Update history to reflect merge of #492 [ci skip]

Merge pull request 492

.github/workflows/ci.yml

@@ -0,0 +1,103 @@
+name: Continuous Integration
+
+on:
+  push:
+    branches:
+    - master
+  pull_request:
+    branches:
+    - master
+
+jobs:
+  j4:
+    if: "!contains(github.event.commits[0].message, '[ci skip]')"
+    name: "Jekyll ${{ matrix.jekyll_version }} (Ruby ${{ matrix.ruby_version }})"
+    runs-on: 'ubuntu-latest'
+    env:
+      JEKYLL_VERSION: ${{ matrix.jekyll_version }}
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby_version:
+        - 2.5
+        - 2.7
+        - 3.1
+        jekyll_version:
+        - "~> 4.0"
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 5
+      - name: "Set up Ruby ${{ matrix.ruby_version }}"
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby_version }}
+          bundler-cache: true
+      - name: Execute tests
+        run:  bundle exec rspec
+  j3:
+    if: "!contains(github.event.commits[0].message, '[ci skip]')"
+    name: "Jekyll ${{ matrix.jekyll_version }} (Ruby ${{ matrix.ruby_version }})"
+    runs-on: 'ubuntu-latest'
+    env:
+      JEKYLL_VERSION: ${{ matrix.jekyll_version }}
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby_version:
+        - 2.5
+        jekyll_version:
+        - "~> 3.9"
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 5
+      - name: "Set up Ruby ${{ matrix.ruby_version }}"
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby_version }}
+          bundler-cache: true
+      - name: Execute tests
+        run:  bundle exec rspec
+
+  style_check:
+    if: "!contains(github.event.commits[0].message, '[ci skip]')"
+    name: "Code Style Check (Ruby ${{ matrix.ruby_version }})"
+    runs-on: 'ubuntu-latest'
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby_version:
+        - 2.5
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 5
+      - name: "Set up Ruby ${{ matrix.ruby_version }}"
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby_version }}
+          bundler-cache: true
+      - name: Check Style Offenses
+        run:  bundle exec rubocop -S -D
+
+  gem_build:
+    if: "!contains(github.event.commits[0].message, '[ci skip]')"
+    name: "Test Gem build (Ruby ${{ matrix.ruby_version }})"
+    runs-on: 'ubuntu-latest'
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby_version:
+        - 2.5
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 5
+      - name: "Set up Ruby ${{ matrix.ruby_version }}"
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby_version }}
+          bundler-cache: true
+      - name: Test Gem build
+        run:  bundle exec gem build jekyll-seo-tag.gemspec

.github/workflows/release.yml

@@ -0,0 +1,33 @@
+name: Release Gem
+
+on:
+  push:
+    branches:
+      - master
+    paths:
+      - "lib/**/version.rb"
+
+jobs:
+  release:
+    if: "github.repository_owner == 'jekyll'"
+    name: "Release Gem (Ruby ${{ matrix.ruby_version }})"
+    runs-on: "ubuntu-latest"
+    strategy:
+      fail-fast: true
+      matrix:
+        ruby_version:
+          - 2.7
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v2
+      - name: "Set up Ruby ${{ matrix.ruby_version }}"
+        uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby_version }}
+          bundler-cache: true
+      - name: Build and Publish Gem
+        uses: ashmaroli/release-gem@dist
+        with:
+          gemspec_name: jekyll-seo-tag
+        env:
+          GEM_HOST_API_KEY: ${{ secrets.RUBYGEMS_GEM_PUSH_API_KEY }}

.github/workflows/scripts/memprof

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+
+bundle exec ruby jekyll-seo-tag/.github/workflows/scripts/memprof.rb sandbox
+exit 0

.github/workflows/scripts/memprof.rb

@@ -0,0 +1,14 @@
+# frozen_string_literal: true
+
+require 'jekyll'
+require 'memory_profiler'
+
+MemoryProfiler.report(allow_files: 'lib/jekyll-seo-tag') do
+  Jekyll::PluginManager.require_from_bundler
+  Jekyll::Commands::Build.process({
+    "source"             => File.expand_path(ARGV[0]),
+    "destination"        => File.expand_path("#{ARGV[0]}/_site"),
+    "disable_disk_cache" => true,
+  })
+  puts ''
+end.pretty_print(scale_bytes: true, normalize_paths: true)

.github/workflows/third-party.yml

@@ -0,0 +1,63 @@
+name: Third-Party Repository Profiling
+
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    branches:
+      - master
+jobs:
+  build_n_profile:
+    name: "Third-Party Repo Profile (Jekyll ${{ matrix.jekyll_version }}, Ruby ${{ matrix.ruby_version }})"
+    runs-on: "ubuntu-latest"
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby_version:
+          - 2.7
+        jekyll_version:
+          - "~> 4.0"
+          - "~> 3.9"
+    env:
+      BUNDLE_GEMFILE: "sandbox/Gemfile"
+      BUNDLE_PATH: "vendor/bundle"
+      BUNDLE_JOBS: 4
+      BUNDLE_RETRY: 3
+      JEKYLL_VERSION: ${{ matrix.jekyll_version }}
+    steps:
+    - name: Checkout Jekyll SEO Tag
+      uses: actions/checkout@v2
+      with:
+        fetch-depth: 5
+        path: jekyll-seo-tag
+    - name: Checkout Third-Party Repository (WITHOUT SEO Tag)
+      uses: actions/checkout@v2
+      with:
+        repository: ashmaroli/tomjoht.github.io
+        ref: "no-seo-tag"
+        path: alpha-sandbox
+    - name: Checkout Same Third-Party Repository (WITH SEO Tag)
+      uses: actions/checkout@v2
+      with:
+        repository: ashmaroli/tomjoht.github.io
+        ref: "seo-tag"
+        path: sandbox
+    - name: "Set up Ruby ${{ matrix.ruby_version }}"
+      uses: ruby/setup-ruby@v1
+      with:
+        ruby-version: ${{ matrix.ruby_version }}
+        bundler-cache: true
+    - name: Run Jekyll Build (WITHOUT SEO Tag) 3 times
+      run: |
+        bundle exec jekyll build -s alpha-sandbox -d alpha-sandbox/_site --trace
+        bundle exec jekyll build -s alpha-sandbox -d alpha-sandbox/_site --trace
+        bundle exec jekyll build -s alpha-sandbox -d alpha-sandbox/_site --trace
+    - name: Run Jekyll Build (WITH SEO Tag) 3 times
+      run: |
+        bundle exec jekyll build -s sandbox -d sandbox/_site --trace
+        bundle exec jekyll build -s sandbox -d sandbox/_site --trace
+        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.')"

.gitignore

@@ -1,10 +1,9 @@
-/.bundle/
-/.yardoc
-/Gemfile.lock
-/_yardoc/
-/coverage/
-/doc/
-/pkg/
-/spec/reports/
-/tmp/
 *.gem
+.bundle
+.yardoc
+Gemfile.lock
+pkg
+spec/reports/
+spec/fixtures/.jekyll-cache
+tmp/
+vendor/bundle

.rspec

@@ -1,2 +1,2 @@
---format documentation
 --color
+--require spec_helper

.rubocop.yml

@@ -1,17 +1,42 @@
-inherit_gem:
-  jekyll: .rubocop.yml
+inherit_from: .rubocop_todo.yml
 
-Metrics/LineLength:
+require: rubocop-jekyll
+inherit_gem:
+  rubocop-jekyll: .rubocop.yml
+
+AllCops:
+  TargetRubyVersion: 2.5
+  SuggestExtensions: false
+  Exclude:
+    - vendor/**/*
+
+Layout/LineEndStringConcatenationIndentation:
+  Enabled: true
+Layout/LineLength:
   Exclude:
     - spec/**/*
     - jekyll-seo-tag.gemspec
 
-Style/Documentation:
+Lint/EmptyInPattern:
   Enabled: false
 
-Style/FileName:
-  Enabled: false
-
-AllCops:
+Metrics/BlockLength:
   Exclude:
-    - vendor/**/*
+    - spec/**/*
+
+Naming/InclusiveLanguage:
+  Enabled: false
+
+Performance/MapCompact:
+  Enabled: true
+Performance/RedundantEqualityComparisonBlock:
+  Enabled: true
+Performance/RedundantSplitRegexpArgument:
+  Enabled: true
+
+Style/InPatternThen:
+  Enabled: false
+Style/MultilineInPatternThen:
+  Enabled: false
+Style/QuotedSymbols:
+  Enabled: true

.rubocop_todo.yml

@@ -0,0 +1,46 @@
+# This configuration was generated by
+# `rubocop --auto-gen-config --auto-gen-only-exclude`
+# on 2021-09-17 06:40:32 UTC using RuboCop version 1.18.4.
+# The point is for the user to remove these configuration records
+# one by one as the offenses are removed from the code base.
+# Note that changes in the inspected code, or installation of new
+# versions of RuboCop, may require this file to be generated again.
+
+# Offense count: 1
+# Configuration parameters: AllowComments.
+Lint/EmptyClass:
+  Exclude:
+    - 'lib/jekyll-seo-tag/version.rb'
+
+# Offense count: 3
+Lint/NoReturnInBeginEndBlocks:
+  Exclude:
+    - 'lib/jekyll-seo-tag/author_drop.rb'
+    - 'lib/jekyll-seo-tag/drop.rb'
+
+# Offense count: 1
+# Cop supports --auto-correct.
+Lint/ToJSON:
+  Exclude:
+    - 'lib/jekyll-seo-tag/json_ld_drop.rb'
+
+# Offense count: 1
+# Configuration parameters: IgnoredMethods, Max.
+Metrics/PerceivedComplexity:
+  Exclude:
+    - 'lib/jekyll-seo-tag/drop.rb'
+
+# Offense count: 1
+# Configuration parameters: MinSize.
+Performance/CollectionLiteralInLoop:
+  Exclude:
+    - 'spec/jekyll_seo_tag/author_drop_spec.rb'
+
+# Offense count: 9
+# Cop supports --auto-correct.
+Style/RedundantBegin:
+  Exclude:
+    - 'lib/jekyll-seo-tag.rb'
+    - 'lib/jekyll-seo-tag/author_drop.rb'
+    - 'lib/jekyll-seo-tag/drop.rb'
+    - 'lib/jekyll-seo-tag/image_drop.rb'

.travis.yml

@@ -1,7 +0,0 @@
-rvm:
-  - 2.2.3
-before_install: gem install bundler
-language: ruby
-script: script/cibuild
-sudo: false
-cache: bundler

Gemfile

@@ -1,17 +1,8 @@
-source 'https://rubygems.org'
-require 'json'
-require 'open-uri'
+# frozen_string_literal: true
+
+source "https://rubygems.org"
 
 gemspec
 
-group :development, :test do
-  versions = JSON.parse(open('https://pages.github.com/versions.json').read)
-  versions.delete('ruby')
-  versions.delete('jekyll-seo-tag')
-  versions.delete('github-pages')
-  versions.delete('jekyll') # Remove this line when GitHub Pages supports 3.3.0
-
-  versions.each do |dep, version|
-    gem dep, version
-  end
-end
+gem "jekyll", ENV["JEKYLL_VERSION"] if ENV["JEKYLL_VERSION"]
+gem "kramdown-parser-gfm" if ENV["JEKYLL_VERSION"] == "~> 3.9"

History.markdown

@@ -0,0 +1,228 @@
+## HEAD
+
+### Development Fixes
+
+  * Add Ruby 3.1 to CI matrix (#459)
+  * chore: remove git.io (#462)
+
+### Minor Enhancements
+
+  * 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
+
+### Minor Enhancements
+
+  * Allow to set type for author (#427)
+  * Allow setting `author.url` (#453)
+  * Implement Facebook domain verification (#455)
+  * Add `og:image:alt` and `twitter:image:alt` (#438)
+  * Sort JSON-LD data by key (#458)
+
+### Bug Fixes
+
+  * Set the default `og:type` to 'website' (#391)
+  * Template: Remove double new line (#454)
+
+### Development Fixes
+
+  * Fix typo in source code comment (#449)
+  * Set up Continuous Integration via GH Actions (#450)
+  * Bump RuboCop to v1.18.x (#452)
+  * Add workflow to release gem via GH Actions
+
+## 2.7.1 / 2020-10-18
+
+### Development Fixes
+
+  * refactor: mutate site payload instead of duplicating it (#419)
+
+## 2.7.0 / 2020-10-18
+
+### Minor Enhancements
+
+  * Change pagination message with `seo_paginator_message` option (#324)
+  * Make Twitter Summary Card without having Twitter account (#284)
+  * Prefer site.tagline to site.description for page title (#356)
+  * Render og:locale meta only when defined explicitly (#388)
+
+### Bug Fixes
+
+  * Ensure a single leading `@` for twitter usernames (#367)
+
+### Development Fixes
+
+  * chore(deps): require Ruby > 2.4.0 EOL
+  * test: fix locale specs that use the fallback locale (#360)
+  * refactor: Replace read-only empty hash with private constant (#418)
+  * refactor: Mutate hash literals instead of duplicating them (#417)
+  * refactor: Reduce allocations of instance-agnostic objects (#376)
+  * refactor: Memoize #author_hash in SeoTag::AuthorDrop (#342)
+  * refactor: simplify conditional in SeoTag::Drop#date_modified (#343)
+  * chore(ci): profile seo-tag plugin on a third-party repository (#414)
+  * chore(ci): Jekyll v4.0 (#372)
+  * chore(ci): test against current stable Ruby 2.5 and 2.7 (#385)
+  * style: align with latest jekyll-rubocop (#382)
+  * fix: Travis builds for Jekyll 3.x (#415)
+
+### Documentation
+
+  * Structured Data Testing Tool is deprecated (#409)
+  * Rename Google webmaster tools to Google Search Console  (#403)
+  * Improve documentation on plugin usage (#399)
+  * remove Google+ from example snippet (#358)
+  * HTTPS link to https://ogp.me/ (#359)
+  * HTTPS links to schema.org (#350)
+  * use example.com for example URL (#351)
+
+## 2.6.1 / 2019-05-17
+
+### Development Fixes
+
+  * Test against Jekyll 4.x (#336)
+
+## 2.6.0 / 2019-03-16
+
+### Minor Enhancements
+
+  * Twitter Image and Title (#330)
+
+### Bug Fixes
+
+  * Do not cache the drop payload for SeoTag (#306)
+  * Update url of schema website (#296)
+
+### Development Fixes
+
+  * Relax version constraint on Bundler (#325)
+  * chore(ci): Add Ruby 2.6, drop Ruby 2.3 (#326)
+  * chore (ci): remove deprecated `sudo: false` in .travis.yml (#333)
+  * Lint Ruby code with rubocop-jekyll gem (#302)
+  * chore(deps): bump rubocop-jekyll to v0.4 (#320)
+  * chore(deps): bump rubocop-jekyll to v0.3 (#316)
+  * Correct RuboCop offenses in spec files (#319)
+
+### Documentation
+
+  * Rectify error in Usage documentation (#328)
+
+## 2.5.0 / 2018-05-18
+
+  * Docs: Prevent GitHub Pages from processing Liquid raw tag (#276)
+
+### Documentation
+
+  * Use gems config key for Jekyll < 3.5.0 (#255)
+  * docs/usage - replace "below" with correct link (#280)
+
+### Development Fixes
+
+  * Test against Ruby 2.5 (#260)
+  * add tests for twitter.card types (#289)
+  * Target Ruby 2.3 and Rubocop 0.56.0 (#292)
+
+### Minor Enhancements
+
+  * Add webmaster_verifications for baidu (#263)
+  * Include page number in title (#250)
+  * Configure default Twitter summary card type (V2) (#225)
+
+## 2.4.0 / 2017-12-04
+
+### Minor
+
+  * Add meta generator (#236)
+  * Consistently use self-closing tags (#246)
+  * Strip null values from JSON-LD hash (#249)
+
+### Documentation
+
+  * Avoid deprecation warning when building docs (#243)
+
+### Development Fixes
+
+  * Test against latest Rubies (#242)
+  * Use Nokigiri on CI (#181)
+
+## 2.3.0
+
+### Minor Enhancements
+
+  * Use canonical_url specified in page if present #211
+  * Fix for image.path causing an invalid url error #228
+  * Ensure `site.data.authors` is properly formatted before attempting to retrieve author meta #227
+  * Convert author, image, and JSON-LD to dedicated drops #229
+  * Cache parsed template #231
+  * Define path with `__dir__` #232
+
+### Documentation
+
+  * gems: is deprecated in current Jekyll version of github-pages #230
+
+## 2.2.3
+
+  * Guard against the author's Twitter handle being Nil when stripping @'s #203
+  * Guard against empty title or description strings #206
+
+## 2.2.2
+
+### Minor Enhancements
+
+  * Guard against arrays in subhashes #197
+  * Guard against invalid or missing URLs #199
+
+### Development fixes
+
+  * Remove dynamic GitHub Pages logic from Gemfile #194
+
+## 2.2.1
+
+  * Convert template logic to a Liquid Drop (significant performance improvement) (#184)
+  * Fix for JSON-LD validation warning for images missing required properties (#183)
+
+## 2.2.0
+
+### Major Enhancements
+
+  * Add author meta (#103)
+  * Add og:locale support #166
+  * Add support for Bing and Yandex webmaster tools. Closes #147 (#148)
+  * Add SEO author and date modified to validate JSON-LD output (#151)
+
+### Minor Enhancements
+
+  * Use `|` for title separator (#162)
+  * Use `og:image` for twitter image (#174)
+
+### Development Fixes
+
+  * Style fixes (#170, #157, #149)
+  * Test against latest version of Jekyll (#171)
+  * Bump dev dependencies (#172)
+  * Remove Rake dependency (#180)
+
+## 2.1.0
+
+### Major Enhancement
+
+  * Use new URL filters (#123)
+
+### Minor Enhancements
+
+  * Wraps logo image json data in a publisher property (#133)
+  * Fix duplicated `escape_once` (#93)
+  * Simplify minify regex (#125)
+  * Don't mangle text with newlines #126
+
+### Documentation
+
+  * Add front matter default example for image (#132)
+  * Fix tiny typo (#106)
+  * add example usage of social profiles (#139)
+
+### Development
+
+  * Inherit Jekyll's rubocop config for consistency (#109)
+  * Correct spelling in .travis.yml (#112)

