Merge branch 'master' into patch-1

.github/workflows/third-party.yml

@@ -60,3 +60,4 @@ jobs:
         bundle exec jekyll build -s sandbox -d sandbox/_site --trace
     - name: Memory Analysis of Jekyll Build (WITH SEO Tag)
       run: bash jekyll-seo-tag/.github/workflows/scripts/memprof
+      if: "!contains(matrix.jekyll_version, '3.')"

History.markdown

@@ -9,6 +9,7 @@
 
   * Added the `twitter:description` tag (#464)
   * Support for image paths relative to the page's directory (#466)
+  * Truncate the description value to 100 words (#492)
 
 ## 2.8.0 / 2022-02-04
 

docs/advanced-usage.md

@@ -80,6 +80,27 @@ You can set it the same way as the other author properties. For example, you can
     url: https://example.com/
   ```
 
+### Customizing description length
+
+By default, the description is limited to the first 100 words of the full content.
+
+You can adjust this limit at the page level, by using the `seo_description_max_words` page property:
+
+```yml
+  seo_description_max_words: 200
+```
+
+You can also set a default site-wide value for all pages using [Front Matter defaults](https://jekyllrb.com/docs/configuration/front-matter-defaults/) in your `_config.yml` file:
+
+```yml
+defaults:
+  - scope:
+      path: ""
+    values:
+      seo_description_max_words: 200
+```
+
+
 ### Customizing JSON-LD output
 
 The following options can be set for any particular page. While the default options are meant to serve most users in the most common circumstances, there may be situations where more precise control is necessary.

lib/jekyll-seo-tag/drop.rb

@@ -89,7 +89,8 @@ module Jekyll
 
       def description
         @description ||= begin
-          format_string(page["description"] || page["excerpt"]) || site_description
+          value = format_string(page["description"] || page["excerpt"]) || site_description
+          snippet(value, description_max_words)
         end
       end
 
@@ -175,6 +176,10 @@ module Jekyll
         end
       end
 
+      def description_max_words
+        @description_max_words ||= page["seo_description_max_words"] || 100
+      end
+
       private
 
       def filters
@@ -217,6 +222,13 @@ module Jekyll
         string unless string.empty?
       end
 
+      def snippet(string, max_words)
+        return string if string.nil?
+
+        result = string.split(%r!\s+!, max_words + 1)[0...max_words].join(" ")
+        result.length < string.length ? result.concat("…") : result
+      end
+
       def seo_name
         @seo_name ||= format_string(page_seo["name"]) if page_seo["name"]
       end

lib/jekyll-seo-tag/image_drop.rb

@@ -61,6 +61,7 @@ module Jekyll
 
       def absolute_url
         return unless raw_path
+
         @absolute_url ||= build_absolute_path
       end
 

spec/jekyll_seo_tag/drop_spec.rb

@@ -233,6 +233,24 @@ RSpec.describe Jekyll::SeoTag::Drop do
           expect(subject.description).to be_nil
         end
       end
+
+      context "truncation" do
+        context "without seo_description_max_words" do
+          let(:page_meta) { { "description" => "word " * 150 } }
+
+          it "truncates the description to the first 200 words" do
+            expect(subject.description).to eql(("word " * 100).chop.concat("…"))
+          end
+        end
+
+        context "with an explicit seo_description_max_words property" do
+          let(:page_meta) { { "description" => "For a long time, I went to bed early", "seo_description_max_words" => 6 } }
+
+          it "truncates the description to the configured words count" do
+            expect(subject.description).to eql("For a long time, I went…")
+          end
+        end
+      end
     end
 
     context "author" do

spec/jekyll_seo_tag_integration_spec.rb

@@ -350,6 +350,23 @@ RSpec.describe Jekyll::SeoTag do
       it "removes null values from JSON-LD" do
         expect(output).to_not match(%r!:null!)
       end
+
+      context "description" do
+        context "with page.seo_description_max_words" do
+          let(:meta) do
+            {
+              "title"                     => "post",
+              "description"               => "For a long time, I went to bed early",
+              "image"                     => "/img.png",
+              "seo_description_max_words" => 6,
+            }
+          end
+
+          it "truncates the description" do
+            expect(json_data["description"]).to eql("For a long time, I went…")
+          end
+        end
+      end
     end
   end