Add the methods for rendering XML from cmark-gfm

.gitignore

@@ -4,6 +4,7 @@
 /.bundle
 /.config
 /coverage/
+/.idea
 /InstalledFiles
 /pkg/
 /spec/reports/

bin/commonmarker

@@ -15,13 +15,16 @@ def parse_options
   extensions = CommonMarker.extensions
   parse_options = CommonMarker::Config::OPTS.fetch(:parse)
   render_options = CommonMarker::Config::OPTS.fetch(:render)
+  format_options = CommonMarker::Config::OPTS.fetch(:format)
 
   options.active_extensions = []
   options.active_parse_options = [:DEFAULT]
   options.active_render_options = [:DEFAULT]
+  options.output_format = :html
 
   option_parser = OptionParser.new do |opts|
     opts.banner = 'Usage: commonmarker [--html-renderer] [--extension=EXTENSION]'
+    opts.separator '                    [--to=FORMAT]'
     opts.separator '                    [--parse-option=OPTION]'
     opts.separator '                    [--render-option=OPTION]'
     opts.separator '                    [FILE..]'
@@ -43,6 +46,7 @@ def parse_options
     opts.on('-h', '--help', 'Prints this help') do
       puts opts
       puts
+      puts "Available formats: #{format_options.join(', ')}"
       puts "Available extentions: #{extensions.join(', ')}"
       puts "Available parse options: #{parse_options.keys.join(', ')}"
       puts "Available render options: #{render_options.keys.join(', ')}"
@@ -51,7 +55,12 @@ def parse_options
       exit
     end
 
-    opts.on('--html-renderer', 'Use the HtmlRenderer renderer rather than the native C renderer') do
+    opts.on('-tFORMAT', '--to=FORMAT', String, 'Specify output format (html, xml)') do |value|
+      value = value.to_sym
+      options.output_format = value if format_options.include?(value)
+    end
+
+    opts.on('--html-renderer', 'Use the HtmlRenderer renderer rather than the native C renderer (only valid when format is html)') do
       options.renderer = true
     end
 
@@ -87,12 +96,16 @@ def parse_options
 end
 
 options = parse_options
-
 doc = CommonMarker.render_doc(ARGF.read, options.active_parse_options, options.active_extensions)
 
-if options.renderer
-  renderer = CommonMarker::HtmlRenderer.new(extensions: options.active_extensions)
-  $stdout.write(renderer.render(doc))
-else
-  $stdout.write(doc.to_html(options.active_render_options, options.active_extensions))
+case options.output_format
+when :html
+  if options.renderer
+    renderer = CommonMarker::HtmlRenderer.new(options: options.active_render_options, extensions: options.active_extensions)
+    $stdout.write(renderer.render(doc))
+  else
+    $stdout.write(doc.to_html(options.active_render_options, options.active_extensions))
+  end
+when :xml
+  $stdout.write(doc.to_xml(options.active_render_options))
 end

ext/commonmarker/commonmarker.c

@@ -186,6 +186,40 @@ static VALUE rb_markdown_to_html(VALUE self, VALUE rb_text, VALUE rb_options, VA
   return ruby_html;
 }
 
+/*
+ * Internal: Parses a Markdown string into an HTML string.
+ *
+ */
+static VALUE rb_markdown_to_xml(VALUE self, VALUE rb_text, VALUE rb_options, VALUE rb_extensions) {
+  char *str, *xml;
+  int len;
+  cmark_parser *parser;
+  cmark_node *doc;
+  Check_Type(rb_text, T_STRING);
+  Check_Type(rb_options, T_FIXNUM);
+
+  parser = prepare_parser(rb_options, rb_extensions, cmark_get_arena_mem_allocator());
+
+  str = (char *)RSTRING_PTR(rb_text);
+  len = RSTRING_LEN(rb_text);
+
+  cmark_parser_feed(parser, str, len);
+  doc = cmark_parser_finish(parser);
+  if (doc == NULL) {
+    cmark_arena_reset();
+    rb_raise(rb_eNodeError, "error parsing document");
+  }
+
+  cmark_mem *default_mem = cmark_get_default_mem_allocator();
+  xml = cmark_render_xml_with_mem(doc, FIX2INT(rb_options), default_mem);
+  cmark_arena_reset();
+
+  VALUE ruby_xml = rb_str_new2(xml);
+  default_mem->free(xml);
+
+  return ruby_xml;
+}
+
 /*
  * Internal: Creates a node based on a node type.
  *
@@ -574,6 +608,28 @@ static VALUE rb_render_html(VALUE self, VALUE rb_options, VALUE rb_extensions) {
   return ruby_html;
 }
 