LICENSE.txt

@@ -1,6 +1,6 @@
 The MIT License (MIT)
 
-Copyright (c) 2015 Ben Balter
+Copyright (c) 2015-present Ben Balter and the jekyll-seo-tag contributors
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -1,189 +0,0 @@
-# Jekyll SEO Tag
-
-A Jekyll plugin to add metadata tags for search engines and social networks to better index and display your site's content.
-
-[![Gem Version](https://badge.fury.io/rb/jekyll-seo-tag.svg)](https://badge.fury.io/rb/jekyll-seo-tag) [![Build Status](https://travis-ci.org/jekyll/jekyll-seo-tag.svg)](https://travis-ci.org/jekyll/jekyll-seo-tag)
-
-## What it does
-
-Jekyll SEO Tag adds the following meta tags to your site:
-
-* Pages title (with site title appended when available)
-* Page description
-* Canonical URL
-* Next and previous URLs on paginated pages
-* [JSON-LD Site and post metadata](https://developers.google.com/structured-data/) for richer indexing
-* [Open graph](http://ogp.me/) title, description, site title, and URL (for Facebook, LinkedIn, etc.)
-* [Twitter summary card](https://dev.twitter.com/cards/overview) metadata
-
-While you could theoretically add the necessary metadata tags yourself, Jekyll SEO Tag provides a battle-tested template of crowdsourced best-practices.
-
-## What it doesn't do
-
-Jekyll SEO tag is designed to output machine-readable metadata for search engines and social networks to index and display. If you're looking for something to analyze your Jekyll site's structure and content (e.g., more traditional SEO optimization), take a look at [The Jekyll SEO Gem](https://github.com/pmarsceill/jekyll-seo-gem).
-
-Jekyll SEO tag isn't designed to accommodate every possible use case. It should work for most site out of the box and without a laundry list of configuration options that serve only to confuse most users.
-
-## Installation
-
-1. Add the following to your site's `Gemfile`:
-
-  ```ruby
-  gem 'jekyll-seo-tag'
-  ```
-
-2. Add the following to your site's `_config.yml`:
-
-  ```yml
-  gems:
-    - jekyll-seo-tag
-  ```
-
-3. Add the following right before `</head>` in your site's template(s):
-
-  ```liquid
-  {% seo %}
-  ```
-
-## Usage
-
-The SEO tag will respect any of the following if included in your site's `_config.yml` (and simply not include them if they're not defined):
-
-* `title` - Your site's title (e.g., Ben's awesome site, The GitHub Blog, etc.)
-* `description` - A short description (e.g., A blog dedicated to reviewing cat gifs)
-* `url` - The full URL to your site. Note: `site.github.url` will be used by default.
-* `author` - global author information (see below)
-* `twitter:username` - The site's Twitter handle. You'll want to describe it like so:
-
-  ```yml
-  twitter:
-    username: benbalter
-  ```
-
-* `facebook` - The following properties are available:
-  * `facebook:app_id` - a Facebook app ID for Facebook insights
-  * `facebook:publisher` - a Facebook page URL or ID of the publishing entity
-  * `facebook:admins` - a Facebook user ID for domain insights linked to a personal account
-
-  You'll want to describe one or more like so:
-
-   ```yml
-  facebook:
-    app_id: 1234
-    publisher: 1234
-    admins: 1234
-   ```
-
-* `logo` - URL to a site-wide logo (e.g., `/assets/your-company-logo.png`)
-* `social` - For [specifying social profiles](https://developers.google.com/structured-data/customize/social-profiles). The following properties are available:
-  * `name` - If the user or organization name differs from the site's name
-  * `links` - An array of links to social media profiles.
-* `google_site_verification` for verifying ownership via Google webmaster tools
-
-The SEO tag will respect the following YAML front matter if included in a post, page, or document:
-
-* `title` - The title of the post, page, or document
-* `description` - A short description of the page's content
-* `image` - URL to an image associated with the post, page, or document (e.g., `/assets/page-pic.jpg`)
-* `author` - Page-, post-, or document-specific author information (see below)
-
-## Advanced usage
-
-Jekyll SEO Tag is designed to implement SEO best practices by default and to be the right fit for most sites right out of the box. If for some reason, you need more control over the output, read on:
-
-### Disabling `<title>` output
-
-If for some reason, you don't want the plugin to output `<title>` tags on each page, simply invoke the plugin within your template like so:
-
-```
-{% seo title=false %}
-```
-
-### Author information
-
-Author information is used to propagate the `creator` field of Twitter summary cards. This should be an author-specific, not site-wide Twitter handle (the site-wide username be stored as `site.twitter.username`).
-
-*TL;DR: In most cases, put `author: [your Twitter handle]` in the document's front matter, for sites with multiple authors. If you need something more complicated, read on.*
-
-There are several ways to convey this author-specific information. Author information is found in the following order of priority:
-
-1. An `author` object, in the documents's front matter, e.g.:
-
-  ```yml
-  author:
-    twitter: benbalter
-  ```
-
-2. An `author` object, in the site's `_config.yml`, e.g.:
-
-  ```yml
-  author:
-    twitter: benbalter
-  ```
-
-3. `site.data.authors[author]`, if an author is specified in the document's front matter, and a corresponding key exists in `site.data.authors`. E.g., you have the following in the document's front matter:
-
-  ```yml
-  author: benbalter
-  ```
-
-  And you have the following in `_data/authors.yml`:
-
-  ```yml
-  benbalter:
-    picture: /img/benbalter.png
-    twitter: jekyllrb
-
-  potus:
-    picture: /img/potus.png
-    twitter: whitehouse
-  ```
-
-  In the above example, the author `benbalter`'s Twitter handle will be resolved to `@jekyllrb`. This allows you to centralize author information in a single `_data/authors` file for site with many authors that require more than just the author's username.
-
-  *Pro-tip: If `authors` is present in the document's front matter as an array (and `author` is not), the plugin will use the first author listed, as Twitter supports only one author.*
-
-4. An author in the document's front matter (the simplest way), e.g.:
-
-  ```yml
-  author: benbalter
-  ```
-
-5. An author in the site's `_config.yml`, e.g.:
-
-  ```yml
-  author: benbalter
-  ```
-
-### 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.
-
-* `seo`
-  * `name` - If the name of the thing that the page represents is different from the page title. (i.e.: "Frank's Café" vs "Welcome to Frank's Café")
-  * `type` - The type of things that the page represents. This must be a [Schema.org type](http://schema.org/docs/schemas.html), and will probably usually be something like [`BlogPosting`](http://schema.org/BlogPosting), [`NewsArticle`](http://schema.org/NewsArticle), [`Person`](http://schema.org/Person), [`Organization`](http://schema.org/Organization), etc.
-  * `links` - An array of other URLs that represent the same thing that this page represents. For instance, Jane's bio page might include links to Jane's GitHub and Twitter profiles.
-
-### Customizing image output
-
-For most users, setting `image: [path-to-image]` on a per-page basis should be enough. If you need more control over how images are represented, the `image` property can also be an object, with the following options:
-
-* `path` - The relative path to the image. Same as `image: [path-to-image]`
-* `twitter` - The relative path to a Twitter-specific image.
-* `facebook` - The relative path to a Facebook-specific image.
-* `height` - The height of the Facebook (`og:image`) image
-* `width` - The width of the Facebook (`og:image`) image
-
-You can use any of the above, optional properties, like so:
-
-```yml
-image:
-  twitter: /img/twitter.png
-  facebook: /img/facebook.png
-  height: 100
-  width: 100
-```
-
-### SmartyPants Titles
-
-Titles will be processed using [Jekyll's `smartify` filter](https://jekyllrb.com/docs/templates/). This will use SmartyPants to translate plain ASCII punctuation into "smart" typographic punctuation. This will not render or strip any Markdown you may be using in a page title.

Rakefile

@@ -1,6 +0,0 @@
-require 'bundler/gem_tasks'
-require 'rspec/core/rake_task'
-
-RSpec::Core::RakeTask.new(:spec)
-
-task default: :spec

docs/README.md

@@ -0,0 +1,33 @@
+## About Jekyll SEO Tag
+
+A Jekyll plugin to add metadata tags for search engines and social networks to better index and display your site's content.
+
+[![Gem Version](https://badge.fury.io/rb/jekyll-seo-tag.svg)](https://badge.fury.io/rb/jekyll-seo-tag) [![Build Status](https://travis-ci.org/jekyll/jekyll-seo-tag.svg)](https://travis-ci.org/jekyll/jekyll-seo-tag)
+
+## What it does
+
+Jekyll SEO Tag adds the following meta tags to your site:
+
+* Page title, with site title or description appended
+* Page description
+* Canonical URL
+* Next and previous URLs on paginated pages
+* [JSON-LD Site and post metadata](https://developers.google.com/search/docs/guides/intro-structured-data) for richer indexing
+* [Open Graph](https://ogp.me/) title, description, site title, and URL (for Facebook, LinkedIn, etc.)
+* [Twitter Summary Card](https://dev.twitter.com/cards/overview) metadata
+
+While you could theoretically add the necessary metadata tags yourself, Jekyll SEO Tag provides a battle-tested template of crowdsourced best-practices.
+
+## What it doesn't do
+
+Jekyll SEO tag is designed to output machine-readable metadata for search engines and social networks to index and display. If you're looking for something to analyze your Jekyll site's structure and content (e.g., more traditional SEO optimization), take a look at [The Jekyll SEO Gem](https://github.com/pmarsceill/jekyll-seo-gem).
+
+Jekyll SEO tag isn't designed to accommodate every possible use case. It should work for most site out of the box and without a laundry list of configuration options that serve only to confuse most users.
+
+## Documentation
+
+For more information, see:
+
+* [Installation](installation.md)
+* [Usage](usage.md)
+* [Advanced usage](advanced-usage.md)

docs/_config.yml

@@ -0,0 +1,10 @@
+title: Jekyll SEO Tag
+description: A Jekyll plugin to add metadata tags for search engines and social networks to better index and display your site's content.
+
+permalink: pretty
+
+plugins:
+  - jekyll-seo-tag
+  - jekyll-sitemap
+
+theme: jekyll-theme-primer

docs/_layouts/default.html

@@ -0,0 +1,18 @@
+<!DOCTYPE html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <link href="{{ '/assets/css/style.css?v=' | append: site.github.build_revision | relative_url }}" rel="stylesheet">
+    {% seo %}
+  </head>
+  <body>
+    <div class="container markdown-body">
+      <h1>{{ site.title }}</h1>
+      
+      {{ content }}
+    </div>
+    <script src="{{ "assets/javascript/anchor-js/anchor.min.js" | relative_url }}"></script>
+    <script>anchors.add();</script>
+  </body>
+</html>

