Finish 2.2.2

ruby-rdf · Oct 10, 2017 · 59b6526 · 59b6526
2 parents 72f648d + 5181927
commit 59b6526
Show file tree

Hide file tree

Showing 19 changed files with 2,653 additions and 179 deletions.
diff --git a/Gemfile b/Gemfile
@@ -2,18 +2,19 @@ source "http://rubygems.org"
 
 gemspec
 
-gem "rdf",            github: "ruby-rdf/rdf",      branch: "develop"
-gem "rdf-rdfa",       github: "ruby-rdf/rdf-rdfa", branch: "develop"
-gem "rdf-xsd",        github: "ruby-rdf/rdf-xsd",  branch: "develop"
+gem "rdf",            git: "https://github.com/ruby-rdf/rdf",      branch: "develop"
+gem "rdf-rdfa",       git: "https://github.com/ruby-rdf/rdf-rdfa", branch: "develop"
+gem "rdf-xsd",        git: "https://github.com/ruby-rdf/rdf-xsd",  branch: "develop"
 gem "nokogumbo",      '~> 1.4'
 
 group :development do
-  gem 'ebnf',               github: "gkellogg/ebnf",                branch: "develop"
-  gem 'rdf-aggregate-repo', github: "ruby-rdf/rdf-aggregate-repo",  branch: "develop"
-  gem 'rdf-isomorphic',     github: "ruby-rdf/rdf-isomorphic",      branch: "develop"
-  gem "rdf-spec",           github: "ruby-rdf/rdf-spec",            branch: "develop"
-  gem 'rdf-turtle',         github: "ruby-rdf/rdf-turtle",          branch: "develop"
-  gem 'sxp',                github: "dryruby/sxp.rb",               branch: "develop"
+  gem 'linkeddata'
+  gem 'ebnf',               git: "https://github.com/gkellogg/ebnf",                branch: "develop"
+  gem 'rdf-aggregate-repo', git: "https://github.com/ruby-rdf/rdf-aggregate-repo",  branch: "develop"
+  gem 'rdf-isomorphic',     git: "https://github.com/ruby-rdf/rdf-isomorphic",      branch: "develop"
+  gem "rdf-spec",           git: "https://github.com/ruby-rdf/rdf-spec",            branch: "develop"
+  gem 'rdf-turtle',         git: "https://github.com/ruby-rdf/rdf-turtle",          branch: "develop"
+  gem 'sxp',                git: "https://github.com/dryruby/sxp.rb",               branch: "develop"
 end
 
 group :debug do

diff --git a/README.md b/README.md
@@ -60,11 +60,24 @@ Full documentation available on [Rubydoc.info][Microdata doc]
 * {RDF::Microdata::Reader}
   * {RDF::Microdata::Reader::Nokogiri}
 
-### Additional vocabularies
+
+### RDFa-based Reader
+There is an experimental reader based on transforming Microdata to RDFa within the DOM. To invoke
+this, add the `rdfa: true` option to the {RDF::Microdata::Reader.new}, or
+use {RDF::Microdata::RdfaReader} directly.
+
+The reader exposes a `#rdfa` method, which can be used to retrieve the transformed HTML+RDFa
+
+### JSON-lD-based Reader
+There is an experimental reader based on transforming Microdata to JSON-LD. To invoke
+this, add the `jsonld: true` option to the {RDF::Microdata::Reader.new}, or
+use {RDF::Microdata::JsonLdReader} directly.
+
+The reader exposes a `#json` method, which can be used to retrieve the generated JSON-LD
 
 ## Resources
 * [RDF.rb][RDF.rb]
-* [Documentation](http://rdf.rubyforge.org/microdata)
+* [Documentation](http://www.rubydoc.info/github/ruby-rdf/rdf-microdata/)
 * [History](file:History.md)
 * [Microdata][]
 * [Microdata RDF][]

diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-2.2.1
+2.2.2
diff --git a/examples/blog_posting.html b/examples/blog_posting.html
@@ -0,0 +1,34 @@
+<!DOCTYPE HTML>
+<title>My Blog</title>
+<article itemscope itemtype="https://schema.org/BlogPosting">
+ <header>
+  <h1 itemprop="headline">Progress report</h1>
+  <p><time itemprop="datePublished" datetime="2013-08-29">today</time></p>
+  <link itemprop="url" href="?comments=0">
+ </header>
+ <p>All in all, he's doing well with his swim lessons. The biggest thing was he had trouble
+ putting his head in, but we got it down.</p>
+ <section>
+  <h1>Comments</h1>
+  <article itemprop="comment" itemscope itemtype="https://schema.org/Comment" id="c1">
+   <link itemprop="url" href="#c1">
+   <footer>
+    <p>Posted by: <span itemprop="creator" itemscope itemtype="https://schema.org/Person">
+     <span itemprop="name">Greg</span>
+    </span></p>
+    <p><time itemprop="dateCreated" datetime="2013-08-29">15 minutes ago</time></p>
+   </footer>
+   <p>Ha!</p>
+  </article>
+  <article itemprop="comment" itemscope itemtype="https://schema.org/Comment" id="c2">
+   <link itemprop="url" href="#c2">
+   <footer>
+    <p>Posted by: <span itemprop="creator" itemscope itemtype="https://schema.org/Person">
+     <span itemprop="name">Charlotte</span>
+    </span></p>
+    <p><time itemprop="dateCreated" datetime="2013-08-29">5 minutes ago</time></p>
+   </footer>
+   <p>When you say "we got it down"...</p>
+  </article>
+ </section>
+</article>
diff --git a/examples/itemref.html b/examples/itemref.html
@@ -0,0 +1,7 @@
+<div id="x">
+ <p itemprop="a">1</p>
+</div>
+<div itemscope itemref="x" itemtype="http://schema.org/Thing">
+ <p itemprop="b">test</p>
+ <p itemprop="a">2</p>
+</div>
diff --git a/examples/locomotive.html b/examples/locomotive.html
@@ -0,0 +1,11 @@
+<dl itemscope itemtype="http://md.example.com/loco
+                        http://md.example.com/lighting">
+ <dt>Name:
+ <dd itemprop="name">Tank Locomotive (DB 80)
+ <dt>Product code:
+ <dd itemprop="product-code">33041
+ <dt>Scale:
+ <dd itemprop="scale">HO
+ <dt>Digital:
+ <dd itemprop="digital">Delta
+</dl>
diff --git a/examples/to_jsonld.html b/examples/to_jsonld.html
@@ -0,0 +1,34 @@
+<!DOCTYPE HTML>
+<title>My Blog</title>
+<article itemscope itemtype="https://schema.org/BlogPosting">
+ <header>
+  <h1 itemprop="headline">Progress report</h1>
+  <p><time itemprop="datePublished" datetime="2013-08-29">today</time></p>
+  <link itemprop="url" href="?comments=0">
+ </header>
+ <p>All in all, he's doing well with his swim lessons. The biggest thing was he had trouble
+ putting his head in, but we got it down.</p>
+ <section>
+  <h1>Comments</h1>
+  <article itemprop="comment" itemscope itemtype="https://schema.org/Comment" id="c1">
+   <link itemprop="url" href="#c1">
+   <footer>
+    <p>Posted by: <span itemprop="creator" itemscope itemtype="https://schema.org/Person">
+     <span itemprop="name">Greg</span>
+    </span></p>
+    <p><time itemprop="dateCreated" datetime="2013-08-29">15 minutes ago</time></p>
+   </footer>
+   <p>Ha!</p>
+  </article>
+  <article itemprop="comment" itemscope itemtype="https://schema.org/Comment" id="c2">
+   <link itemprop="url" href="#c2">
+   <footer>
+    <p>Posted by: <span itemprop="creator" itemscope itemtype="https://schema.org/Person">
+     <span itemprop="name">Charlotte</span>
+    </span></p>
+    <p><time itemprop="dateCreated" datetime="2013-08-29">5 minutes ago</time></p>
+   </footer>
+   <p>When you say "we got it down"...</p>
+  </article>
+ </section>
+</article>
diff --git a/lib/rdf/microdata.rb b/lib/rdf/microdata.rb
@@ -15,18 +15,22 @@ module RDF
   #     end
   #   end
   #
-  # @see http://rdf.rubyforge.org/
+  # @see http://www.rubydoc.info/github/ruby-rdf/rdf/
   # @see http://www.w3.org/TR/2011/WD-microdata-20110525/
   #
   # @author [Gregg Kellogg](http://greggkellogg.net/)
   module Microdata
     USES_VOCAB = RDF::URI("http://www.w3.org/ns/rdfa#usesVocabulary")
+    DEFAULT_REGISTRY = File.expand_path("../../../etc/registry.json", __FILE__)
 