+/* Internal: Convert the node to an XML string.
+ *
+ * Returns a {String}.
+ */
+static VALUE rb_render_xml(VALUE self, VALUE rb_options) {
+  int options;
+  int i;
+  cmark_node *node;
+  Check_Type(rb_options, T_FIXNUM);
+
+  options = FIX2INT(rb_options);
+
+  Data_Get_Struct(self, cmark_node, node);
+
+  char *xml = cmark_render_xml(node, options);
+  VALUE ruby_xml = rb_str_new2(xml);
+
+  free(xml);
+
+  return ruby_xml;
+}
+
 /* Internal: Convert the node to a CommonMark string.
  *
  * Returns a {String}.
@@ -1216,6 +1272,8 @@ __attribute__((visibility("default"))) void Init_commonmarker() {
   rb_cNode = rb_define_class_under(module, "Node", rb_cObject);
   rb_define_singleton_method(rb_cNode, "markdown_to_html", rb_markdown_to_html,
                              3);
+  rb_define_singleton_method(rb_cNode, "markdown_to_xml", rb_markdown_to_xml,
+                             3);
   rb_define_singleton_method(rb_cNode, "new", rb_node_new, 1);
   rb_define_singleton_method(rb_cNode, "parse_document", rb_parse_document, 4);
   rb_define_method(rb_cNode, "string_content", rb_node_get_string_content, 0);
@@ -1228,6 +1286,7 @@ __attribute__((visibility("default"))) void Init_commonmarker() {
   rb_define_method(rb_cNode, "next", rb_node_next, 0);
   rb_define_method(rb_cNode, "insert_before", rb_node_insert_before, 1);
   rb_define_method(rb_cNode, "_render_html", rb_render_html, 2);
+  rb_define_method(rb_cNode, "_render_xml", rb_render_xml, 1);
   rb_define_method(rb_cNode, "_render_commonmark", rb_render_commonmark, -1);
   rb_define_method(rb_cNode, "_render_plaintext", rb_render_plaintext, -1);
   rb_define_method(rb_cNode, "insert_after", rb_node_insert_after, 1);

lib/commonmarker/config.rb

@@ -24,7 +24,8 @@ module Config
         FULL_INFO_STRING: (1 << 16),
         UNSAFE: (1 << 17),
         FOOTNOTES: (1 << 13)
-      }.freeze
+      }.freeze,
+      format: %i[html xml].freeze
     }.freeze
 
     def self.process_options(option, type)

lib/commonmarker/node.rb

@@ -30,6 +30,16 @@ def to_html(options = :DEFAULT, extensions = [])
       _render_html(opts, extensions).force_encoding('utf-8')
     end
 
+    # Public: Convert the node to an XML string.
+    #
+    # options - A {Symbol} or {Array of Symbol}s indicating the render options
+    #
+    # Returns a {String}.
+    def to_xml(options = :DEFAULT)
+      opts = Config.process_options(options, :render)
+      _render_xml(opts).force_encoding('utf-8')
+    end
+
     # Public: Convert the node to a CommonMark string.
     #
     # options - A {Symbol} or {Array of Symbol}s indicating the render options

test/test_commands.rb

@@ -28,4 +28,10 @@ def test_understands_multiple_extensions
     assert_includes out, '<p><del>hi</del>'
     %w[<table> <tr> <th> a </th> <td> c </td>].each { |html| assert_includes out, html }
   end
+
+  def test_understands_format
+    out = make_bin('strong.md', '--to=xml')
+    assert_includes out, '<?xml version="1.0" encoding="UTF-8"?>'
+    assert_includes out, '<text xml:space="preserve">strong</text>'
+  end
 end