docs/advanced-usage.md

@@ -0,0 +1,191 @@
+## Advanced usage
+
+Jekyll SEO Tag is designed to implement SEO best practices by default and to be the right fit for most sites right out of the box. If for some reason, you need more control over the output, read on:
+
+### Disabling `<title>` output
+
+If for some reason, you don't want the plugin to output `<title>` tags on each page, simply invoke the plugin within your template like so:
+
+<!-- {% raw %} -->
+```
+{% seo title=false %}
+```
+<!-- {% endraw %} -->
+
+### Author information
+
+Author information is used to propagate the `creator` field of Twitter summary cards. This should be an author-specific, not site-wide Twitter handle (the site-wide username be stored as `site.twitter.username`).
+
+*TL;DR: In most cases, put `author: [your Twitter handle]` in the document's front matter, for sites with multiple authors. If you need something more complicated, read on.*
+
+There are several ways to convey this author-specific information. Author information is found in the following order of priority:
+
+1. An `author` object, in the documents's front matter, e.g.:
+
+  ```yml
+  author:
+    twitter: benbalter
+  ```
+
+2. An `author` object, in the site's `_config.yml`, e.g.:
+
+  ```yml
+  author:
+    twitter: benbalter
+  ```
+
+3. `site.data.authors[author]`, if an author is specified in the document's front matter, and a corresponding key exists in `site.data.authors`. E.g., you have the following in the document's front matter:
+
+  ```yml
+  author: benbalter
+  ```
+
+  And you have the following in `_data/authors.yml`:
+
+  ```yml
+  benbalter:
+    picture: /img/benbalter.png
+    twitter: jekyllrb
+
+  potus:
+    picture: /img/potus.png
+    twitter: whitehouse
+  ```
+
+  In the above example, the author `benbalter`'s Twitter handle will be resolved to `@jekyllrb`. This allows you to centralize author information in a single `_data/authors` file for site with many authors that require more than just the author's username.
+
+  *Pro-tip: If `authors` is present in the document's front matter as an array (and `author` is not), the plugin will use the first author listed, as Twitter supports only one author.*
+
+4. An author in the document's front matter (the simplest way), e.g.:
+
+  ```yml
+  author: benbalter
+  ```
+
+5. An author in the site's `_config.yml`, e.g.:
+
+  ```yml
+  author: benbalter
+  ```
+
+#### Setting author url
+
+Starting from August 6, 2021 [Google recommends](https://developers.google.com/search/updates) to set the `author.url` property. This property helps Google to disambiguate the correct author of the article.
+
+You can set it the same way as the other author properties. For example, you can put it in an `author` object, in the site's `_config.yml`, e.g.:
+
+  ```yml
+  author:
+    name: My Name
+    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.
+
+* `seo`
+  * `name` - If the name of the thing that the page represents is different from the page title. (i.e.: "Frank's Café" vs "Welcome to Frank's Café")
+  * `type` - The type of things that the page represents. This must be a [Schema.org type](https://schema.org/docs/schemas.html), and will probably usually be something like [`BlogPosting`](https://schema.org/BlogPosting), [`NewsArticle`](https://schema.org/NewsArticle), [`Person`](https://schema.org/Person), [`Organization`](https://schema.org/Organization), etc.
+  * `links` - An array of other URLs that represent the same thing that this page represents. For instance, Jane's bio page might include links to Jane's GitHub and Twitter profiles.
+  * `date_modified` - Manually specify the `dateModified` field in the JSON-LD output to override Jekyll's own `dateModified`.
+  This field will take **first priority** for the `dateModified` JSON-LD output. This is useful when the file timestamp does not match the true time that the content was modified. A user may also install [Last Modified At](https://github.com/gjtorikian/jekyll-last-modified-at) which will offer an alternative way of providing for the `dateModified` field.
+
+### Customizing image output
+
+For most users, setting `image: [path-to-image]` on a per-page basis should be enough. If you need more control over how images are represented, the `image` property can also be an object, with the following options:
+
+* `path` - The relative path to the image. Same as `image: [path-to-image]`
+* `height` - The height of the Open Graph (`og:image`) image
+* `width` - The width of the Open Graph (`og:image`) image
+* `alt` - The alternative image text for Open Graph (`og:image:alt`) and Twitter (`twitter:image:alt`)
+
+You can use any of the above, optional properties, like so:
+
+```yml
+image:
+  path: /img/twitter.png
+  height: 100
+  width: 100
+  alt: Twitter Logo
+```
+
+### Setting a default image
+
+You can define a default image using [Front Matter defaults](https://jekyllrb.com/docs/configuration/front-matter-defaults/), to provide a default Twitter Card or OGP image to all of your posts and pages.
+
+Here is a very basic example, that you are encouraged to adapt to your needs:
+
+```yml
+defaults:
+  - scope:
+      path: ""
+    values:
+      image: /assets/images/default-card.png
+```
+
+### SmartyPants Titles
+
+Titles will be processed using [Jekyll's `smartify` filter](https://jekyllrb.com/docs/liquid/filters/). This will use SmartyPants to translate plain ASCII punctuation into "smart" typographic punctuation. This will not render or strip any Markdown you may be using in a page title.
+
+### Setting customized Canonical URL
+
+You can set custom Canonical URL for a page by specifying canonical_url option in page front matter.
+E.g., you have the following in the page's front matter:
+```yml
+layout: post
+title: Title of Your Post
+canonical_url: 'https://github.com/jekyll/jekyll-seo-tag/'
+```
+
+Which will generate canonical_url with specified link in canonical_url.
+```html
+<link rel="canonical" href="https://github.com/jekyll/jekyll-seo-tag/" />
+```
+
+If no canonical_url option was specified, then uses page url for generating canonical_url.
+E.g., you have not specified canonical_url in front matter:
+```yml
+layout: post
+title: Title of Your Post
+```
+
+Which will generate following canonical_url:
+```html
+<link rel="canonical" href="https://example.com/title-of-your-post" />
+```
+
+### Customizing title modifier for paginated pages
+
+You can override the default title modifier for paginated pages from `Page %{current} of %{total} for ` to a string of your
+choice by setting a `seo_paginator_message` key in your `_config.yml`.
+
+For example:
+
+```yml
+seo_paginator_message: "%<current>s / %<total>s | "
+```
+
+While the value can be any string text, we recommend using a Ruby string-template containing the variables `current` and `total`
+similar to the example above, to incorporate the current page-number and total number of paginated pages in the title.

docs/installation.md

@@ -0,0 +1,24 @@
+# Installing Jekyll SEO Tag
+
+1. Add the following to your site's `Gemfile`:
+
+  ```ruby
+  gem 'jekyll-seo-tag'
+  ```
+
+2. Add the following to your site's `_config.yml`:
+
+  ```yml
+  plugins:
+    - jekyll-seo-tag
+  ```
+
+If you are using a Jekyll version less than `3.5.0`, use the `gems` key instead of `plugins`.
+
+3. Add the following right before `</head>` in your site's template(s):
+
+<!-- {% raw %} -->
+  ```liquid
+  {% seo %}
+  ```
+<!-- {% endraw %} -->

docs/usage.md

@@ -0,0 +1,81 @@
+## Usage
+
+The SEO tag will respect any of the following if included in your site's `_config.yml` (and simply not include them if
+they're not defined):
+
+* `title` - Your site's title (e.g., *Ben's Awesome Site*, *The GitHub Blog*, etc.), used as part of the title tag like
+`Home | Ben's Awesome Site`.
+* `tagline` - A short description (e.g., *A blog dedicated to reviewing cat gifs*), used as part of the title tag like
+`Ben's Awesome Site | A blog dedicated to reviewing cat gifs` instead of `Ben's Awesome Site | Long description about a
+blog dedicated to reviewing cat gifs` that would be used when `page.title` is not defined.
+* `description` - A longer description used for the description meta tag. Also used as fallback for pages that don't
+provide their own `description`, and also as part of the page's title tag if neither `page.title` nor `site.tagline`
+has been defined.
+* `url` - The full URL to your site. Note: `site.github.url` will be used by default.
+* `author` - global author information (see [Advanced usage](advanced-usage.md#author-information))
+* `twitter` - The following properties are available:
+  * `twitter:card` - The site's default card type
+  * `twitter:username` - The site's Twitter handle.
+
+  You'll want to describe them like so:
+
+  ```yml
+  twitter:
+    username: benbalter
+    card: summary
+  ```
+* `facebook` - The following properties are available:
+  * `facebook:app_id` - a Facebook app ID for Facebook insights
+  * `facebook:publisher` - a Facebook page URL or ID of the publishing entity
+  * `facebook:admins` - a Facebook user ID for domain insights linked to a personal account
+
+  You'll want to describe one or more like so:
+
+  ```yml
+  facebook:
+    app_id: 1234
+    publisher: 1234
+    admins: 1234
+  ```
+* `logo` - URL to a site-wide logo (e.g., `/assets/your-company-logo.png`) - If you would like the "publisher" property
+to be present, you must add this field to your site's configuration, during the validation of the structured data by
+Google Search Console, if the `logo` field is not validated, you will find errors inherent to the publisher in the
+[Rich Results Testing Tool](https://search.google.com/test/rich-results)
+* `social` - For [specifying social profiles](https://developers.google.com/search/docs/guides/enhance-site#add-your-sites-name-logo-and-social-links).
+The following properties are available:
+  * `name` - If the user or organization name differs from the site's name
+  * `links` - An array of links to social media profiles.
+
+  ```yml
+  social:
+    name: Ben Balter
+    links:
+      - https://twitter.com/BenBalter
+      - https://www.facebook.com/ben.balter
+      - https://www.linkedin.com/in/BenBalter
+      - https://github.com/benbalter
+      - https://keybase.io/benbalter
+  ```
+* `google_site_verification` for verifying ownership via Google Search Console
+* Alternatively, verify ownership with several services at once using the following format:
+  ```yml
+  webmaster_verifications:
+    google: 1234
+    bing: 1234
+    alexa: 1234
+    yandex: 1234
+    baidu: 1234
+    facebook: 1234
+  ```
+* `locale` - The locale these tags are marked up in. Of the format `language_TERRITORY`. Default is `en_US`. Takes priority
+over existing config key `lang`.
+
+The SEO tag will respect the following YAML front matter if included in a post, page, or document:
+
+* `title` - The title of the post, page, or document
+* `description` - A short description of the page's content
+* `image` - URL to an image associated with the post, page, or document (e.g., `/assets/page-pic.jpg`)
+* `author` - Page-, post-, or document-specific author information (see [Advanced usage](advanced-usage.md#author-information))
+* `locale` - Page-, post-, or document-specific locale information. Takes priority over existing front matter attribute `lang`.
+
+*Note:* Front matter defaults can be used for any of the above values as described in advanced usage with an image example.

jekyll-seo-tag.gemspec

@@ -1,34 +1,34 @@
-# coding: utf-8
-lib = File.expand_path('../lib', __FILE__)
-$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
-require 'jekyll-seo-tag/version'
+# frozen_string_literal: true
+
+require_relative "lib/jekyll-seo-tag/version"
 
 Gem::Specification.new do |spec|
-  spec.name          = 'jekyll-seo-tag'
+  spec.name          = "jekyll-seo-tag"
   spec.version       = Jekyll::SeoTag::VERSION
-  spec.authors       = ['Ben Balter']
-  spec.email         = ['ben.balter@github.com']
+  spec.authors       = ["Ben Balter"]
+  spec.email         = ["ben.balter@github.com"]
   spec.summary       = "A Jekyll plugin to add metadata tags for search engines and social networks to better index and display your site's content."
-  spec.homepage      = 'https://github.com/benbalter/jekyll-seo-tag'
-  spec.license       = 'MIT'
+  spec.homepage      = "https://github.com/jekyll/jekyll-seo-tag"
+  spec.license       = "MIT"
 
   # Prevent pushing this gem to RubyGems.org by setting 'allowed_push_host', or
   # delete this section to allow pushing this gem to any host.
   if spec.respond_to?(:metadata)
-    spec.metadata['allowed_push_host'] = 'https://rubygems.org'
+    spec.metadata["allowed_push_host"] = "https://rubygems.org"
   else
-    raise 'RubyGems 2.0 or newer is required to protect against public gem pushes.'
+    raise "RubyGems 2.0 or newer is required to protect against public gem pushes."
   end
 
-  spec.files         = `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
-  spec.bindir        = 'exe'
-  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
-  spec.require_paths = ['lib']
+  spec.required_ruby_version = ">= 2.5.0"
 
-  spec.add_dependency 'jekyll', '~> 3.3'
-  spec.add_development_dependency 'bundler', '~> 1.10'
-  spec.add_development_dependency 'rake', '~> 10.0'
-  spec.add_development_dependency 'rspec', '~> 3.3'
-  spec.add_development_dependency 'html-proofer', '~> 2.5'
-  spec.add_development_dependency 'rubocop', '~> 0.37'
+  spec.files         = `git ls-files -z`.split("\x0").reject { |f| f.match(%r!^(test|spec|features)/!) }
+  spec.bindir        = "exe"
+  spec.executables   = spec.files.grep(%r!^exe/!) { |f| File.basename(f) }
+  spec.require_paths = ["lib"]
+
+  spec.add_dependency "jekyll", ">= 3.8", "< 5.0"
+  spec.add_development_dependency "bundler", ">= 1.15"
+  spec.add_development_dependency "html-proofer", "~> 3.7"
+  spec.add_development_dependency "rspec", "~> 3.5"
+  spec.add_development_dependency "rubocop-jekyll", "~> 0.12.0"
 end

lib/jekyll-seo-tag.rb

@@ -1,7 +1,18 @@
+# frozen_string_literal: true
+
+require "jekyll"
 require "jekyll-seo-tag/version"
 
 module Jekyll
   class SeoTag < Liquid::Tag
+    autoload :JSONLD,     "jekyll-seo-tag/json_ld"
+    autoload :AuthorDrop, "jekyll-seo-tag/author_drop"
+    autoload :ImageDrop,  "jekyll-seo-tag/image_drop"
+    autoload :JSONLDDrop, "jekyll-seo-tag/json_ld_drop"
+    autoload :UrlHelper,  "jekyll-seo-tag/url_helper"
+    autoload :Drop,       "jekyll-seo-tag/drop"
+    autoload :Filters,    "jekyll-seo-tag/filters"
+
     attr_accessor :context
 
     # Matches all whitespace that follows either
@@ -13,7 +24,7 @@ module Jekyll
     # We will not strip any whitespace if the next character is a '-'
     #   so that we do not interfere with the HTML comment at the
     #   very begining
-    MINIFY_REGEX = %r!(?<=[{}]|[>,]\n)\s+(?\!-)!
+    MINIFY_REGEX = %r!(?<=[{}]|[>,]\n)\s+(?\!-)!.freeze
 
     def initialize(_tag_name, text, _tokens)
       super
@@ -22,7 +33,7 @@ module Jekyll
 
     def render(context)
       @context = context
-      template.render!(payload, info)
+      SeoTag.template.render!(payload, info)
     end
 
     private
@@ -30,43 +41,51 @@ module Jekyll
     def options
       {
         "version" => Jekyll::SeoTag::VERSION,
-        "title"   => title?
+        "title"   => title?,
       }
     end
 
     def payload
-      {
-        "page"      => context.registers[:page],
-        "site"      => context.registers[:site].site_payload["site"],
-        "paginator" => context["paginator"],
-        "seo_tag"   => options
-      }
+      # site_payload is an instance of UnifiedPayloadDrop. See https://github.com/jekyll/jekyll/blob/22f2724a1f117a94cc16d18c499a93d5915ede4f/lib/jekyll/site.rb#L261-L276
+      context.registers[:site].site_payload.tap do |site_payload|
+        site_payload["page"]      = context.registers[:page]
+        site_payload["paginator"] = context["paginator"]
+        site_payload["seo_tag"]   = drop
+      end
     end
 
-    def title?
-      !(@text =~ %r!title=false!i)
+    def drop
+      if context.registers[:site].liquid_renderer.respond_to?(:cache)
+        Jekyll::SeoTag::Drop.new(@text, @context)
+      else
+        @drop ||= Jekyll::SeoTag::Drop.new(@text, @context)
+      end
     end
 
     def info
       {
         :registers => context.registers,
-        :filters   => [Jekyll::Filters]
+        :filters   => [Jekyll::Filters],
       }
     end
 
-    def template
-      @template ||= Liquid::Template.parse template_contents
-    end
-
-    def template_contents
-      @template_contents ||= begin
-        File.read(template_path).gsub(MINIFY_REGEX, "")
+    class << self
+      def template
+        @template ||= Liquid::Template.parse template_contents
       end
-    end
 
-    def template_path
-      @template_path ||= begin
-        File.expand_path "./template.html", File.dirname(__FILE__)
+      private
+
+      def template_contents
+        @template_contents ||= begin
+          File.read(template_path).gsub(MINIFY_REGEX, "")
+        end
+      end
+
+      def template_path
+        @template_path ||= begin
+          File.expand_path "./template.html", File.dirname(__FILE__)
+        end
       end
     end
   end

lib/jekyll-seo-tag/author_drop.rb

@@ -0,0 +1,93 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    # A drop representing the current page's author
+    #
+    # Author name will be pulled from:
+    #
+    # 1. The page's `author` key
+    # 2. The first author in the page's `authors` key
+    # 3. The `author` key in the site config
+    #
+    # If the result from the name search is a string, we'll also check
+    # for additional author metadata in `site.data.authors`
+    class AuthorDrop < Jekyll::Drops::Drop
+      # Initialize a new AuthorDrop
+      #
+      # page - The page hash (e.g., Page#to_liquid)
+      # site - The Jekyll::Drops::SiteDrop
+      def initialize(page: nil, site: nil)
+        raise ArgumentError unless page && site
+
+        @mutations = {}
+        @page = page
+        @site = site
+      end
+
+      # AuthorDrop#to_s should return name, allowing the author drop to safely
+      # replace `page.author`, if necessary, and remain backwards compatible
+      def name
+        author_hash["name"]
+      end
+      alias_method :to_s, :name
+
+      def twitter
+        return @twitter if defined? @twitter
+
+        twitter = author_hash["twitter"] || author_hash["name"]
+        @twitter = twitter.is_a?(String) ? twitter.sub(%r!^@!, "") : nil
+      end
+
+      private
+
+      attr_reader :site, :page
+
+      # Finds the page author in the page.author, page.authors, or site.author
+      #
+      # Returns a string or hash representing the author
+      def resolved_author
+        return @resolved_author if defined? @resolved_author
+
+        sources = [page["author"]]
+        sources << page["authors"].first if page["authors"].is_a?(Array)
+        sources << site["author"]
+        @resolved_author = sources.find { |s| !s.to_s.empty? }
+      end
+
+      # If resolved_author is a string, attempts to find coresponding author
