commonmarker 0.18.2 → 0.19.0

data/ext/commonmarker/ext_scanners.h

@@ -11,11 +11,13 @@ bufsize_t _scan_table_start(const unsigned char *p);
 bufsize_t _scan_table_cell(const unsigned char *p);
 bufsize_t _scan_table_cell_end(const unsigned char *p);
 bufsize_t _scan_table_row_end(const unsigned char *p);
+bufsize_t _scan_tasklist(const unsigned char *p);
 
 #define scan_table_start(c, l, n) _ext_scan_at(&_scan_table_start, c, l, n)
 #define scan_table_cell(c, l, n) _ext_scan_at(&_scan_table_cell, c, l, n)
 #define scan_table_cell_end(c, l, n) _ext_scan_at(&_scan_table_cell_end, c, l, n)
 #define scan_table_row_end(c, l, n) _ext_scan_at(&_scan_table_row_end, c, l, n)
+#define scan_tasklist(c, l, n) _ext_scan_at(&_scan_tasklist, c, l, n)
 
 #ifdef __cplusplus
 }

data/ext/commonmarker/extconf.rb

@@ -1,5 +1,7 @@
+# frozen_string_literal: true
+
 require 'mkmf'
 
-$CFLAGS << " -std=c99"
+$CFLAGS << ' -std=c99'
 
 create_makefile('commonmarker/commonmarker')

data/ext/commonmarker/houdini_href_e.c

@@ -15,7 +15,7 @@
  *	- The characters which are *not* safe to be in
  *	an URL because they are RESERVED characters.
  *
- * We asume (lazily) that any RESERVED char that
+ * We assume (lazily) that any RESERVED char that
  * appears inside an URL is actually meant to
  * have its native function (i.e. as an URL
  * component/separator) and hence needs no escaping.

data/ext/commonmarker/inlines.c

@@ -760,9 +760,10 @@ static delimiter *S_insert_emph(subject *subj, delimiter *opener,
   }
   cmark_node_insert_after(opener_inl, emph);
 
-  emph->start_line = emph->end_line = subj->line;
-  emph->start_column = opener_inl->start_column + subj->column_offset;
-  emph->end_column = closer_inl->end_column + subj->column_offset;
+  emph->start_line = opener_inl->start_line;
+  emph->end_line = closer_inl->end_line;
+  emph->start_column = opener_inl->start_column;
+  emph->end_column = closer_inl->end_column;
 
   // if opener has 0 characters, remove it and its associated inline
   if (opener_num_chars == 0) {

data/ext/commonmarker/render.c

@@ -57,6 +57,7 @@ static void S_out(cmark_renderer *renderer, cmark_node *node,
       }
     }
     renderer->column = 0;
+    renderer->last_breakable = 0;
     renderer->begin_line = true;
     renderer->begin_content = true;
     renderer->need_cr -= 1;
@@ -188,7 +189,7 @@ char *cmark_render(cmark_mem *mem, cmark_node *root, int options, int width,
   }
 
   // ensure final newline
-  if (renderer.buffer->ptr[renderer.buffer->size - 1] != '\n') {
+  if (renderer.buffer->size == 0 || renderer.buffer->ptr[renderer.buffer->size - 1] != '\n') {
     cmark_strbuf_putc(renderer.buffer, '\n');
   }
 

data/ext/commonmarker/scanners.c

@@ -1,4 +1,4 @@
-/* Generated by re2c 1.0.3 */
+/* Generated by re2c 1.1.1 */
 #include <stdlib.h>
 #include "chunk.h"
 #include "scanners.h"
@@ -9226,7 +9226,7 @@ bufsize_t _scan_open_code_fence(const unsigned char *p)
 		144, 192, 192, 192, 192, 192, 192, 192, 
 		192, 192, 192, 192, 192, 192, 192, 192, 
 		192, 192, 192, 192, 192, 192, 192, 192, 
-		192, 192, 192, 192, 192, 192,  96, 192, 
+		192, 192, 192, 192, 192, 192, 224, 192, 
 		  0,   0,   0,   0,   0,   0,   0,   0, 
 		  0,   0,   0,   0,   0,   0,   0,   0, 
 		  0,   0,   0,   0,   0,   0,   0,   0, 

data/ext/commonmarker/table.c

@@ -114,7 +114,7 @@ static table_row *row_from_string(cmark_syntax_extension *self,
                                   cmark_parser *parser, unsigned char *string,
                                   int len) {
   table_row *row = NULL;
-  bufsize_t cell_matched, pipe_matched, offset;
+  bufsize_t cell_matched = 1, pipe_matched = 1, offset;
 
   row = (table_row *)parser->mem->calloc(1, sizeof(table_row));
   row->n_columns = 0;
@@ -122,7 +122,9 @@ static table_row *row_from_string(cmark_syntax_extension *self,
 
   offset = scan_table_cell_end(string, len, 0);
 