test/test_xml.rb

@@ -0,0 +1,93 @@
+# frozen_string_literal: true
+
+require 'test_helper'
+
+class TestXml < Minitest::Test
+  def setup
+    @markdown = <<~MD
+      Hi *there*!
+
+      1. I am a numeric list.
+      2. I continue the list.
+      * Suddenly, an unordered list!
+      * What fun!
+
+      Okay, _enough_.
+
+      | a   | b   |
+      | --- | --- |
+      | c   | d   |
+    MD
+  end
+
+  def render_doc(doc)
+    CommonMarker.render_doc(doc, :DEFAULT, [:table])
+  end
+
+  def test_to_commonmark
+    compare = render_doc(@markdown).to_xml(:SOURCEPOS)
+
+    assert_equal <<~XML, compare
+      <?xml version="1.0" encoding="UTF-8"?>
+      <!DOCTYPE document SYSTEM "CommonMark.dtd">
+      <document sourcepos="1:1-12:13" xmlns="http://commonmark.org/xml/1.0">
+        <paragraph sourcepos="1:1-1:11">
+          <text sourcepos="1:1-1:3" xml:space="preserve">Hi </text>
+          <emph sourcepos="1:4-1:10">
+            <text sourcepos="1:5-1:9" xml:space="preserve">there</text>
+          </emph>
+          <text sourcepos="1:11-1:11" xml:space="preserve">!</text>
+        </paragraph>
+        <list sourcepos="3:1-4:23" type="ordered" start="1" delim="period" tight="true">
+          <item sourcepos="3:1-3:23">
+            <paragraph sourcepos="3:4-3:23">
+              <text sourcepos="3:4-3:23" xml:space="preserve">I am a numeric list.</text>
+            </paragraph>
+          </item>
+          <item sourcepos="4:1-4:23">
+            <paragraph sourcepos="4:4-4:23">
+              <text sourcepos="4:4-4:23" xml:space="preserve">I continue the list.</text>
+            </paragraph>
+          </item>
+        </list>
+        <list sourcepos="5:1-7:0" type="bullet" tight="true">
+          <item sourcepos="5:1-5:30">
+            <paragraph sourcepos="5:3-5:30">
+              <text sourcepos="5:3-5:30" xml:space="preserve">Suddenly, an unordered list!</text>
+            </paragraph>
+          </item>
+          <item sourcepos="6:1-7:0">
+            <paragraph sourcepos="6:3-6:11">
+              <text sourcepos="6:3-6:11" xml:space="preserve">What fun!</text>
+            </paragraph>
+          </item>
+        </list>
+        <paragraph sourcepos="8:1-8:15">
+          <text sourcepos="8:1-8:6" xml:space="preserve">Okay, </text>
+          <emph sourcepos="8:7-8:14">
+            <text sourcepos="8:8-8:13" xml:space="preserve">enough</text>
+          </emph>
+          <text sourcepos="8:15-8:15" xml:space="preserve">.</text>
+        </paragraph>
+        <table sourcepos="10:1-12:13">
+          <table_header sourcepos="10:1-10:13">
+            <table_cell sourcepos="10:2-10:6">
+              <text sourcepos="10:3-10:3" xml:space="preserve">a</text>
+            </table_cell>
+            <table_cell sourcepos="10:8-10:12">
+              <text sourcepos="10:9-10:9" xml:space="preserve">b</text>
+            </table_cell>
+          </table_header>
+          <table_row sourcepos="12:1-12:13">
+            <table_cell sourcepos="12:2-12:6">
+              <text sourcepos="12:3-12:3" xml:space="preserve">c</text>
+            </table_cell>
+            <table_cell sourcepos="12:8-12:12">
+              <text sourcepos="12:9-12:9" xml:space="preserve">d</text>
+            </table_cell>
+          </table_row>
+        </table>
+      </document>
+    XML
+  end
+end