+      # metadata in `site.data.authors`
+      #
+      # Returns a hash representing additional metadata or an empty hash
+      def site_data_hash
+        @site_data_hash ||= begin
+          return {} unless resolved_author.is_a?(String)
+          return {} unless site.data["authors"].is_a?(Hash)
+
+          author_hash = site.data["authors"][resolved_author]
+          author_hash.is_a?(Hash) ? author_hash : {}
+        end
+      end
+
+      # Returns the normalized author hash representing the page author,
+      # including site-wide metadata if the author is provided as a string,
+      # or an empty hash, if the author cannot be resolved
+      def author_hash
+        @author_hash ||= begin
+          case resolved_author
+          when Hash
+            resolved_author
+          when String
+            { "name" => resolved_author }.merge!(site_data_hash)
+          else
+            {}
+          end
+        end
+      end
+
+      # Since author_hash is aliased to fallback_data, any values in the hash
+      # will be exposed via the drop, allowing support for arbitrary metadata
+      alias_method :fallback_data, :author_hash
+    end
+  end
+end

lib/jekyll-seo-tag/drop.rb

@@ -0,0 +1,259 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    class Drop < Jekyll::Drops::Drop
+      include Jekyll::SeoTag::UrlHelper
+
+      TITLE_SEPARATOR = " | "
+      FORMAT_STRING_METHODS = [
+        :markdownify, :strip_html, :normalize_whitespace, :escape_once,
+      ].freeze
+      HOMEPAGE_OR_ABOUT_REGEX = %r!^/(about/)?(index.html?)?$!.freeze
+
+      EMPTY_READ_ONLY_HASH = {}.freeze
+      private_constant :EMPTY_READ_ONLY_HASH
+
+      def initialize(text, context)
+        @obj = EMPTY_READ_ONLY_HASH
+        @mutations = {}
+        @text = text
+        @context = context
+      end
+
+      def version
+        Jekyll::SeoTag::VERSION
+      end
+
+      # Should the `<title>` tag be generated for this page?
+      def title?
+        return false unless title
+        return @display_title if defined?(@display_title)
+
+        @display_title = (@text !~ %r!title=false!i)
+      end
+
+      def site_title
+        @site_title ||= format_string(site["title"] || site["name"])
+      end
+
+      def site_tagline
+        @site_tagline ||= format_string site["tagline"]
+      end
+
+      def site_description
+        @site_description ||= format_string site["description"]
+      end
+
+      # Page title without site title or description appended
+      def page_title
+        @page_title ||= format_string(page["title"]) || site_title
+      end
+
+      def site_tagline_or_description
+        site_tagline || site_description
+      end
+
+      # Page title with site title or description appended
+      # rubocop:disable Metrics/CyclomaticComplexity
+      def title
+        @title ||= begin
+          if site_title && page_title != site_title
+            page_title + TITLE_SEPARATOR + site_title
+          elsif site_description && site_title
+            site_title + TITLE_SEPARATOR + site_tagline_or_description
+          else
+            page_title || site_title
+          end
+        end
+
+        return page_number + @title if page_number
+
+        @title
+      end
+      # rubocop:enable Metrics/CyclomaticComplexity
+
+      def name
+        return @name if defined?(@name)
+
+        @name = if seo_name
+                  seo_name
+                elsif !homepage_or_about?
+                  nil
+                elsif site_social["name"]
+                  format_string site_social["name"]
+                elsif site_title
+                  site_title
+                end
+      end
+
+      def description
+        @description ||= begin
+          value = format_string(page["description"] || page["excerpt"]) || site_description
+          snippet(value, description_max_words)
+        end
+      end
+
+      # A drop representing the page author
+      def author
+        @author ||= AuthorDrop.new(:page => page, :site => site)
+      end
+
+      # A drop representing the JSON-LD output
+      def json_ld
+        @json_ld ||= JSONLDDrop.new(self)
+      end
+
+      # Returns a Drop representing the page's image
+      # Returns nil if the image has no path, to preserve backwards compatability
+      def image
+        @image ||= ImageDrop.new(:page => page, :context => @context)
+        @image if @image.path
+      end
+
+      def date_modified
+        @date_modified ||= begin
+          date = page_seo["date_modified"] || page["last_modified_at"].to_liquid || page["date"]
+          filters.date_to_xmlschema(date) if date
+        end
+      end
+
+      def date_published
+        @date_published ||= filters.date_to_xmlschema(page["date"]) if page["date"]
+      end
+
+      def type
+        @type ||= begin
+          if page_seo["type"]
+            page_seo["type"]
+          elsif homepage_or_about?
+            "WebSite"
+          elsif page["date"]
+            "BlogPosting"
+          else
+            "WebPage"
+          end
+        end
+      end
+
+      def links
+        @links ||= begin
+          if page_seo["links"]
+            page_seo["links"]
+          elsif homepage_or_about? && site_social["links"]
+            site_social["links"]
+          end
+        end
+      end
+
+      def logo
+        @logo ||= begin
+          return unless site["logo"]
+
+          if absolute_url? site["logo"]
+            filters.uri_escape site["logo"]
+          else
+            filters.uri_escape filters.absolute_url site["logo"]
+          end
+        end
+      end
+
+      def page_lang
+        @page_lang ||= page["lang"] || site["lang"] || "en_US"
+      end
+
+      def page_locale
+        @page_locale ||= (page["locale"] || site["locale"] || page_lang).tr("-", "_")
+      end
+
+      def canonical_url
+        @canonical_url ||= begin
+          if page["canonical_url"].to_s.empty?
+            filters.absolute_url(page["url"]).to_s.gsub(%r!/index\.html$!, "/")
+          else
+            page["canonical_url"]
+          end
+        end
+      end
+
+      def description_max_words
+        @description_max_words ||= page["seo_description_max_words"] || 100
+      end
+
+      private
+
+      def filters
+        @filters ||= Jekyll::SeoTag::Filters.new(@context)
+      end
+
+      def page
+        @page ||= @context.registers[:page].to_liquid
+      end
+
+      def site
+        @site ||= @context.registers[:site].site_payload["site"].to_liquid
+      end
+
+      def homepage_or_about?
+        page["url"] =~ HOMEPAGE_OR_ABOUT_REGEX
+      end
+
+      def page_number
+        return unless @context["paginator"] && @context["paginator"]["page"]
+
+        current = @context["paginator"]["page"]
+        total = @context["paginator"]["total_pages"]
+        paginator_message = site["seo_paginator_message"] || "Page %<current>s of %<total>s for "
+
+        format(paginator_message, :current => current, :total => total) if current > 1
+      end
+
+      attr_reader :context
+
+      def fallback_data
+        @fallback_data ||= {}
+      end
+
+      def format_string(string)
+        string = FORMAT_STRING_METHODS.reduce(string) do |memo, method|
+          filters.public_send(method, memo)
+        end
+
+        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
+
+      def page_seo
+        @page_seo ||= sub_hash(page, "seo")
+      end
+
+      def site_social
+        @site_social ||= sub_hash(site, "social")
+      end
+
+      # Safely returns a sub hash
+      #
+      # hash - the parent hash
+      # key  - the key in the parent hash
+      #
+      # Returns the sub hash or an empty hash, if it does not exist
+      def sub_hash(hash, key)
+        if hash[key].is_a?(Hash)
+          hash[key]
+        else
+          EMPTY_READ_ONLY_HASH
+        end
+      end
+    end
+  end
+end

lib/jekyll-seo-tag/filters.rb

@@ -0,0 +1,14 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    class Filters
+      include Jekyll::Filters
+      include Liquid::StandardFilters
+
+      def initialize(context)
+        @context = context
+      end
+    end
+  end
+end

lib/jekyll-seo-tag/image_drop.rb

@@ -0,0 +1,83 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    # A drop representing the page image
+    # The image path will be pulled from:
+    #
+    # 1. The `image` key if it's a string
+    # 2. The `image.path` key if it's a hash
+    # 3. The `image.facebook` key
+    # 4. The `image.twitter` key
+    class ImageDrop < Jekyll::Drops::Drop
+      include Jekyll::SeoTag::UrlHelper
+
+      # Initialize a new ImageDrop
+      #
+      # page - The page hash (e.g., Page#to_liquid)
+      # context - the Liquid::Context
+      def initialize(page: nil, context: nil)
+        raise ArgumentError unless page && context
+
+        @mutations = {}
+        @page = page
+        @context = context
+      end
+
+      # Called path for backwards compatability, this is really
+      # the escaped, absolute URL representing the page's image
+      # Returns nil if no image path can be determined
+      def path
+        @path ||= filters.uri_escape(absolute_url) if absolute_url
+      end
+      alias_method :to_s, :path
+
+      private
+
+      attr_accessor :page, :context
+
+      # The normalized image hash with a `path` key (which may be nil)
+      def image_hash
+        @image_hash ||= begin
+          image_meta = page["image"]
+
+          case image_meta
+          when Hash
+            { "path" => nil }.merge!(image_meta)
+          when String
+            { "path" => image_meta }
+          else
+            { "path" => nil }
+          end
+        end
+      end
+      alias_method :fallback_data, :image_hash
+
+      def raw_path
+        @raw_path ||= begin
+          image_hash["path"] || image_hash["facebook"] || image_hash["twitter"]
+        end
+      end
+
+      def absolute_url
+        return unless raw_path
+
+        @absolute_url ||= build_absolute_path
+      end
+
+      def build_absolute_path
+        return raw_path unless raw_path.is_a?(String) && absolute_url?(raw_path) == false
+        return filters.absolute_url(raw_path) if raw_path.start_with?("/")
+
+        page_dir = @page["url"]
+        page_dir = File.dirname(page_dir) unless page_dir.end_with?("/")
+
+        filters.absolute_url File.join(page_dir, raw_path)
+      end
+
+      def filters
+        @filters ||= Jekyll::SeoTag::Filters.new(context)
+      end
+    end
+  end
+end

lib/jekyll-seo-tag/json_ld.rb

@@ -0,0 +1,31 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    # This module is deprecated, but is included in the Gem to avoid a breaking
+    # change and should be removed at the next major version bump
+    module JSONLD
+      METHODS_KEYS = {
+        :json_context   => "@context",
+        :type           => "@type",
+        :name           => "name",
+        :page_title     => "headline",
+        :json_author    => "author",
+        :json_image     => "image",
+        :date_published => "datePublished",
+        :date_modified  => "dateModified",
+        :description    => "description",
+        :publisher      => "publisher",
+        :main_entity    => "mainEntityOfPage",
+        :links          => "sameAs",
+        :canonical_url  => "url",
+      }.freeze
+
+      # Self should be a Jekyll::SeoTag::Drop instance (when extending the module)
+      def json_ld
+        Jekyll.logger.warn "Jekyll::SeoTag::JSONLD is deprecated"
+        @json_ld ||= JSONLDDrop.new(self)
+      end
+    end
+  end
+end

lib/jekyll-seo-tag/json_ld_drop.rb

@@ -0,0 +1,104 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    class JSONLDDrop < Jekyll::Drops::Drop
+      extend Forwardable
+
+      def_delegator :page_drop, :name,           :name
+      def_delegator :page_drop, :description,    :description
+      def_delegator :page_drop, :canonical_url,  :url
+      def_delegator :page_drop, :page_title,     :headline
+      def_delegator :page_drop, :date_modified,  :dateModified
+      def_delegator :page_drop, :date_published, :datePublished
+      def_delegator :page_drop, :links,          :sameAs
+      def_delegator :page_drop, :logo,           :logo
+      def_delegator :page_drop, :type,           :type
+
+      # Expose #type and #logo as private methods and #@type as a public method
+      alias_method :@type, :type
+      private :type, :logo
+
+      VALID_ENTITY_TYPES = %w(BlogPosting CreativeWork).freeze
+      VALID_AUTHOR_TYPES = %w(Organization Person).freeze
+      private_constant :VALID_ENTITY_TYPES, :VALID_AUTHOR_TYPES
+
+      # page_drop should be an instance of Jekyll::SeoTag::Drop
+      def initialize(page_drop)
+        @mutations = {}
+        @page_drop = page_drop
+      end
+
+      def fallback_data
+        @fallback_data ||= {
+          "@context" => "https://schema.org",
+        }
+      end
+
+      def author
+        return unless page_drop.author["name"]
+
+        author_type = page_drop.author["type"]
+        return if author_type && !VALID_AUTHOR_TYPES.include?(author_type)
+
+        hash = {
+          "@type" => author_type || "Person",
+          "name"  => page_drop.author["name"],
+        }
+
+        author_url = page_drop.author["url"]
+        hash["url"] = author_url if author_url
+
+        hash
+      end
+
+      def image
+        return unless page_drop.image
+        return page_drop.image.path if page_drop.image.keys.length == 1
+
+        hash = page_drop.image.to_h
+        hash["url"]   = hash.delete("path")
+        hash["@type"] = "imageObject"
+        hash
+      end
+
+      def publisher
+        return unless logo
+
+        output = {
+          "@type" => "Organization",
+          "logo"  => {
+            "@type" => "ImageObject",
+            "url"   => logo,
+          },
+        }
+        output["name"] = page_drop.author.name if page_drop.author.name
+        output
+      end
+
+      def main_entity
+        return unless VALID_ENTITY_TYPES.include?(type)
+
+        {
+          "@type" => "WebPage",
+          "@id"   => page_drop.canonical_url,
+        }
+      end
+      alias_method :mainEntityOfPage, :main_entity
+      private :main_entity
+
+      # Returns a JSON-encoded object containing the JSON-LD data.
+      # Keys are sorted.
+      def to_json(state = nil)
+        keys.sort.each_with_object({}) do |(key, _), result|
+          v = self[key]
+          result[key] = v unless v.nil?
+        end.to_json(state)
+      end
+
+      private
+
+      attr_reader :page_drop
+    end
+  end
+end

lib/jekyll-seo-tag/url_helper.rb

@@ -0,0 +1,23 @@
+# frozen_string_literal: true
+
+module Jekyll
+  class SeoTag
+    # Mixin to share common URL-related methods between class
+    module UrlHelper
+      private
+
+      # Determines if the given string is an absolute URL
+      #
+      # Returns true if an absolute URL
+      # Returns false if it's a relative URL
+      # Returns nil if it is not a string or can't be parsed as a URL
+      def absolute_url?(string)
+        return unless string
+
+        Addressable::URI.parse(string).absolute?
+      rescue Addressable::URI::InvalidURIError
+        nil
+      end
+    end
+  end
+end

lib/jekyll-seo-tag/version.rb

@@ -1,8 +1,10 @@
+# frozen_string_literal: true
+
 # Prevent bundler errors
 module Liquid; class Tag; end; end
 
 module Jekyll
   class SeoTag < Liquid::Tag
-    VERSION = "2.0.0".freeze
+    VERSION = "2.8.0"
   end
 end

lib/template.html

@@ -1,160 +1,82 @@
 <!-- Begin Jekyll SEO tag v{{ seo_tag.version }} -->
-
-{% if page.url == "/" or page.url == "/about/" %}
-  {% assign seo_homepage_or_about = true %}
+{% if seo_tag.title? %}
+  <title>{{ seo_tag.title }}</title>
 {% endif %}
 
-{% assign seo_site_title = site.title | default: site.name %}
+<meta name="generator" content="Jekyll v{{ jekyll.version }}" />
 
-{% if page.title %}
-  {% assign seo_title = page.title %}
-  {% assign seo_page_title = page.title %}
+{% if seo_tag.page_title %}
+  <meta property="og:title" content="{{ seo_tag.page_title }}" />
+{% endif %}
 
-  {% if seo_site_title %}
-    {% assign seo_title = seo_title | append:" - " | append: seo_site_title %}
+{% if seo_tag.author.name %}
+  <meta name="author" content="{{ seo_tag.author.name }}" />
+{% endif %}
+
+<meta property="og:locale" content="{{ seo_tag.page_locale }}" />
+
+{% if seo_tag.description %}
+  <meta name="description" content="{{ seo_tag.description }}" />
+  <meta property="og:description" content="{{ seo_tag.description }}" />
+  <meta property="twitter:description" content="{{ seo_tag.description }}" />
+{% endif %}
+
+{% if site.url %}
+  <link rel="canonical" href="{{ seo_tag.canonical_url }}" />
+  <meta property="og:url" content="{{ seo_tag.canonical_url }}" />
+{% endif %}
+
+{% if seo_tag.site_title %}
+  <meta property="og:site_name" content="{{ seo_tag.site_title }}" />
+{% endif %}
+
+{% if seo_tag.image %}
+  <meta property="og:image" content="{{ seo_tag.image.path }}" />
+  {% if seo_tag.image.height %}
+    <meta property="og:image:height" content="{{ seo_tag.image.height }}" />
   {% endif %}
-{% elsif seo_site_title %}
-  {% assign seo_title = seo_site_title %}
-  {% assign seo_page_title = seo_site_title %}
-
-  {% if site.description %}
-    {% assign seo_title = seo_title | append:" - " | append: site.description %}
+  {% if seo_tag.image.width %}
+    <meta property="og:image:width" content="{{ seo_tag.image.width }}" />
   {% endif %}
-{% endif %}
-
-{% if page.seo and page.seo.name %}
-    {% assign seo_name = page.seo.name %}
-{% elsif seo_homepage_or_about and site.social and site.social.name %}
-    {% assign seo_name = site.social.name %}
-{% elsif seo_homepage_or_about and seo_site_title %}
-    {% assign seo_name = seo_site_title %}
-{% endif %}
-{% if seo_name %}
-  {% assign seo_name = seo_name | smartify | strip_html | strip_newlines | escape_once %}
-{% endif %}
-
-{% if seo_title %}
-  {% assign seo_title = seo_title | smartify | strip_html | strip_newlines | escape_once %}
-{% endif %}
-
-{% if seo_site_title %}
-  {% assign seo_site_title = seo_site_title | smartify | strip_html | strip_newlines | escape_once %}
-{% endif %}
-
-{% if seo_page_title %}
-  {% assign seo_page_title = seo_page_title | smartify | strip_html | strip_newlines | escape_once %}
-{% endif %}
-
-{% assign seo_description = page.description | default: page.excerpt | default: site.description %}
-{% if seo_description %}
-  {% assign seo_description = seo_description | markdownify | strip_html | strip_newlines | escape_once %}
-{% endif %}
-
-{% assign seo_author = page.author | default: page.authors[0] | default: site.author %}
-{% if seo_author %}
-  {% if seo_author.twitter %}
-    {% assign seo_author_twitter = seo_author.twitter %}
-  {% else %}
-    {% if site.data.authors and site.data.authors[seo_author] %}
-      {% assign seo_author_twitter = site.data.authors[seo_author].twitter  %}
-    {% else %}
-      {% assign seo_author_twitter = seo_author  %}
-    {% endif %}
+  {% if seo_tag.image.alt %}
+    <meta property="og:image:alt" content="{{ seo_tag.image.alt }}" />