-  do {
+  // Parse the cells of the row. Stop if we reach the end of the input, or if we
+  // cannot detect any more cells.
+  while (offset < len && (cell_matched || pipe_matched)) {
     cell_matched = scan_table_cell(string, len, offset);
     pipe_matched = scan_table_cell_end(string, len, offset + cell_matched);
 
@@ -149,7 +151,7 @@ static table_row *row_from_string(cmark_syntax_extension *self,
       pipe_matched = scan_table_row_end(string, len, offset);
       offset += pipe_matched;
     }
-  } while ((cell_matched || pipe_matched) && offset < len);
+  }
 
   if (offset != len || !row->n_columns) {
     free_table_row(parser->mem, row);

data/ext/commonmarker/tasklist.c

@@ -0,0 +1,135 @@
+#include "tasklist.h"
+#include <parser.h>
+#include <render.h>
+#include <html.h>
+#include "ext_scanners.h"
+
+typedef enum {
+  CMARK_TASKLIST_NOCHECKED,
+  CMARK_TASKLIST_CHECKED,
+} cmark_tasklist_type;
+
+static const char *get_type_string(cmark_syntax_extension *extension, cmark_node *node) {
+  return "tasklist";
+}
+
+char *cmark_gfm_extensions_get_tasklist_state(cmark_node *node) {
+  if (!node || ((int)node->as.opaque != CMARK_TASKLIST_CHECKED && (int)node->as.opaque != CMARK_TASKLIST_NOCHECKED))
+    return 0;
+
+  if ((int)node->as.opaque != CMARK_TASKLIST_CHECKED) {
+    return "checked";
+  }
+  else {
+    return "unchecked";
+  }
+}
+
+static bool parse_node_item_prefix(cmark_parser *parser, const char *input,
+                                   cmark_node *container) {
+  bool res = false;
+
+  if (parser->indent >=
+      container->as.list.marker_offset + container->as.list.padding) {
+    cmark_parser_advance_offset(parser, input, container->as.list.marker_offset +
+                                        container->as.list.padding,
+                     true);
+    res = true;
+  } else if (parser->blank && container->first_child != NULL) {
+    // if container->first_child is NULL, then the opening line
+    // of the list item was blank after the list marker; in this
+    // case, we are done with the list item.
+    cmark_parser_advance_offset(parser, input, parser->first_nonspace - parser->offset,
+                     false);
+    res = true;
+  }
+  return res;
+}
+
+static int matches(cmark_syntax_extension *self, cmark_parser *parser,
+                   unsigned char *input, int len,
+                   cmark_node *parent_container) {
+  return parse_node_item_prefix(parser, (const char*)input, parent_container);
+}
+
+static int can_contain(cmark_syntax_extension *extension, cmark_node *node,
+                       cmark_node_type child_type) {
+  return (node->type == CMARK_NODE_ITEM) ? 1 : 0;
+}
+
+static cmark_node *open_tasklist_item(cmark_syntax_extension *self,
+                                      int indented, cmark_parser *parser,
+                                      cmark_node *parent_container,
+                                      unsigned char *input, int len) {
+  cmark_node_type node_type = cmark_node_get_type(parent_container);
+  if (node_type != CMARK_NODE_ITEM) {
+    return NULL;
+  }
+
+  bufsize_t matched = scan_tasklist(input, len, 0);
+  if (!matched) {
+    return NULL;
+  }
+
+  cmark_node_set_syntax_extension(parent_container, self);
+  cmark_parser_advance_offset(parser, (char *)input, 3, false);
+
+  if (strstr((char*)input, "[x]")) {
+    parent_container->as.opaque = (void *)CMARK_TASKLIST_CHECKED;
+  } else {
+    parent_container->as.opaque = (void *)CMARK_TASKLIST_NOCHECKED;
+  }
+
+  return NULL;
+}
+
+static void commonmark_render(cmark_syntax_extension *extension,
+                              cmark_renderer *renderer, cmark_node *node,
+                              cmark_event_type ev_type, int options) {
+  bool entering = (ev_type == CMARK_EVENT_ENTER);
+  if (entering) {
+    renderer->cr(renderer);
+    if ((int)node->as.opaque == CMARK_TASKLIST_CHECKED) {
+      renderer->out(renderer, node, "- [x] ", false, LITERAL);
+    } else {
+      renderer->out(renderer, node, "- [ ] ", false, LITERAL);
+    }
+    cmark_strbuf_puts(renderer->prefix, "  ");
+  } else {
+    cmark_strbuf_truncate(renderer->prefix, renderer->prefix->size - 2);
+    renderer->cr(renderer);
+  }
+}
+
+static void html_render(cmark_syntax_extension *extension,
+                        cmark_html_renderer *renderer, cmark_node *node,
+                        cmark_event_type ev_type, int options) {
+  bool entering = (ev_type == CMARK_EVENT_ENTER);
+  if (entering) {
+    cmark_html_render_cr(renderer->html);
+    cmark_strbuf_puts(renderer->html, "<li");
+    cmark_html_render_sourcepos(node, renderer->html, options);
+    cmark_strbuf_putc(renderer->html, '>');
+    if ((int)node->as.opaque == CMARK_TASKLIST_CHECKED) {
+      cmark_strbuf_puts(renderer->html, "<input type=\"checkbox\" checked=\"\" disabled=\"\" /> ");
+    } else {
+      cmark_strbuf_puts(renderer->html, "<input type=\"checkbox\" disabled=\"\" /> ");
+    }
+  } else {
+    cmark_strbuf_puts(renderer->html, "</li>\n");
+  }
+}
+
+cmark_syntax_extension *create_tasklist_extension(void) {
+  cmark_syntax_extension *ext = cmark_syntax_extension_new("tasklist");
+
+  cmark_syntax_extension_set_match_block_func(ext, matches);
+  cmark_syntax_extension_set_get_type_string_func(ext, get_type_string);
+  cmark_syntax_extension_set_open_block_func(ext, open_tasklist_item);
+  cmark_syntax_extension_set_can_contain_func(ext, can_contain);
+  cmark_syntax_extension_set_commonmark_render_func(ext, commonmark_render);
+  cmark_syntax_extension_set_plaintext_render_func(ext, commonmark_render);
+  cmark_syntax_extension_set_html_render_func(ext, html_render);
+
+  return ext;
+}