     require 'rdf/microdata/format'
     require 'rdf/microdata/vocab'
-    autoload :Expansion,  'rdf/microdata/expansion'
-    autoload :Profile,    'rdf/microdata/profile'
-    autoload :Reader,     'rdf/microdata/reader'
-    autoload :VERSION,    'rdf/microdata/version'
+    autoload :Expansion,    'rdf/microdata/expansion'
+    autoload :JsonLdReader, 'rdf/microdata/jsonld_reader'
+    autoload :Profile,      'rdf/microdata/profile'
+    autoload :RdfaReader,   'rdf/microdata/rdfa_reader'
+    autoload :Reader,       'rdf/microdata/reader'
+    autoload :Registry,     'rdf/microdata/registry'
+    autoload :VERSION,      'rdf/microdata/version'
   end
 end
diff --git a/lib/rdf/microdata/format.rb b/lib/rdf/microdata/format.rb
@@ -41,5 +41,117 @@ class Format < RDF::Format
     def self.detect(sample)
       !!sample.match(/<[^>]*(itemprop|itemtype|itemref|itemscope|itemid)[^>]*>/m)
     end
+
+    ##
+    # Hash of CLI commands appropriate for this format
+    # @return [Hash{Symbol => Hash}]
+    def self.cli_commands
+      {
+        "to-rdfa": {
+          description: "Transform HTML+Microdata into HTML+RDFa",
+          parse: false,
+          help: "to-rdfa files ...\nTransform HTML+Microdata into HTML+RDFa",
+          filter: {
+            format: :microdata
+          },
+          option_use: {output_format: :disabled},
+          lambda: ->(files, options) do
+            out = options[:output] || $stdout
+            xsl = Nokogiri::XSLT(%(<xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
+              <xsl:param name="indent-increment" select="'  '"/>
+              <xsl:output method="html" doctype-system="about:legacy-compat"/>
+ 
+              <xsl:template name="newline">
+                <xsl:text disable-output-escaping="yes">
+            </xsl:text>
+              </xsl:template>
+ 
+              <xsl:template match="comment() | processing-instruction()">
+                <xsl:param name="indent" select="''"/>
+                <xsl:call-template name="newline"/>
+                <xsl:value-of select="$indent"/>
+                <xsl:copy />
+              </xsl:template>
+ 
+              <xsl:template match="text()">
+                <xsl:param name="indent" select="''"/>
+                <xsl:call-template name="newline"/>
+                <xsl:value-of select="$indent"/>
+                <xsl:value-of select="normalize-space(.)"/>
+              </xsl:template>
+ 
+              <xsl:template match="text()[normalize-space(.)='']"/>
+ 
+              <xsl:template match="*">
+                <xsl:param name="indent" select="''"/>
+                <xsl:call-template name="newline"/>
+                <xsl:value-of select="$indent"/>
+                  <xsl:choose>
+                   <xsl:when test="count(child::*) > 0">
+                    <xsl:copy>
+                     <xsl:copy-of select="@*"/>
+                     <xsl:apply-templates select="*|text()">
+                       <xsl:with-param name="indent" select="concat ($indent, $indent-increment)"/>
+                     </xsl:apply-templates>
+                     <xsl:call-template name="newline"/>
+                     <xsl:value-of select="$indent"/>
+                    </xsl:copy>
+                   </xsl:when>
+                   <xsl:otherwise>
+                    <xsl:copy-of select="."/>
+                   </xsl:otherwise>
+                 </xsl:choose>
+              </xsl:template>
+            </xsl:stylesheet>).gsub(/^            /, ''))
+            if files.empty?
+              # If files are empty, either use options[::evaluate]
+              input = options[:evaluate] ? StringIO.new(options[:evaluate]) : STDIN
+              input.set_encoding(options.fetch(:encoding, Encoding::UTF_8))
+              RDF::Microdata::Reader.new(input, options.merge(rdfa: true)) do |reader|
+                reader.rdfa.xpath("//text()").each do |txt|
+                  txt.content = txt.content.to_s.strip
+                end
+                out.puts xsl.apply_to(reader.rdfa).to_s
+              end
+            else
+              files.each do |file|
+                RDF::Microdata::Reader.open(file, options.merge(rdfa: true)) do |reader|
+                  reader.rdfa.xpath("//text()").each do |txt|
+                    txt.content = txt.content.to_s.strip
+                  end
+                  out.puts xsl.apply_to(reader.rdfa).to_s
+                end
+              end
+            end
+          end
+        },
+        "to-jsonld": {
+          description: "Transform HTML+Microdata into JSON-LD",
+          parse: false,
+          help: "to-jsonld files ...\nTransform HTML+Microdata into JSON-LD",
+          filter: {
+            format: :microdata
+          },
+          option_use: {output_format: :disabled},
+          lambda: ->(files, options) do
+            out = options[:output] || $stdout
+            if files.empty?
+              # If files are empty, either use options[::evaluate]
+              input = options[:evaluate] ? StringIO.new(options[:evaluate]) : STDIN
+              input.set_encoding(options.fetch(:encoding, Encoding::UTF_8))
+              RDF::Microdata::Reader.new(input, options.merge(jsonld: true)) do |reader|
+                out.puts reader.jsonld.to_json(::JSON::LD::JSON_STATE)
+              end
+            else
+              files.each do |file|
+                RDF::Microdata::Reader.open(file, options.merge(jsonld: true)) do |reader|
+                  out.puts reader.jsonld.to_json(::JSON::LD::JSON_STATE)
+                end
+              end
+            end
+          end
+        },
+      }
+    end
   end
 end