   {% endif %}
-  {% assign seo_author_twitter = seo_author_twitter | replace:"@","" %}
-{% endif %}
-
-{% if page.seo and page.seo.type %}
-    {% assign seo_type = page.seo.type %}
-{% elsif seo_homepage_or_about %}
-    {% assign seo_type = "WebSite" %}
-{% elsif page.date %}
-    {% assign seo_type = "BlogPosting" %}
-{% else %}
-    {% assign seo_type = "WebPage" %}
-{% endif %}
-
-{% if page.seo and page.seo.links %}
-    {% assign seo_links = page.seo.links %}
-{% elsif seo_homepage_or_about and site.social and site.social.links %}
-    {% assign seo_links = site.social.links %}
-{% endif %}
-
-{% if site.logo %}
-  {% assign seo_site_logo = site.logo %}
-  {% unless seo_site_logo contains "://" %}
-    {% assign seo_site_logo = seo_site_logo | absolute_url %}
-  {% endunless %}
-  {% assign seo_site_logo = seo_site_logo | escape %}
-{% endif %}
-
-{% if page.image %}
-  {% assign seo_page_image = page.image.path | default: page.image.facebook | default: page.image %}
-  {% unless seo_page_image contains "://" %}
-    {% assign seo_page_image = seo_page_image | absolute_url %}
-  {% endunless %}
-  {% assign seo_page_image = seo_page_image | escape %}
-{% endif %}
-
-{% if seo_tag.title and seo_title %}
-  <title>{{ seo_title }}</title>
-{% endif %}
-
-{% if seo_page_title %}
-  <meta property="og:title" content="{{ seo_page_title }}" />
-{% endif %}
-
-{% if seo_description %}
-  <meta name="description" content="{{ seo_description }}" />
-  <meta property="og:description" content="{{ seo_description }}" />
-{% endif %}
-
-{% if page.url %}
-  <link rel="canonical" href="{{ page.url | replace:'/index.html','/' | absolute_url }}" />
-  <meta property="og:url" content="{{ page.url | replace:'/index.html','/' | absolute_url }}" />
-{% endif %}
-
-{% if seo_site_title %}
-  <meta property="og:site_name" content="{{ seo_site_title }}" />
-{% endif %}
-
-{% if seo_page_image %}
-<meta property="og:image" content="{{ seo_page_image }}" />
-  {% if page.image.height %}
-    <meta property="og:image:height" content="{{ page.image.height }}" />
-  {% endif %}
-  {% if page.image.width %}
-    <meta property="og:image:width" content="{{ page.image.width }}" />
-  {% endif %}
-{% endif %}
-
-{% if page.image.twitter %}
-  <meta name="twitter:image" content="{{ page.image.twitter | absolute_url }}" />
 {% endif %}
 
 {% if page.date %}
   <meta property="og:type" content="article" />
   <meta property="article:published_time" content="{{ page.date | date_to_xmlschema }}" />
+{% else %}
+  <meta property="og:type" content="website" />
 {% endif %}
 
 {% if paginator.previous_page %}
-  <link rel="prev" href="{{ paginator.previous_page_path | absolute_url }}">
+  <link rel="prev" href="{{ paginator.previous_page_path | absolute_url }}" />
 {% endif %}
 {% if paginator.next_page %}
-  <link rel="next" href="{{ paginator.next_page_path | absolute_url }}">
+  <link rel="next" href="{{ paginator.next_page_path | absolute_url }}" />
+{% endif %}
+
+{% if seo_tag.image %}
+  <meta name="twitter:card" content="{{ page.twitter.card | default: site.twitter.card | default: "summary_large_image" }}" />
+  <meta property="twitter:image" content="{{ seo_tag.image.path }}" />
+{% else %}
+  <meta name="twitter:card" content="summary" />
+{% endif %}
+
+{% if seo_tag.image.alt %}
+  <meta name="twitter:image:alt" content="{{ seo_tag.image.alt }}" />
+{% endif %}
+
+{% if seo_tag.page_title %}
+  <meta property="twitter:title" content="{{ seo_tag.page_title }}" />
 {% endif %}
 
 {% if site.twitter %}
-  {% if seo_page_image or page.image.twitter %}
-    <meta name="twitter:card" content="summary_large_image" />
-  {% else %}
-    <meta name="twitter:card" content="summary" />
-  {% endif %}
+  <meta name="twitter:site" content="@{{ site.twitter.username | remove:'@' }}" />
 
-  <meta name="twitter:site" content="@{{ site.twitter.username | replace:"@","" }}" />
-
-  {% if seo_author_twitter %}
-    <meta name="twitter:creator" content="@{{ seo_author_twitter }}" />
+  {% if seo_tag.author.twitter %}
+    <meta name="twitter:creator" content="@{{ seo_tag.author.twitter | remove:'@' }}" />
   {% endif %}
 {% endif %}
 
@@ -172,52 +94,36 @@
   {% endif %}
 {% endif %}
 
-{% if site.google_site_verification %}
+{% if site.webmaster_verifications %}
+  {% if site.webmaster_verifications.google %}
+    <meta name="google-site-verification" content="{{ site.webmaster_verifications.google }}" />
+  {% endif %}
+
+  {% if site.webmaster_verifications.bing %}
+    <meta name="msvalidate.01" content="{{ site.webmaster_verifications.bing }}" />
+  {% endif %}
+
+  {% if site.webmaster_verifications.alexa %}
+    <meta name="alexaVerifyID" content="{{ site.webmaster_verifications.alexa }}" />
+  {% endif %}
+
+  {% if site.webmaster_verifications.yandex %}
+    <meta name="yandex-verification" content="{{ site.webmaster_verifications.yandex }}" />
+  {% endif %}
+
+  {% if site.webmaster_verifications.baidu %}
+    <meta name="baidu-site-verification" content="{{ site.webmaster_verifications.baidu }}" />
+  {% endif %}
+
+  {% if site.webmaster_verifications.facebook %}
+    <meta name="facebook-domain-verification" content="{{ site.webmaster_verifications.facebook }}" />
+  {% endif %}
+{% elsif site.google_site_verification %}
   <meta name="google-site-verification" content="{{ site.google_site_verification }}" />
 {% endif %}
 
 <script type="application/ld+json">
-  {
-    "@context": "http://schema.org",
-
-{% if seo_type %}
-    "@type": {{ seo_type | jsonify }},
-{% endif %}
-
-{% if seo_name %}
-    "name": {{ seo_name | jsonify }},
-{% endif %}
-
-{% if seo_page_title %}
-    "headline": {{ seo_page_title | jsonify }},
-{% endif %}
-
-{% if seo_page_image %}
-    "image": {{ seo_page_image | jsonify }},
-{% endif %}
-
-{% if page.date %}
-    "datePublished": {{ page.date | date_to_xmlschema | jsonify }},
-{% endif %}
-
-{% if page.last_modified_at %}
-    "dateModified": {{ page.last_modified_at | date_to_xmlschema | jsonify }},
-{% endif %}
-
-{% if seo_description %}
-    "description": {{ seo_description | jsonify }},
-{% endif %}
-
-{% if seo_site_logo %}
-    "logo": {{ seo_site_logo | jsonify }},
-{% endif %}
-
-{% if seo_links %}
-    "sameAs": {{ seo_links | jsonify }},
-{% endif %}
-
-    "url": {{ page.url | replace:'/index.html','/' | absolute_url | jsonify }}
-  }
+  {{ seo_tag.json_ld | jsonify }}
 </script>
 
 <!-- End Jekyll SEO tag -->

script/cibuild

@@ -2,6 +2,6 @@
 
 set -ex
 
-bundle exec rake spec
+bundle exec rspec
 bundle exec rubocop -S -D
 bundle exec gem build jekyll-seo-tag.gemspec

script/site

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+bundle exec jekyll serve --source docs

spec/fixtures/_posts/2017-01-03-.md

@@ -0,0 +1,6 @@
+---
+---
+
+# A post without a title
+
+*See* https://github.com/jekyll/jekyll-seo-tag/commit/4f80ea50b773d1995985e35a8a63806516c353c4#commitcomment-22010563

spec/jekyll_seo_tag/author_drop_spec.rb

@@ -0,0 +1,164 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag::AuthorDrop do
+  let(:data) { {} }
+  let(:config) { { "author" => "site_author" } }
+  let(:site) do
+    site = make_site(config)
+    site.data = data
+    site
+  end
+  let(:site_payload) { site.site_payload["site"] }
+
+  let(:name) { "foo" }
+  let(:twitter) { "foo" }
+  let(:picture) { nil }
+  let(:expected_hash) do
+    {
+      "name"    => name,
+      "twitter" => twitter,
+    }
+  end
+
+  let(:page_meta) { { "title" => "page title" } }
+  let(:page)      { make_page(page_meta) }
+  subject { described_class.new(:page => page.to_liquid, :site => site_payload.to_liquid) }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  it "returns the author's name for #to_s" do
+    expect(subject.to_s).to eql("site_author")
+  end
+
+  context "with site.authors as an array" do
+    let("data") { { "authors" => %w(foo bar) } }
+    let(:page_meta) { { "author" => "foo" } }
+
+    it "doesn't error" do
+      expect(subject.to_h).to eql(expected_hash)
+    end
+  end
+
+  context "with site.authors[author] as string" do
+    let("data") { { "authors" => { "foo" => "bar" } } }
+    let(:page_meta) { { "author" => "foo" } }
+
+    it "doesn't error" do
+      expect(subject.to_h).to eql(expected_hash)
+    end
+  end
+
+  [:with, :without].each do |site_data_type|
+    context "#{site_data_type} site.author data" do
+      let(:data) do
+        if site_data_type == :with
+          {
+            "authors" => {
+              "author"        => { "name" => "data_author", "image" => "author.png" },
+              "array_author"  => { "image" => "author.png" },
+              "string_author" => { "image" => "author.png" },
+              "site_author"   => { "image" => "author.png" },
+            },
+          }
+        else
+          {}
+        end
+      end
+
+      {
+        :string       => { "author" => "string_author" },
+        :array        => { "authors" => %w(array_author author2) },
+        :empty_string => { "author" => "" },
+        :nil          => { "author" => nil },
+        :hash         => { "author" => { "name" => "hash_author" } },
+      }.each do |author_type, data|
+        context "with author as #{author_type}" do
+          let(:page_meta) { data }
+          let(:expected_author) do
+            "#{author_type}_author".sub("nil_", "site_").sub("empty_string_", "site_")
+          end
+
+          it "returns the name" do
+            expect(subject["name"]).to eql(expected_author)
+          end
+
+          it "returns the twitter handle" do
+            expect(subject["twitter"]).to eql(expected_author)
+          end
+
+          if site_data_type == :with && author_type != :hash
+            it "returns arbitrary metadata" do
+              expect(subject["image"]).to eql("author.png")
+            end
+          end
+        end
+      end
+    end
+  end
+
+  context "with author as a front matter default" do
+    let(:config) do
+      {
+        "defaults" => [
+          {
+            "scope"  => { "path" => "" },
+            "values" => { "author" => "front matter default" },
+          },
+        ],
+      }
+    end
+
+    it "uses the author from the front matter default" do
+      expect(subject["name"]).to eql("front matter default")
+    end
+  end
+
+  context "twitter" do
+    let(:page_meta) { { "author" => "author" } }
+
+    it "pulls the handle from the author" do
+      expect(subject["twitter"]).to eql("author")
+    end
+
+    context "with an @" do
+      let(:page_meta) do
+        {
+          "author" => {
+            "name"    => "author",
+            "twitter" => "@twitter",
+          },
+        }
+      end
+
+      it "strips the @" do
+        expect(subject["twitter"]).to eql("twitter")
+      end
+    end
+
+    # See https://github.com/jekyll/jekyll-seo-tag/issues/202
+    context "without an author name or handle" do
+      let(:page_meta) { { "author" => { "foo" => "bar" } } }
+
+      it "dosen't blow up" do
+        expect(subject["twitter"]).to be_nil
+      end
+    end
+
+    context "with an explicit handle" do
+      let(:page_meta) do
+        {
+          "author" => {
+            "name"    => "author",
+            "twitter" => "twitter",
+          },
+        }
+      end
+
+      it "pulls the handle from the hash" do
+        expect(subject["twitter"]).to eql("twitter")
+      end
+    end
+  end
+end

spec/jekyll_seo_tag/drop_spec.rb

@@ -0,0 +1,538 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag::Drop do
+  let(:config)    { { "title" => "site title" } }
+  let(:page_meta) { { "title" => "page title" } }
+  let(:page)      { make_page(page_meta) }
+  let(:site)      { make_site(config) }
+  let(:context)   { make_context(:page => page, :site => site) }
+  let(:text) { "" }
+  subject { described_class.new(text, context) }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  # Drop includes liquid filters which expect arguments
+  # By default, in drops, `to_h` will call each public method with no arugments
+  # Here, that would cause the filters to explode. This test ensures that all
+  # public methods don't explode when called without arguments. Don't explode.
+  it "doesn't blow up on to_h" do
+    expect { subject.to_h }.to_not raise_error
+  end
+
+  it "returns the version" do
+    expect(subject.version).to eql(Jekyll::SeoTag::VERSION)
+  end
+
+  context "title?" do
+    it "knows to include the title" do
+      expect(subject.title?).to be_truthy
+    end
+
+    context "with title=false" do
+      let(:text) { "title=false" }
+
+      it "knows not to include the title" do
+        expect(subject.title?).to be_falsy
+      end
+    end
+
+    context "site title" do
+      it "knows the site title" do
+        expect(subject.site_title).to eql("site title")
+      end
+
+      context "with site.name" do
+        let(:config) { { "name" => "site title" } }
+
+        it "knows the site title" do
+          expect(subject.site_title).to eql("site title")
+        end
+      end
+    end
+
+    context "page title" do
+      it "knows the page title" do
+        expect(subject.page_title).to eql("page title")
+      end
+
+      context "without a page title" do
+        let(:page) { make_page }
+
+        it "knows the page title" do
+          expect(subject.page_title).to eql("site title")
+        end
+      end
+    end
+
+    context "title" do
+      context "with a page and site title" do
+        it "builds the title" do
+          expect(subject.title).to eql("page title | site title")
+        end
+      end
+
+      context "with a site description but no page title" do
+        let(:page)  { make_page }
+        let(:config) do
+          { "title" => "site title", "description" => "site description" }
+        end
+
+        it "builds the title" do
+          expect(subject.title).to eql("site title | site description")
+        end
+      end
+
+      context "with a site tagline but no page title" do
+        let(:page)  { make_page }
+        let(:config) do
+          { "title" => "site title", "description" => "site description", "tagline" => "site tagline" }
+        end
+
+        it "builds the title" do
+          expect(subject.title).to eql("site title | site tagline")
+        end
+      end
+
+      context "with just a page title" do
+        let(:site)  { make_site }
+
+        it "builds the title" do
+          expect(subject.title).to eql("page title")
+        end
+      end
+
+      context "with just a site title" do
+        let(:page)  { make_page }
+
+        it "builds the title" do
+          expect(subject.title).to eql("site title")
+        end
+      end
+
+      context "without a page or site title" do
+        let(:page)  { make_page }
+        let(:site)  { make_site }
+
+        it "returns nil" do
+          expect(subject.title).to be_nil
+        end
+      end
+
+      context "with an empty page title" do
+        let(:page_meta) { { :title => "" } }
+
+        it "builds the title" do
+          expect(subject.title).to eql("site title")
+        end
+      end
+
+      context "with an empty site title" do
+        let(:config) { { :title => "" } }
+
+        it "builds the title" do
+          expect(subject.title).to eql("page title")
+        end
+      end
+
+      context "with an empty page and site title" do
+        let(:page_meta) { { :title => "" } }
+        let(:config) { { :title => "" } }
+
+        it "returns nil" do
+          expect(subject.title).to be_nil
+        end
+      end
+    end
+  end
+
+  context "name" do
+    context "with seo.name" do
+      let(:page_meta) do
+        { "seo" => { "name" => "seo name" } }
+      end
+
+      it "uses the seo name" do
+        expect(subject.name).to eql("seo name")
+      end
+    end
+
+    context "the index" do
+      let(:page_meta) { { "permalink" => "/" } }
+
+      context "with site.social.name" do
+        let(:config) { { "social" => { "name" => "social name" } } }
+
+        it "uses site.social.name" do
+          expect(subject.name).to eql("social name")
+        end
+      end
+
+      context "with site.social as an array" do
+        let(:config) { { "social" => %w(a b) } }
+
+        it "uses site.social.name" do
+          expect(subject.name).to be_nil
+        end
+      end
+
+      it "uses the site title" do
+        expect(subject.name).to eql("site title")
+      end
+    end
+
+    context "site description" do
+      context "with a site description" do
+        let(:config) { { :description => "site description " } }
+
+        it "returns the site discription" do
+          expect(subject.site_description).to eql("site description")
+        end
+      end
+
+      context "without a site description" do
+        let(:site) { make_site }
+
+        it "returns nil" do
+          expect(subject.site_description).to be_nil
+        end
+      end
+    end
+
+    context "page description" do
+      context "with a page description" do
+        let(:page_meta) { { "description"=> "page description" } }
+
+        it "uses the page description" do
+          expect(subject.description).to eql("page description")
+        end
+      end
+
+      context "with a page excerpt" do
+        let(:page_meta) { { "description"=> "page excerpt" } }
+
+        it "uses the page description" do
+          expect(subject.description).to eql("page excerpt")
+        end
+      end
+
+      context "with a site description" do
+        let(:config) { { "description"=> "site description" } }
+
+        it "uses the page description" do
+          expect(subject.description).to eql("site description")
+        end
+      end
+
+      context "with no descriptions" do
+        let(:page_meta) { { "description" => nil, "excerpt" => nil } }
+        let(:config) { { "description"=> nil } }
+
+        it "uses returns nil" 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
+      let(:page_meta) { { "author" => "foo" } }
+
+      it "returns an AuthorDrop" do
+        expect(subject.author).to be_a(Jekyll::SeoTag::AuthorDrop)
+      end
+
+      it "passes page information" do
+        expect(subject.author.name).to eql("foo")
+      end
+
+      # Regression test to ensure to_liquid is called on site and page
+      # before being passed to AuthorDrop
+      context "with author as a front matter default" do
+        let(:page_meta) { {} }
+        let(:config) do
+          {
+            "defaults" => [
+              {
+                "scope"  => { "path" => "" },
+                "values" => { "author" => "front matter default" },
+              },
+            ],
+          }
+        end
+
+        it "uses the author from the front matter default" do
+          expect(subject.author["name"]).to eql("front matter default")
+        end
+      end
+    end
+  end
+
+  context "date published" do
+    let(:config) { { "timezone" => "America/New_York" } }
+    let(:page_meta) { { "date" => "2017-01-01" } }
+
+    it "uses page.date" do
+      expect(subject.date_published).to eql("2017-01-01T00:00:00-05:00")
+    end
+  end
+
+  context "date modified" do
+    let(:config) { { "timezone" => "America/New_York" } }
+
+    context "with seo.date_modified" do
