-insecure-requests; frame-ancestors 'self' https://stackexchange.com
      Date:
-      - Mon, 13 Jan 2020 12:20:49 GMT
+      - Fri, 04 Feb 2022 14:00:21 GMT
    body:
      encoding: ASCII-8BIT
-      string: '{"items":[{"last_activity_date":1565806685,"post_type":"question","post_id":57496168}],"has_more":false,"quota_max":300,"quota_remaining":291}'
-    http_version: 
-  recorded_at: Mon, 13 Jan 2020 12:20:50 GMT
+      string: '{"items":[{"last_activity_date":1565806685,"post_type":"question","post_id":57496168}],"has_more":false,"quota_max":300,"quota_remaining":194}'
+  recorded_at: Fri, 04 Feb 2022 14:00:21 GMT
 - request:
    method: get
-    uri: https://api.stackexchange.com/2.2/questions/57496168?filter=!*1SgQGDOL9bPBHULz9sKS.y6qv7V9fYNszvdhDuv5&key=&site=stackoverflow
+    uri: https://api.stackexchange.com/2.3/questions/57496168?filter=!*1SgQGDOL9bPBHULz9sKS.y6qv7V9fYNszvdhDuv5&key=&site=stackoverflow
    body:
      encoding: US-ASCII
      string: ''
@ -59,12 +62,14 @@ http_interactions:
      code: 200
      message: OK
    headers:
+      Cache-Control:
+      - private
      Content-Length:
-      - '532'
+      - '530'
      Content-Type:
      - application/json; charset=utf-8
      Strict-Transport-Security:
-      - max-age=2592000
+      - max-age=15552000
      Access-Control-Allow-Origin:
      - "*"
      Access-Control-Allow-Methods:
@ -74,18 +79,19 @@ http_interactions:
      X-Content-Type-Options:
      - nosniff
      X-Request-Guid:
-      - 644cb806-8678-4b08-bfbf-6f0cdb72a10d
+      - a9c1799b-3e84-4869-8316-7013814a36b7
+      Content-Security-Policy:
+      - upgrade-insecure-requests; frame-ancestors 'self' https://stackexchange.com
      Date:
-      - Mon, 13 Jan 2020 12:20:50 GMT
+      - Fri, 04 Feb 2022 14:00:21 GMT
    body:
      encoding: ASCII-8BIT
-      string: '{"items":[{"tags":["laravel","foreach","view","relationship"],"comment_count":1,"view_count":61,"answer_count":4,"score":0,"last_activity_date":1565806685,"creation_date":1565790943,"question_id":57496168,"link":"https://stackoverflow.com/questions/57496168/use-where-and-limit-to-child-in-foreach","title":"Use
+      string: '{"items":[{"tags":["laravel","foreach","view","relationship"],"comment_count":1,"view_count":79,"answer_count":4,"score":0,"last_activity_date":1565806685,"creation_date":1565790943,"question_id":57496168,"link":"https://stackoverflow.com/questions/57496168/use-where-and-limit-to-child-in-foreach","title":"Use
        &quot;where&quot; and &quot;limit&quot; to child in @foreach","body":"<p>I
        want to display all user''s profile into views and they posts. It''s pretty
        simple:</p>\n\n<pre><code>@foreach($profiles as $profile)\n{{ $profile-&gt;name
        }}\n @foreach($profile-&gt;posts as $post)\n   {{$post-&gt;title}}\n @endforeach\n@endforeach\n</code></pre>\n\n<p>But
        I want to display only the latests posts (->orderBy(''created_at'', ''desc'')->limit(4)
-        ) and only accepted posts (->where(''accept'', 1)). How can I do that?</p>\n"}],"has_more":false,"quota_max":300,"quota_remaining":290}'
-    http_version: 
-  recorded_at: Mon, 13 Jan 2020 12:20:51 GMT
-recorded_with: VCR 5.0.0
+        ) and only accepted posts (->where(''accept'', 1)). How can I do that?</p>\n"}],"has_more":false,"quota_max":300,"quota_remaining":193}'
+  recorded_at: Fri, 04 Feb 2022 14:00:21 GMT
+recorded_with: VCR 6.0.0