+      let(:page_meta) { { "seo" => { "date_modified" => "2017-01-01" } } }
+
+      it "uses seo.date_modified" do
+        expect(subject.date_modified).to eql("2017-01-01T00:00:00-05:00")
+      end
+    end
+
+    context "with page.last_modified_at" do
+      let(:page_meta) { { "last_modified_at" => "2017-01-01" } }
+
+      it "uses page.last_modified_at" do
+        expect(subject.date_modified).to eql("2017-01-01T00:00:00-05:00")
+      end
+    end
+
+    context "date" do
+      let(:page_meta) { { "date" => "2017-01-01" } }
+
+      it "uses page.date" do
+        expect(subject.date_modified).to eql("2017-01-01T00:00:00-05:00")
+      end
+    end
+  end
+
+  context "type" do
+    context "with seo.type set" do
+      let(:page_meta) { { "seo" => { "type" => "test" } } }
+
+      it "uses seo.type" do
+        expect(subject.type).to eql("test")
+      end
+    end
+
+    context "with seo as an array" do
+      let(:page_meta) { { "seo" => %w(a b) } }
+
+      it "uses seo.type" do
+        expect(subject.type).to eql("WebPage")
+      end
+    end
+
+    context "the homepage" do
+      let(:page_meta) { { "permalink" => "/" } }
+
+      it "is a website" do
+        expect(subject.type).to eql("WebSite")
+      end
+    end
+
+    context "the about page" do
+      let(:page) { make_page("permalink" => "/about/") }
+
+      it "is a website" do
+        expect(subject.type).to eql("WebSite")
+      end
+    end
+
+    context "a blog post" do
+      let(:page_meta) { { "date" => "2017-01-01" } }
+
+      it "is a blog post" do
+        expect(subject.type).to eql("BlogPosting")
+      end
+    end
+
+    it "is a webpage" do
+      expect(subject.type).to eql("WebPage")
+    end
+  end
+
+  context "links" do
+    context "with seo.links" do
+      let(:page_meta) { { "seo" => { "links" => %w(foo bar) } } }
+
+      it "uses seo.links" do
+        expect(subject.links).to eql(%w(foo bar))
+      end
+    end
+
+    context "with site.social.links" do
+      let(:config) { { "social" => { "links"=> %w(a b) } } }
+
+      it "doesn't use site.social.links" do
+        expect(subject.links).to be_nil
+      end
+
+      context "the homepage" do
+        let(:page_meta) { { "permalink" => "/" } }
+
+        it "uses site.social.links" do
+          expect(subject.links).to eql(%w(a b))
+        end
+      end
+    end
+  end
+
+  context "logo" do
+    context "without site.logo" do
+      it "returns nothing" do
+        expect(subject.logo).to be_nil
+      end
+    end
+
+    context "with an absolute site.logo" do
+      let(:config) { { "logo" => "http://example.com/image.png" } }
+
+      it "uses site.logo" do
+        expect(subject.logo).to eql("http://example.com/image.png")
+      end
+    end
+
+    context "with a relative site.logo" do
+      let(:config) do
+        {
+          "logo" => "image.png",
+          "url"  => "http://example.com",
+        }
+      end
+
+      it "uses site.logo" do
+        expect(subject.logo).to eql("http://example.com/image.png")
+      end
+    end
+
+    context "with a uri-escaped logo" do
+      let(:config) { { "logo" => "some image.png" } }
+
+      it "escapes the logo" do
+        expect(subject.logo).to eql("/some%20image.png")
+      end
+    end
+  end
+
+  context "image" do
+    let(:image) { "/foo.png" }
+    let(:page_meta) { { "image" => image } }
+
+    it "returns a Drop" do
+      expect(subject.image).to be_a(Jekyll::SeoTag::ImageDrop)
+    end
+
+    it "returns the image" do
+      expect(subject.image["path"]).to eql("/foo.png")
+    end
+  end
+
+  context "lang" do
+    context "with page.lang" do
+      let(:page_meta) { { "lang" => "en_GB" } }
+
+      it "uses page.lang" do
+        expect(subject.page_lang).to eql("en_GB")
+      end
+    end
+
+    context "with site.lang" do
+      let(:config) { { "lang" => "en_GB" } }
+
+      it "uses site.lang" do
+        expect(subject.page_lang).to eql("en_GB")
+      end
+    end
+
+    context "with nothing" do
+      it "defaults" do
+        expect(subject.page_lang).to eql("en_US")
+      end
+    end
+  end
+
+  context "homepage_or_about?" do
+    [
+      "/", "/index.html", "index.html", "/index.htm",
+      "/about/", "/about/index.html",
+    ].each do |permalink|
+      context "when passed '#{permalink}' as a permalink" do
+        let(:page_meta) { { "permalink" => permalink } }
+
+        it "knows it's the home or about page" do
+          expect(subject.send(:homepage_or_about?)).to be_truthy
+        end
+      end
+    end
+
+    context "a random URL" do
+      let(:page_meta) { { "permalink" => "/about-foo/" } }
+
+      it "knows it's not the home or about page" do
+        expect(subject.send(:homepage_or_about?)).to be_falsy
+      end
+    end
+  end
+
+  context "canonical url" do
+    let(:config) { { :url => "http://example.com" } }
+
+    context "when canonical url is specified for a page" do
+      let(:canonical_url) { "https://github.com/jekyll/jekyll-seo-tag/" }
+      let(:page_meta) { { "title" => "page title", "canonical_url" => canonical_url } }
+
+      it "uses specified canonical url" do
+        expect(subject.canonical_url).to eq(canonical_url)
+      end
+    end
+
+    context "when canonical url is not specified for a page" do
+      it "uses site specific canonical url" do
+        expect(subject.canonical_url).to eq("http://example.com/page.html")
+      end
+    end
+  end
+
+  context "pagination" do
+    let(:context) do
+      make_context(
+        { :page => page, :site => site },
+        "paginator" => { "page" => 2, "total_pages" => 10 }
+      )
+    end
+
+    it "render default pagination title" do
+      expect(subject.send(:page_number)).to eq("Page 2 of 10 for ")
+    end
+
+    context "render custom pagination title" do
+      let(:config) { { "seo_paginator_message" => "%<current>s of %<total>s" } }
+
+      it "renders the correct page number" do
+        expect(subject.send(:page_number)).to eq("2 of 10")
+      end
+    end
+  end
+
+  it "exposes the JSON-LD drop" do
+    expect(subject.json_ld).to be_a(Jekyll::SeoTag::JSONLDDrop)
+  end
+end

spec/jekyll_seo_tag/filters_spec.rb

@@ -0,0 +1,24 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag::Filters do
+  let(:page)      { make_page }
+  let(:site)      { make_site }
+  let(:context)   { make_context(:page => page, :site => site) }
+  subject { described_class.new(context) }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  it "stores the context" do
+    expect(subject.instance_variable_get("@context")).to be_a(Liquid::Context)
+  end
+
+  it "exposes jekyll filters" do
+    expect(subject).to respond_to(:markdownify)
+  end
+
+  it "exposes liquid standard filters" do
+    expect(subject).to respond_to(:normalize_whitespace)
+  end
+end

spec/jekyll_seo_tag/image_drop_spec.rb

@@ -0,0 +1,121 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag::ImageDrop do
+  let(:config)    { { "title" => "site title" } }
+  let(:image)     { nil }
+  let(:page_meta) { { "image" => image, "dir" => "foo" } }
+  let(:page)      { make_page(page_meta) }
+  let(:site)      { make_site(config) }
+  let(:context)   { make_context(:page => page, :site => site) }
+  let(:text) { "" }
+  subject { described_class.new(:page => page.to_liquid, :context => context) }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  context "with a post object" do
+    let(:image) { "image.png" }
+    let(:page_meta) { { "image" => image, "date" => "2017-01-01" } }
+    let(:page) { make_post(page_meta) }
+
+    it "returns the image url relative to the post directory" do
+      expect(subject["path"]).to eql("/2017/01/01/image.png")
+    end
+  end
+
+  context "with image as a relative path" do
+    let(:image) { "image.png" }
+
+    it "returns the image with the page dir prepended" do
+      expect(subject["path"]).to eql("/foo/image.png")
+    end
+
+    context "with site.url" do
+      let(:config) { { "url" => "http://example.com" } }
+
+      it "makes the path absolute" do
+        expect(subject["path"]).to eql("http://example.com/foo/image.png")
+      end
+    end
+  end
+
+  context "with image as an absolute path" do
+    let(:image) { "/image.png" }
+
+    it "returns the image" do
+      expect(subject["path"]).to eql("/image.png")
+    end
+
+    context "with site.url" do
+      let(:config) { { "url" => "http://example.com" } }
+
+      it "makes the path absolute" do
+        expect(subject["path"]).to eql("http://example.com/image.png")
+      end
+    end
+
+    context "with a URL-escaped path" do
+      let(:image) { "/some image.png" }
+
+      it "URL-escapes the image" do
+        expect(subject["path"]).to eql("/some%20image.png")
+      end
+    end
+  end
+
+  context "with image as a hash" do
+    context "with an absolute path" do
+      let(:image) { { "path" => "/image.png" } }
+
+      it "returns the image" do
+        expect(subject["path"]).to eql("/image.png")
+      end
+    end
+
+    context "with facebook" do
+      let(:image) { { "facebook" => "/image.png" } }
+
+      it "returns the image" do
+        expect(subject["path"]).to eql("/image.png")
+      end
+    end
+
+    context "with twitter" do
+      let(:image) { { "twitter" => "/image.png" } }
+
+      it "returns the image" do
+        expect(subject["path"]).to eql("/image.png")
+      end
+    end
+
+    context "with some random hash" do
+      let(:image) { { "foo" => "bar" } }
+
+      it "returns nil" do
+        expect(subject["path"]).to be_nil
+      end
+
+      it "returns arbitrary values" do
+        expect(subject["foo"]).to eql("bar")
+      end
+    end
+
+    context "with an invalid path" do
+      let(:image) { ":" }
+
+      it "returns the path" do
+        expect(subject["path"]).to eql(":")
+      end
+    end
+
+    context "with height and width" do
+      let(:image) { { "path" => "image.png", "height" => 5, "width" => 10 } }
+
+      it "returns the height and width" do
+        expect(subject["height"]).to eql(5)
+        expect(subject["width"]).to eql(10)
+      end
+    end
+  end
+end

spec/jekyll_seo_tag/json_ld_drop_spec.rb

@@ -0,0 +1,201 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag::JSONLDDrop do
+  let(:author) { "author" }
+  let(:image) { "image" }
+  let(:metadata) do
+    {
+      "title"       => "title",
+      "author"      => author,
+      "image"       => image,
+      "date"        => "2017-01-01",
+      "description" => "description",
+      "seo"         => {
+        "name"          => "seo name",
+        "date_modified" => "2017-01-02",
+        "links"         => %w(a b),
+      },
+    }
+  end
+  let(:config) do
+    {
+      "logo"     => "logo",
+      "timezone" => "America/New_York",
+    }
+  end
+  let(:page)      { make_page(metadata) }
+  let(:site)      { make_site(config) }
+  let(:context)   { make_context(:page => page, :site => site) }
+  let(:page_drop) { Jekyll::SeoTag::Drop.new("", context) }
+  subject { described_class.new(page_drop) }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  it "returns the context" do
+    expect(subject).to have_key("@context")
+    expect(subject["@context"]).to eql("https://schema.org")
+  end
+
+  it "returns the type" do
+    expect(subject).to have_key("@type")
+    expect(subject["@type"]).to eql("BlogPosting")
+  end
+
+  it "returns the name" do
+    expect(subject).to have_key("name")
+    expect(subject["name"]).to eql("seo name")
+  end
+
+  it "returns the headline" do
+    expect(subject).to have_key("headline")
+    expect(subject["headline"]).to eql("title")
+  end
+
+  context "author" do
+    {
+      "string" => "author",
+      "hash"   => { "name" => "author" },
+    }.each do |key, value|
+      context "when passed as a #{key}" do
+        let(:author) { value }
+
+        it "returns the author" do
+          expect(subject).to have_key("author")
+          expect(subject["author"]).to be_a(Hash)
+          expect(subject["author"]).to have_key("@type")
+          expect(subject["author"]["@type"]).to eql("Person")
+          expect(subject["author"]).to have_key("name")
+          expect(subject["author"]["name"]).to be_a(String)
+          expect(subject["author"]["name"]).to eql("author")
+          expect(subject["author"]["url"]).to be nil
+        end
+      end
+    end
+
+    context "when type Organization" do
+      let(:author) { { "name" => "organization", "type" => "Organization" } }
+
+      it "returns the author with type" do
+        expect(subject).to have_key("author")
+        expect(subject["author"]).to be_a(Hash)
+        expect(subject["author"]).to have_key("@type")
+        expect(subject["author"]["@type"]).to eql("Organization")
+        expect(subject["author"]).to have_key("name")
+        expect(subject["author"]["name"]).to be_a(String)
+        expect(subject["author"]["name"]).to eql("organization")
+      end
+    end
+
+    context "when invalid type" do
+      let(:author) { { "name" => "organization", "type" => "Invalid" } }
+
+      it "returns the author with type" do
+        expect(subject).to have_key("author")
+        expect(subject["author"]).to be nil
+      end
+    end
+
+    context "when url is present" do
+      let(:author) { { "name" => "author", "url" => "https://example.com" } }
+
+      it "returns the author with url" do
+        expect(subject).to have_key("author")
+        expect(subject["author"]).to be_a(Hash)
+        expect(subject["author"]).to have_key("url")
+        expect(subject["author"]["url"]).to eql("https://example.com")
+        expect(subject["author"]).to have_key("name")
+        expect(subject["author"]["name"]).to be_a(String)
+        expect(subject["author"]["name"]).to eql("author")
+      end
+    end
+  end
+
+  context "image" do
+    context "with image as an absolute path" do
+      let(:image) { "/image" }
+
+      it "returns the image as a string" do
+        expect(subject).to have_key("image")
+        expect(subject["image"]).to be_a(String)
+        expect(subject["image"]).to eql("/image")
+      end
+    end
+
+    context "with image as a hash" do
+      let(:image) { { "path" => "/image", "height" => 5, "width" => 10 } }
+
+      it "returns the image as a hash" do
+        expect(subject).to have_key("image")
+        expect(subject["image"]).to be_a(Hash)
+        expect(subject["image"]).to eql(
+          "@type"  => "imageObject",
+          "url"    => "/image",
+          "height" => 5,
+          "width"  => 10
+        )
+      end
+    end
+  end
+
+  it "returns the datePublished" do
+    expect(subject).to have_key("datePublished")
+    expect(subject["datePublished"]).to eql("2017-01-01T00:00:00-05:00")
+  end
+
+  it "returns the dateModified" do
+    expect(subject).to have_key("dateModified")
+    expect(subject["dateModified"]).to eql("2017-01-02T00:00:00-05:00")
+  end
+
+  it "returns the description" do
+    expect(subject).to have_key("description")
+    expect(subject["description"]).to eql("description")
+  end
+
+  it "returns the publisher" do
+    expect(subject).to have_key("publisher")
+
+    publisher = subject["publisher"]
+    expect(publisher).to be_a(Hash)
+
+    expect(publisher).to have_key("@type")
+    expect(publisher["@type"]).to eql("Organization")
+    expect(publisher).to have_key("logo")
+
+    logo = publisher["logo"]
+    expect(logo).to have_key("@type")
+    expect(logo["@type"]).to eql("ImageObject")
+    expect(logo).to have_key("url")
+    expect(logo["url"]).to eql("/logo")
+  end
+
+  it "returns the main entity of page" do
+    expect(subject).to have_key("mainEntityOfPage")
+    expect(subject["mainEntityOfPage"]).to be_a(Hash)
+    expect(subject["mainEntityOfPage"]).to have_key("@type")
+    expect(subject["mainEntityOfPage"]["@type"]).to eql("WebPage")
+    expect(subject["mainEntityOfPage"]).to have_key("@id")
+    expect(subject["mainEntityOfPage"]["@id"]).to eql("/page.html")
+  end
+
+  it "returns sameAs" do
+    expect(subject).to have_key("sameAs")
+    expect(subject["sameAs"]).to be_a(Array)
+    expect(subject["sameAs"]).to eql(%w(a b))
+  end
+
+  it "returns the url" do
+    expect(subject).to have_key("url")
+    expect(subject["url"]).to eql("/page.html")
+  end
+
+  context "with null values" do
+    let(:metadata) { {} }
+
+    it "does not return null values as json" do
+      expect(subject.to_json).to_not match(%r!:null!)
+    end
+  end
+end

spec/jekyll_seo_tag_integration_spec.rb

@@ -0,0 +1,720 @@
+# frozen_string_literal: true
+
+RSpec.describe Jekyll::SeoTag do
+  let(:page)      { make_page }
+  let(:site)      { make_site }
+  let(:post)      { make_post }
+  let(:context)   { make_context(:page => page, :site => site) }
+  let(:tag)       { "seo" }
+  let(:text)      { "" }
+  let(:output)    { Liquid::Template.parse("{% #{tag} #{text} %}").render!(context, {}) }
+  let(:json)      { output.match(%r!<script type="application/ld\+json">(.*)</script>!m)[1] }
+  let(:json_data) { JSON.parse(json) }
+  let(:paginator) { { "previous_page" => true, "previous_page_path" => "foo", "next_page" => true, "next_page_path" => "bar" } }
+
+  before do
+    Jekyll.logger.log_level = :error
+  end
+
+  it "builds" do
+    expect(output).to match(%r!Jekyll SEO tag!i)
+  end
+
+  it "outputs the plugin version" do
+    version = Jekyll::SeoTag::VERSION
+    expect(output).to match(%r!Jekyll SEO tag v#{version}!i)
+  end
+
+  it "outputs meta generator" do
+    version = Jekyll::VERSION
+    expect(output).to match(%r!Jekyll v#{version}!i)
+  end
+
+  it "outputs JSON sorted by key" do
+    expect(json.strip).to eql('{"@context":"https://schema.org","@type":"WebPage","url":"/page.html"}')
+  end
+
+  it "outputs valid HTML" do
+    site.process
+    options = {
+      :check_html       => true,
+      :checks_to_ignore => %w(ScriptCheck LinkCheck ImageCheck),
+    }
+    status = HTMLProofer.check_directory(dest_dir, options).run
+    expect(status).to eql(true)
+  end
+
+  context "with page.title" do
+    let(:page) { make_page("title" => "foo") }
+
+    it "builds the title with a page title only" do
+      expect(output).to match(%r!<title>foo</title>!)
+      expected = %r!<meta property="og:title" content="foo" />!
+      expect(output).to match(expected)
+    end
+
+    context "with site.name" do
+      let(:site) { make_site("name" => "Site Name") }
+
+      it "builds the title with a page title and site name" do
+        expect(output).to match(%r!<title>foo \| Site Name</title>!)
+      end
+    end
+
+    context "with site.title" do
+      let(:site) { make_site("title" => "bar") }
+
+      it "builds the title with a page title and site title" do
+        expect(output).to match(%r!<title>foo \| bar</title>!)
+      end
+    end
+
+    context "with site.description" do
+      let(:site) { make_site("description" => "Site Description") }
+
+      it "builds the title without the site description" do
+        expect(output).not_to match(%r!<title>foo \| Site Description</title>!)
+      end
+    end
+
+    context "with site.title and site.description" do
+      let(:site) { make_site("title" => "Site Title", "description" => "Site Description") }
+
+      it "builds the title with a page title and site title" do
+        expect(output).to match(%r!<title>foo \| Site Title</title>!)
+      end
+
+      it "does not build the title with the site description" do
+        expect(output).not_to match(%r!<title>foo \| Site Description</title>!)
+      end
+    end
+
+    context "with site.title and site.description" do
+      let(:site) { make_site("title" => "Site Title", "description" => "Site Description") }
+
+      it "builds the title with a page title and site title" do
+        expect(output).to match(%r!<title>foo \| Site Title</title>!)
+      end
+
+      it "does not build the title with the site description" do
+        expect(output).not_to match(%r!<title>Page Title \| Site Description</title>!)
+      end
+    end
+  end
+
+  context "with site.title" do
+    let(:site) { make_site("title" => "Site title") }
+
+    it "builds the title with only a site title" do
+      expect(output).to match(%r!<title>Site title</title>!)
+    end
+  end
+
+  context "with site.title and site.description" do
+    let(:site) { make_site("title" => "Site Title", "description" => "Site Description") }
+
+    it "builds the title with site title and description" do
+      expect(output).to match(%r!<title>Site Title \| Site Description</title>!)
+    end
+  end
+
+  context "with page.description" do
+    let(:page) { make_page("description" => "foo") }
+
+    it "uses the page description" do
+      expect(output).to match(%r!<meta name="description" content="foo" />!)
+      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
+      expect(output).to match(%r!<meta property="twitter:description" content="foo" />!)
+    end
+  end
+
+  context "with page.excerpt" do
+    let(:page) { make_page("excerpt" => "foo") }
+
+    it "uses the page excerpt when no page description exists" do
+      expect(output).to match(%r!<meta name="description" content="foo" />!)
+      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
+      expect(output).to match(%r!<meta property="twitter:description" content="foo" />!)
+    end
+  end
+
+  context "with site.description" do
+    let(:site) { make_site("description" => "foo") }
+
+    it "uses the site description when no page description nor excerpt exist" do
+      expect(output).to match(%r!<meta name="description" content="foo" />!)
+      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
+      expect(output).to match(%r!<meta property="twitter:description" content="foo" />!)
+    end
+  end
+
+  context "with site.url" do
+    let(:site) { make_site("url" => "http://example.invalid") }
+
+    it "uses the site url to build the seo url" do
+      expected = %r!<link rel="canonical" href="http://example.invalid/page.html" />!
+      expect(output).to match(expected)
+      expected = %r!<meta property="og:url" content="http://example.invalid/page.html" />!
+      expect(output).to match(expected)
+    end
+
+    context "with page.permalink" do
+      let(:page) { make_page("permalink" => "/page/index.html") }
+
+      it "uses replaces '/index.html' with '/'" do
+        expected = %r!<link rel="canonical" href="http://example.invalid/page/" />!
+        expect(output).to match(expected)
+
+        expected = %r!<meta property="og:url" content="http://example.invalid/page/" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "with site.baseurl" do
+      let(:site) { make_site("url" => "http://example.invalid", "baseurl" => "/foo") }
+
+      it "uses baseurl to build the seo url" do
+        expected = %r!<link rel="canonical" href="http://example.invalid/foo/page.html" />!
+        expect(output).to match(expected)
+        expected = %r!<meta property="og:url" content="http://example.invalid/foo/page.html" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "with relative page.image as a string" do
+      let(:page) { make_page("image" => "/img/foo.png") }
+
+      it "outputs an open graph image" do
+        expected = '<meta property="og:image" content="http://example.invalid/img/foo.png" />'
+        expect(output).to include(expected)
+      end
+    end
+
+    context "with absolute page.image" do
+      let(:page) { make_page("image" => "http://cdn.example.invalid/img/foo.png") }
+
+      it "outputs an open graph image" do
+        expected = '<meta property="og:image" content="http://cdn.example.invalid/img/foo.png" />'
+        expect(output).to include(expected)
+      end
+    end
+
+    context "with page.image as an object" do
+      context "when given a path" do
+        let(:page) { make_page("image" => { "path" => "/img/foo.png" }) }
+
+        it "outputs an open graph image" do
+          expected = %r!<meta property="og:image" content="http://example.invalid/img/foo.png" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "when given a facebook image" do
+        let(:page) { make_page("image" => { "facebook" => "/img/facebook.png" }) }
+
+        it "outputs an open graph image" do
+          expected = %r!<meta property="og:image" content="http://example.invalid/img/facebook.png" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "when given a twitter image" do
+        let(:page) { make_page("image" => { "twitter" => "/img/twitter.png" }) }
+
+        it "outputs an open graph image" do
+          expected = %r!<meta property="og:image" content="http://example.invalid/img/twitter.png" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "when given an image height and width" do
+        let(:image) { { "path" => "/img/foo.png", "height" => 1, "width" => 2 } }
+        let(:page) { make_page("image" => image) }
+
+        it "outputs an open graph image width and height" do
+          expected = %r!<meta property="og:image:height" content="1" />!
+          expect(output).to match(expected)
+          expected = %r!<meta property="og:image:width" content="2" />!
+          expect(output).to match(expected)
+        end
+      end
+    end
+
+    context "with site.logo" do
+      let(:site) { make_site("logo" => "/logo.png", "url" => "http://example.invalid") }
+
+      it "outputs the logo" do
+        expect(json_data["publisher"]["logo"]["url"]).to eql("http://example.invalid/logo.png")
+      end
+    end
+
+    context "with absolute site.logo" do
+      let(:site) { make_site("logo" => "http://cdn.example.invalid/logo.png", "url" => "http://example.invalid") }
+
+      it "outputs the logo" do
+        expect(json_data["publisher"]["logo"]["url"]).to eql("http://cdn.example.invalid/logo.png")
+      end
+    end
+
+    context "with site.logo and page.author" do
+      let(:site) { make_site("logo" => "http://cdn.example.invalid/logo.png", "url" => "http://example.invalid", "author" => "Mr. Foo") }
+
+      it "outputs the author" do
+        expect(json_data["publisher"]["name"]).to eql("Mr. Foo")
+      end
+    end
+
+    context "with page author" do
+      let(:site) { make_site("logo" => "/logo.png", "url" => "http://example.invalid") }
+      let(:page) { make_post("author" => "Mr. Foo") }
+
+      it "outputs the author" do
+        expect(json_data["author"]["@type"]).to eql("Person")
+        expect(json_data["author"]["name"]).to eql("Mr. Foo")
+      end
+
+      it "outputs the publisher author" do
+        expect(json_data["publisher"]["name"]).to eql("Mr. Foo")
+      end
+    end
+
+    context "with seo type is BlogPosting" do
+      let(:site) { make_site("url" => "http://example.invalid") }
+      let(:page) { make_post("seo" => { "type" => "BlogPosting" }, "permalink" => "/foo/") }
+
+      it "outputs the mainEntityOfPage" do
+        expect(json_data["mainEntityOfPage"]["@type"]).to eql("WebPage")
+        expect(json_data["mainEntityOfPage"]["@id"]).to eql("http://example.invalid/foo/")
+      end
+    end
+
+    context "with seo type is CreativeWork" do
+      let(:site) { make_site("url" => "http://example.invalid") }
+      let(:page) { make_post("seo" => { "type" => "CreativeWork" }, "permalink" => "/foo/") }
+
+      it "outputs the mainEntityOfPage" do
+        expect(json_data["mainEntityOfPage"]["@type"]).to eql("WebPage")
+        expect(json_data["mainEntityOfPage"]["@id"]).to eql("http://example.invalid/foo/")
+      end
+    end
+
+    context "with site.title" do
+      let(:site) { make_site("title" => "Foo", "url" => "http://example.invalid") }
+
+      it "outputs the site title meta" do
+        expect(output).to match(%r!<meta property="og:site_name" content="Foo" />!)
+      end
+
+      it "minifies the output" do
+        version = Jekyll::SeoTag::VERSION
+        expected = <<~HTML
+          <!-- Begin Jekyll SEO tag v#{version} -->
+          <title>Foo</title>
+          <meta name="generator" content="Jekyll v#{Jekyll::VERSION}" />
+          <meta property="og:title" content="Foo" />
+          <meta property="og:locale" content="en_US" />
+          <link rel="canonical" href="http://example.invalid/page.html" />
+          <meta property="og:url" content="http://example.invalid/page.html" />
+          <meta property="og:site_name" content="Foo" />
+        HTML
+        expect(output).to match(expected)
+      end
+    end
+  end
+
+  context "posts" do
+    context "with post meta" do
+      let(:site) { make_site("url" => "http://example.invalid") }
+      let(:meta) do
+        {
+          "title"       => "post",
+          "description" => "description",
+          "image"       => "/img.png",
+        }
+      end
+      let(:page) { make_post(meta) }
+
+      it "outputs post meta" do
+        expected = %r!<meta property="og:type" content="article" />!
+        expect(output).to match(expected)
+
+        expect(json_data["headline"]).to eql("post")
+        expect(json_data["description"]).to eql("description")
+        expect(json_data["image"]).to eql("http://example.invalid/img.png")
+      end
+
+      it "minifies JSON-LD" do
+        expect(output).to_not match(%r!{.*?\s.*?}!)
+      end
+
+      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
+
+  context "facebook" do
+    let(:site_facebook) do
+      {
+        "admins"    => "jekyllrb-fb-admins",
+        "app_id"    => "jekyllrb-fb-app_id",
+        "publisher" => "jekyllrb-fb-publisher",
+      }
+    end
+
+    let(:site) { make_site("facebook" => site_facebook) }
+
+    it "outputs facebook admins meta" do
+      expected = %r!<meta property="fb:admins" content="jekyllrb-fb-admins" />!
+      expect(output).to match(expected)
+    end
+
+    it "outputs facebook app ID meta" do
+      expected = %r!<meta property="fb:app_id" content="jekyllrb-fb-app_id" />!
+      expect(output).to match(expected)
+    end
+
+    it "outputs facebook article publisher meta" do
+      expected = %r!<meta property="article:publisher" content="jekyllrb-fb-publisher" />!
+      expect(output).to match(expected)
+    end
+  end
+
+  context "twitter" do
+    context "with site.twitter.username" do
+      let(:site_twitter) { { "username" => "jekyllrb" } }
+      let(:site) { make_site("twitter" => site_twitter) }
+
+      context "with page.author as a string" do
+        let(:page) { make_page("author" => "benbalter") }
+
+        it "outputs twitter card meta" do
+          expected = %r!<meta name="twitter:card" content="summary" />!
+          expect(output).to match(expected)
+
+          expected = %r!<meta name="twitter:site" content="@jekyllrb" />!
+          expect(output).to match(expected)
+
+          expected = %r!<meta name="twitter:creator" content="@benbalter" />!
+          expect(output).to match(expected)
+        end
+
+        context "with an @" do
+          let(:page) { make_page("author" => "@benbalter") }
+
+          it "outputs the twitter card" do
+            expected = %r!<meta name="twitter:creator" content="@benbalter" />!
+            expect(output).to match(expected)
+          end
+        end
+
+        context "with site.data.authors" do
+          let(:author_data) { {} }
+          let(:data) { { "authors" => author_data } }
+          let(:site) { make_site("data" => data, "twitter" => site_twitter) }
+
+          context "with the author in site.data.authors" do
+            let(:author_data) { { "benbalter" => { "twitter" => "test" } } }
+
+            it "outputs the twitter card" do
+              expected = %r!<meta name="twitter:creator" content="@test" />!
+              expect(output).to match(expected)
+            end
+          end
+
+          context "without the author in site.data.authors" do
+            it "outputs the twitter card" do
+              expected = %r!<meta name="twitter:creator" content="@benbalter" />!
+              expect(output).to match(expected)
+            end
+          end
+        end
+      end
+
+      context "with page.image" do
+        context "without *.twitter.card" do
+          let(:page) { make_page("image" => "/img/foo.png") }
+
+          it "outputs summary card with large image" do
+            expected = %r!<meta name="twitter:card" content="summary_large_image" />!
+            expect(output).to match(expected)
+          end
+        end
+
+        context "with page.twitter.card" do
+          let(:page_twitter) { { "card" => "summary" } }
+          let(:page) { make_page("image" => "/img/foo.png", "twitter" => page_twitter) }
+
+          it "outputs summary card with small image" do
+            expected = %r!<meta name="twitter:card" content="summary" />!
+            expect(output).to match(expected)
+          end
+        end
+
+        context "with site.twitter.card" do
+          let(:site_twitter) { { "username" => "jekyllrb", "card" => "summary" } }
+          let(:site) { make_site("twitter" => site_twitter) }
+          let(:page) { make_page("image" => "/img/foo.png") }
+
+          it "outputs summary card with small image" do
+            expected = %r!<meta name="twitter:card" content="summary" />!
+            expect(output).to match(expected)
+          end
+        end
+      end
+
+      context "with page.author as a hash" do
+        let(:page) { make_page("author" => { "twitter" => "@test" }) }
+
+        it "supports author data as a hash" do
+          expected = %r!<meta name="twitter:creator" content="@test" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "with page.authors as an array" do
+        let(:page) { make_page("authors" => %w(test foo)) }
+
+        it "supports author data as an array" do
+          expected = %r!<meta name="twitter:creator" content="@test" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "with site.author as a hash" do
+        let(:author) { { "twitter" => "@test" } }
+        let(:site) { make_site("author" => author, "twitter" => site_twitter) }
+
+        it "supports author data as an hash" do
+          expected = %r!<meta name="twitter:creator" content="@test" />!
+          expect(output).to match(expected)
+        end
+      end
+    end
+  end
+
+  context "author" do
+    let(:site) { make_site("author" => "Site Author") }
+
+    context "with site.author" do
+      it "outputs site author metadata" do
+        expected = %r!<meta name="author" content="Site Author" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "with page.author" do
+      let(:page) { make_page("author" => "Page Author") }
+
+      it "outputs page author metadata" do
+        expected = %r!<meta name="author" content="Page Author" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "without page.author" do
+      let(:page) { make_page("author" => "") }
+
+      it "outputs site author metadata" do
+        expected = %r!<meta name="author" content="Site Author" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "with site.data.authors" do
+      let(:author_data) { { "renshuki" => { "name" => "Site Data Author" } } }
+      let(:data) { { "authors" => author_data } }
+
+      context "with the author in site.data.authors" do
+        let(:site) { make_site("data" => data, "author" => "renshuki") }
+        it "outputs the author metadata" do
+          expected = %r!<meta name="author" content="Site Data Author" />!
+          expect(output).to match(expected)
+        end
+      end
+
+      context "without the author in site.data.authors" do
+        it "outputs site author metadata" do
+          expected = %r!<meta name="author" content="Site Author" />!
+          expect(output).to match(expected)
+        end
+      end
+    end
+  end
+
+  context "with site.social" do
+    let(:links) { ["http://foo.invalid", "http://bar.invalid"] }
+    let(:social_namespace) { { "name" => "Ben", "links" => links } }
+    let(:site) { make_site("social" => social_namespace) }
+    let(:page) { make_post(meta) }
+
+    context "on homepage" do
+      let(:meta) do
+        {
+          "permalink" => "/",
+          "seo"       => {
+            "type" => "person",
+          },
+        }
+      end
+
+      it "outputs social meta" do
+        expect(json_data["@type"]).to eql("person")
+        expect(json_data["name"]).to eql("Ben")
+        expect(json_data["sameAs"]).to eql(links)
+      end
+    end
+
+    context "on about page" do
+      let(:meta) { { "permalink" => "/about/" } }
+
+      it "outputs sameAs links" do
+        expect(json_data["sameAs"]).to eql(links)
+      end
+    end
+
+    context "on other pages" do
+      let(:meta) { { "permalink" => "/2/" } }
+
+      it "does not output sameAs links" do
+        expect(json_data["sameAs"]).to be_nil
+      end
+    end
+  end
+
+  context "with site.name" do
+    let(:site) { make_site("name" => "Site name") }
+
+    it "uses site.name if site.title is not present" do
+      expected = %r!<meta property="og:site_name" content="Site name" />!
+      expect(output).to match(expected)
+    end
+
+    context "with site.title" do
+      let(:site)  { make_site("name" => "Site Name", "title" => "Site Title") }
+
+      it "uses site.tile if both site.title and site.name are present" do
+        expected = %r!<meta property="og:site_name" content="Site Title" />!
+        expect(output).to match(expected)
+      end
+    end
+  end
+
+  context "with title=false" do
+    let(:text) { "title=false" }
+
+    it "does not output a <title> tag" do
+      expect(output).not_to match(%r!<title>!)
+    end
+  end
+
+  context "with pagination" do
+    let(:context) { make_context({}, "paginator" => paginator) }
+
+    it "outputs pagination links" do
+      expect(output).to match(%r!<link rel="prev" href="/foo" />!)
+      expect(output).to match(%r!<link rel="next" href="/bar" />!)
+    end
+  end
+
+  context "webmaster verification" do
+    context "with site.webmaster_verifications" do
+      let(:site_verifications) do
+        {
+          "google"   => "foo",
+          "bing"     => "bar",
+          "alexa"    => "baz",
+          "yandex"   => "bat",
+          "facebook" => "bas",
+        }
+      end
+
+      let(:site) { make_site("webmaster_verifications" => site_verifications) }
+
+      it "outputs google verification meta" do
+        expected = %r!<meta name="google-site-verification" content="foo" />!
+        expect(output).to match(expected)
+      end
+
+      it "outputs bing verification meta" do
+        expected = %r!<meta name="msvalidate.01" content="bar" />!
+        expect(output).to match(expected)
+      end
+
+      it "outputs alexa verification meta" do
+        expected = %r!<meta name="alexaVerifyID" content="baz" />!
+        expect(output).to match(expected)
+      end
+
+      it "outputs yandex verification meta" do
+        expected = %r!<meta name="yandex-verification" content="bat" />!
+        expect(output).to match(expected)
+      end
+
+      it "outputs facebook verification meta" do
+        expected = %r!<meta name="facebook-domain-verification" content="bas" />!
+        expect(output).to match(expected)
+      end
+    end
+
+    context "with site.google_site_verification" do
+      let(:site) { make_site("google_site_verification" => "foo") }
+
+      it "outputs google verification meta" do
+        expected = %r!<meta name="google-site-verification" content="foo" />!
+        expect(output).to match(expected)
+      end
+    end
+  end
+
+  context "with locale" do
+    it "uses en_US when no locale is specified" do
+      expected = %r!<meta property="og:locale" content="en_US" />!
+      expect(output).to match(expected)
+    end
+
+    context "with site.lang" do
+      let(:site)  { make_site("lang" => "de_DE") }
+
+      it "uses site.lang if page.lang is not present" do
+        expected = %r!<meta property="og:locale" content="de_DE" />!
+        expect(output).to match(expected)
+      end
+
+      context "with page.lang" do
+        let(:page)  { make_page("lang" => "en_UK") }
+
+        it "uses page.lang if both site.lang and page.lang are present" do
+          expected = %r!<meta property="og:locale" content="en_UK" />!
+          expect(output).to match(expected)
+        end
+      end
+    end
+
+    context "with site.lang hyphenated" do
+      let(:site)  { make_site("lang" => "en-AU") }
+
+      it "coerces hyphen to underscore" do
+        expected = %r!<meta property="og:locale" content="en_AU" />!
+        expect(output).to match(expected)
+      end
+    end
+  end
+end

spec/jekyll_seo_tag_spec.rb

@@ -1,443 +1,52 @@
-require "spec_helper"
+# frozen_string_literal: true
 
-describe Jekyll::SeoTag do
-  let(:page)      { make_page }
-  let(:site)      { make_site }
-  let(:post)      { make_post }
-  let(:context)   { make_context(:page => page, :site => site) }
-  let(:tag)       { "seo" }
-  let(:text)      { "" }
-  let(:output)    { Liquid::Template.parse("{% #{tag} #{text} %}").render!(context, {}) }
-  let(:json)      { output.match(%r!<script type=\"application/ld\+json\">(.*)</script>!m)[1] }
-  let(:json_data) { JSON.parse(json) }
-  let(:paginator) { { "previous_page" => true, "previous_page_path" => "foo", "next_page" => true, "next_page_path" => "bar" } }
+RSpec.describe Jekyll::SeoTag do
+  let(:config)    { { "title" => "site title" } }
+  let(:page_meta) { {} }
+  let(:page)      { make_page(page_meta) }
+  let(:site)      { make_site(config) }
+  let(:render_context) { make_context(:page => page, :site => site) }
+  let(:text) { "" }
+  let(:tag_name) { "github_edit_link" }
+  let(:tokenizer) { Liquid::Tokenizer.new("") }
+  let(:parse_context) { Liquid::ParseContext.new }
+  let(:rendered) { subject.render(render_context) }
+  let(:payload) { subject.send(:payload) }
+
+  subject do
+    tag = described_class.parse(tag_name, text, tokenizer, parse_context)
+    tag.instance_variable_set("@context", render_context)
+    tag
+  end
 
   before do
     Jekyll.logger.log_level = :error
   end
 
-  it "builds" do
-    expect(output).to match(%r!Jekyll SEO tag!i)
+  it "returns the template" do
+    expect(described_class.template).to be_a(Liquid::Template)
   end
 
-  it "outputs the plugin version" do
-    version = Jekyll::SeoTag::VERSION
-    expect(output).to match(%r!Jekyll SEO tag v#{version}!i)
-  end
-
-  it "outputs valid HTML" do
-    site.process
-    options = {
-      :check_html       => true,
-      :checks_to_ignore => %w(ScriptCheck LinkCheck ImageCheck)
-    }
-    status = HTML::Proofer.new(dest_dir, options).run
-    expect(status).to eql(true)
-  end
-
-  context "with page.title" do
-    let(:page) { make_page("title" => "foo") }
-
-    it "builds the title with a page title only" do
-      expect(output).to match(%r!<title>foo</title>!)
-      expected = %r!<meta property="og:title" content="foo" />!
-      expect(output).to match(expected)
+  context "payload" do
+    it "contains the drop" do
+      expect(payload["seo_tag"]).to be_a(Jekyll::SeoTag::Drop)
     end
 
-    context "with site.title" do
-      let(:site) { make_site("title" => "bar") }
+    it "contains the Jekyll drop" do
+      expect(payload["jekyll"]).to be_a(Jekyll::Drops::JekyllDrop)
+    end
 
-      it "builds the title with a page title and site title" do
-        expect(output).to match(%r!<title>foo - bar</title>!)
-      end
+    it "contains the page" do
+      expect(payload["page"]).to be_a(Jekyll::Page)
+    end
+
+    it "contains the site" do
+      expect(payload["site"]).to be_a(Jekyll::Drops::SiteDrop)
     end
   end
 
-  context "with site.title" do
-    let(:site) { make_site("title" => "Site title") }
-
-    it "builds the title with only a site title" do
-      expect(output).to match(%r!<title>Site title</title>!)
-    end
-  end
-
-  context "with page.description" do
-    let(:page) { make_page("description" => "foo") }
-
-    it "uses the page description" do
-      expect(output).to match(%r!<meta name="description" content="foo" />!)
-      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
-    end
-  end
-
-  context "with page.excerpt" do
-    let(:page) { make_page("excerpt" => "foo") }
-
-    it "uses the page excerpt when no page description exists" do
-      expect(output).to match(%r!<meta name="description" content="foo" />!)
-      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
-    end
-  end
-
-  context "with site.description" do
-    let(:site) { make_site("description" => "foo") }
-
-    it "uses the site description when no page description nor excerpt exist" do
-      expect(output).to match(%r!<meta name="description" content="foo" />!)
-      expect(output).to match(%r!<meta property="og:description" content="foo" />!)
-    end
-  end
-
-  context "with site.url" do
-    let(:site) { make_site("url" => "http://example.invalid") }
-
-    it "uses the site url to build the seo url" do
-      expected = %r!<link rel="canonical" href="http://example.invalid/page.html" />!
-      expect(output).to match(expected)
-      expected = %r!<meta property="og:url" content="http://example.invalid/page.html" />!
-      expect(output).to match(expected)
-    end
-
-    context "with page.permalink" do
-      let(:page) { make_page("permalink" => "/page/index.html") }
-
-      it "uses replaces '/index.html' with '/'" do
-        expected = %r!<link rel="canonical" href="http://example.invalid/page/" />!
-        expect(output).to match(expected)
-
-        expected = %r!<meta property="og:url" content="http://example.invalid/page/" />!
-        expect(output).to match(expected)
-      end
-    end
-
-    context "with site.baseurl" do
-      let(:site) { make_site("url" => "http://example.invalid", "baseurl" => "/foo") }
-
-      it "uses baseurl to build the seo url" do
-        expected = %r!<link rel="canonical" href="http://example.invalid/foo/page.html" />!
-        expect(output).to match(expected)
-        expected = %r!<meta property="og:url" content="http://example.invalid/foo/page.html" />!
-        expect(output).to match(expected)
-      end
-    end
-
-    context "with relative page.image as a string" do
-      let(:page) { make_page("image" => "/img/foo.png") }
-
-      it "outputs the image" do
-        expected = '<meta property="og:image" content="http://example.invalid/img/foo.png" />'
-        expect(output).to include(expected)
-      end
-    end
-
-    context "with absolute page.image" do
-      let(:page) { make_page("image" => "http://cdn.example.invalid/img/foo.png") }
-
-      it "outputs the image" do
-        expected = '<meta property="og:image" content="http://cdn.example.invalid/img/foo.png" />'
-        expect(output).to include(expected)
-      end
-    end
-
-    context "with page.image as an object" do
-      context "when given a path" do
-        let(:page) { make_page("image" => { "path" => "/img/foo.png" }) }
-
-        it "outputs the image" do
-          expected = %r!<meta property="og:image" content="http://example.invalid/img/foo.png" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "when given a facebook image" do
-        let(:page) { make_page("image" => { "facebook" => "/img/facebook.png" }) }
-
-        it "outputs the image" do
-          expected = %r!<meta property="og:image" content="http://example.invalid/img/facebook.png" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "when given a twitter image" do
-        let(:page) { make_page("image" => { "twitter" => "/img/twitter.png" }) }
-
-        it "outputs the image" do
-          expected = %r!<meta name="twitter:image" content="http://example.invalid/img/twitter.png" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "when given the image height and width" do
-        let(:image) { { "facebook" => "/img/foo.png", "height" => 1, "width" => 2 } }
-        let(:page) { make_page("image" => image) }
-
-        it "outputs the image" do
-          expected = %r!<meta property="og:image:height" content="1" />!
-          expect(output).to match(expected)
-          expected = %r!<meta property="og:image:width" content="2" />!
-          expect(output).to match(expected)
-        end
-      end
-    end
-
-    context "with site.logo" do
-      let(:site) { make_site("logo" => "/logo.png", "url" => "http://example.invalid") }
-
-      it "outputs the logo" do
-        expect(json_data["logo"]).to eql("http://example.invalid/logo.png")
-      end
-    end
-
-    context "with absolute site.logo" do
-      let(:site) { make_site("logo" => "http://cdn.example.invalid/logo.png", "url" => "http://example.invalid") }
-
-      it "outputs the logo" do
-        expect(json_data["logo"]).to eql("http://cdn.example.invalid/logo.png")
-      end
-    end
-
-    context "with site.title" do
-      let(:site) { make_site("title" => "Foo", "url" => "http://example.invalid") }
-
-      it "outputs the site title meta" do
-        expect(output).to match(%r!<meta property="og:site_name" content="Foo" />!)
-      end
-
-      it "minifies the output" do
-        version = Jekyll::SeoTag::VERSION
-        expected = <<-EOS
-<!-- Begin Jekyll SEO tag v#{version} -->
-<title>Foo</title>
-<meta property="og:title" content="Foo" />
-<link rel="canonical" href="http://example.invalid/page.html" />
-<meta property="og:url" content="http://example.invalid/page.html" />
-<meta property="og:site_name" content="Foo" />
-EOS
-        expect(output).to match(expected)
-      end
-    end
-  end
-
-  context "posts" do
-    context "with post meta" do
-      let(:site) { make_site("url" => "http://example.invalid") }
-      let(:meta) do
-        {
-          "title"       => "post",
-          "description" => "description",
-          "image"       => "/img.png"
-        }
-      end
-      let(:page) { make_post(meta) }
-
-      it "outputs post meta" do
-        expected = %r!<meta property="og:type" content="article" />!
-        expect(output).to match(expected)
-
-        expect(json_data["headline"]).to eql("post")
-        expect(json_data["description"]).to eql("description")
-        expect(json_data["image"]).to eql("http://example.invalid/img.png")
-      end
-
-      it "minifies JSON-LD" do
-        expected = <<-EOS
-{"@context": "http://schema.org",
-"@type": "BlogPosting",
-"headline": "post",
-EOS
-        expect(output).to match(expected)
-      end
-    end
-  end
-
-  context "facebook" do
-    let(:site_facebook) do
-      {
-        "admins"    => "jekyllrb-fb-admins",
-        "app_id"    => "jekyllrb-fb-app_id",
-        "publisher" => "jekyllrb-fb-publisher"
-      }
-    end
-
-    let(:site) { make_site("facebook" => site_facebook) }
-
-    it "outputs facebook admins meta" do
-      expected = %r!<meta property="fb:admins" content="jekyllrb-fb-admins" />!
-      expect(output).to match(expected)
-    end
-
-    it "outputs facebook app ID meta" do
-      expected = %r!<meta property="fb:app_id" content="jekyllrb-fb-app_id" />!
-      expect(output).to match(expected)
-    end
-
-    it "outputs facebook article publisher meta" do
-      expected = %r!<meta property="article:publisher" content="jekyllrb-fb-publisher" />!
-      expect(output).to match(expected)
-    end
-  end
-
-  context "twitter" do
-    context "with site.twitter.username" do
-      let(:site_twitter) { { "username" => "jekyllrb" } }
-      let(:site) { make_site("twitter" => site_twitter) }
-
-      context "with page.author as a string" do
-        let(:page) { make_page("author" => "benbalter") }
-
-        it "outputs twitter card meta" do
-          expected = %r!<meta name="twitter:card" content="summary" />!
-          expect(output).to match(expected)
-
-          expected = %r!<meta name="twitter:site" content="@jekyllrb" />!
-          expect(output).to match(expected)
-
-          expected = %r!<meta name="twitter:creator" content="@benbalter" />!
-          expect(output).to match(expected)
-        end
-
-        context "with an @" do
-          let(:page) { make_page("author" => "@benbalter") }
-
-          it "outputs the twitter card" do
-            expected = %r!<meta name="twitter:creator" content="@benbalter" />!
-            expect(output).to match(expected)
-          end
-        end
-
-        context "with site.data.authors" do
-          let(:author_data) { {} }
-          let(:data) { { "authors" => author_data } }
-          let(:site) { make_site("data" => data, "twitter" => site_twitter) }
-
-          context "with the author in site.data.authors" do
-            let(:author_data) { { "benbalter" => { "twitter" => "test" } } }
-            it "outputs the twitter card" do
-              expected = %r!<meta name="twitter:creator" content="@test" />!
-              expect(output).to match(expected)
-            end
-          end
-
-          context "without the author in site.data.authors" do
-            it "outputs the twitter card" do
-              expected = %r!<meta name="twitter:creator" content="@benbalter" />!
-              expect(output).to match(expected)
-            end
-          end
-        end
-      end
-
-      context "with page.image" do
-        let(:page) { make_page("image" => "/img/foo.png") }
-
-        it "outputs summary card with large image" do
-          expected = %r!<meta name="twitter:card" content="summary_large_image" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "with page.author as a hash" do
-        let(:page) { make_page("author" => { "twitter" => "@test" }) }
-
-        it "supports author data as a hash" do
-          expected = %r!<meta name="twitter:creator" content="@test" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "with page.authors as an array" do
-        let(:page) { make_page("authors" => %w(test foo)) }
-
-        it "supports author data as an array" do
-          expected = %r!<meta name="twitter:creator" content="@test" />!
-          expect(output).to match(expected)
-        end
-      end
-
-      context "with site.author as a hash" do
-        let(:author) { { "twitter" => "@test" } }
-        let(:site) { make_site("author" => author, "twitter" => site_twitter) }
-
-        it "supports author data as an hash" do
-          expected = %r!<meta name="twitter:creator" content="@test" />!
-          expect(output).to match(expected)
-        end
-      end
-    end
-  end
-
-  context "with site.social" do
-    let(:links) { ["http://foo.invalid", "http://bar.invalid"] }
-    let(:social_namespace) { { "name" => "Ben", "links" => links } }
-    let(:site) { make_site("social" => social_namespace) }
-    let(:page) { make_post(meta) }
-
-    context "on homepage" do
-      let(:meta) do
-        {
-          "permalink" => "/",
-          "seo"       => {
-            "type" => "person"
-          }
-        }
-      end
-
-      it "outputs social meta" do
-        expect(json_data["@type"]).to eql("person")
-        expect(json_data["name"]).to eql("Ben")
-        expect(json_data["sameAs"]).to eql(links)
-      end
-    end
-
-    context "on about page" do
-      let(:meta) { { "permalink" => "/about/" } }
-
-      it "outputs sameAs links" do
-        expect(json_data["sameAs"]).to eql(links)
-      end
-    end
-
-    context "on other pages" do
-      let(:meta) { { "permalink" => "/2/" } }
-
-      it "does not output sameAs links" do
-        expect(json_data["sameAs"]).to be_nil
-      end
-    end
-  end
-
-  context "with site.name" do
-    let(:site) { make_site("name" => "Site name") }
-
-    it "uses site.name if site.title is not present" do
-      expected = %r!<meta property="og:site_name" content="Site name" />!
-      expect(output).to match(expected)
-    end
-
-    context "with site.title" do
-      let(:site)  { make_site("name" => "Site Name", "title" => "Site Title") }
-
-      it "uses site.tile if both site.title and site.name are present" do
-        expected = %r!<meta property="og:site_name" content="Site Title" />!
-        expect(output).to match(expected)
-      end
-    end
-  end
-
-  context "with title=false" do
-    let(:text) { "title=false" }
-
-    it "does not output a <title> tag" do
-      expect(output).not_to match(%r!<title>!)
-    end
-  end
-
-  context "with pagination" do
-    let(:context) { make_context({}, "paginator" => paginator) }
-
-    it "outputs pagination links" do
-      expect(output).to match(%r!<link rel="prev" href="/foo">!)
-      expect(output).to match(%r!<link rel="next" href="/bar">!)
-    end
+  it "renders" do
+    expected = "<!-- Begin Jekyll SEO tag v#{described_class::VERSION} -->"
+    expect(rendered).to match(expected)
   end
 end

spec/spec_helper.rb

@@ -1,26 +1,37 @@
-$LOAD_PATH.unshift File.expand_path("../../lib", __FILE__)
+# frozen_string_literal: true
+
+$LOAD_PATH.unshift File.expand_path("../lib", __dir__)
 require "jekyll"
 require "jekyll-seo-tag"
-require "html/proofer"
+require "html-proofer"
+
+# Monkey patch Jekyll::Drops::Drop so Rspec's `have_key` works as expected
+module Jekyll
+  module Drops
+    class Drop
+      alias_method :has_key?, :key?
+    end
+  end
+end
 
 ENV["JEKYLL_LOG_LEVEL"] = "error"
 
 def dest_dir
-  File.expand_path("../tmp/dest", File.dirname(__FILE__))
+  File.expand_path("../tmp/dest", __dir__)
 end
 
 def source_dir
-  File.expand_path("./fixtures", File.dirname(__FILE__))
+  File.expand_path("fixtures", __dir__)
 end
 
 CONFIG_DEFAULTS = {
   "source"      => source_dir,
   "destination" => dest_dir,
-  "gems"        => ["jekyll-seo-tag"]
+  "gems"        => ["jekyll-seo-tag"],
 }.freeze
 
 def make_page(options = {})
-  page = Jekyll::Page.new site, CONFIG_DEFAULTS["source"], "", "page.md"
+  page = Jekyll::Page.new site, CONFIG_DEFAULTS["source"], options.delete("dir") || "", "page.md"
   page.data = options
   page
 end