data/ext/commonmarker/tasklist.h

@@ -0,0 +1,8 @@
+#ifndef TASKLIST_H
+#define TASKLIST_H
+
+#include "cmark-gfm-core-extensions.h"
+
+cmark_syntax_extension *create_tasklist_extension(void);
+
+#endif

data/lib/commonmarker.rb

@@ -1,4 +1,6 @@
 #!/usr/bin/env ruby
+# frozen_string_literal: true
+
 require 'commonmarker/commonmarker'
 require 'commonmarker/config'
 require 'commonmarker/node'

data/lib/commonmarker/config.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'ruby-enum'
 module CommonMarker
   # For Ruby::Enum, these must be classes, not modules
@@ -37,7 +39,7 @@ module CommonMarker
         # neckbearding around. the map will both check the opts and then bitwise-OR it
         option.map { |o| check_option(o, type); type.to_h[o] }.inject(0, :|)
       else
-        raise TypeError, 'option type must be a valid symbol or array of symbols'
+        raise TypeError, "option type must be a valid symbol or array of symbols within the #{type} context"
       end
     end
 

data/lib/commonmarker/node.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'commonmarker/node/inspect'
 
 module CommonMarker

data/lib/commonmarker/renderer.rb

@@ -1,3 +1,7 @@
+# frozen_string_literal: true
+# rubocop:disable Standard/RailsViewRenderLiteral
+# rubocop:disable Standard/RailsControllerRenderLiteral
+
 require 'set'
 require 'stringio'
 
@@ -6,7 +10,7 @@ module CommonMarker
     attr_accessor :in_tight, :warnings, :in_plain
     def initialize(options: :DEFAULT, extensions: [])
       @opts = Config.process_options(options, :render)
-      @stream = StringIO.new("".force_encoding("utf-8"))
+      @stream = StringIO.new(''.dup.force_encoding('utf-8'))
       @need_blocksep = false
       @warnings = Set.new []
       @in_tight = false
@@ -116,7 +120,7 @@ module CommonMarker
     end
 
     def sourcepos(node)
-      return "" unless option_enabled?(:SOURCEPOS)
+      return '' unless option_enabled?(:SOURCEPOS)
       s = node.sourcepos
       " data-sourcepos=\"#{s[:start_line]}:#{s[:start_column]}-" \
         "#{s[:end_line]}:#{s[:end_column]}\""

data/lib/commonmarker/renderer/html_renderer.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module CommonMarker
   class HtmlRenderer < Renderer
     def render(node)
@@ -26,7 +28,7 @@ module CommonMarker
           container("<p#{sourcepos(node)}>", '</p>') do
             out(:children)
             if node.parent.type == :footnote_definition && node.next.nil?
-              out(" ")
+              out(' ')
               out_footnote_backref
             end
           end
@@ -60,12 +62,23 @@ module CommonMarker
 
     def list_item(node)
       block do
-        container("<li#{sourcepos(node)}>", '</li>') do
+        tasklist_data = tasklist(node)
+        container("<li#{sourcepos(node)}#{tasklist_data}>#{' ' if tasklist?(node)}", '</li>') do
           out(:children)
         end
       end
     end
 
+    def tasklist(node)
+      return '' unless tasklist?(node)
+      state = if checked?(node)
+        'disabled=""'
+      else
+        'checked="" disabled=""'
+      end
+      return "><input type=\"checkbox\" #{state} /"
+    end
+
     def blockquote(node)
       block do
         container("<blockquote#{sourcepos(node)}>\n", '</blockquote>') do
@@ -239,5 +252,14 @@ module CommonMarker
       out("<a href=\"#fnref#@footnote_ix\" class=\"footnote-backref\">↩</a>")
       true
     end
+
+    def tasklist?(node)
+      node.type_string == 'tasklist'
+    end
+
+    def checked?(node)
+      node.tasklist_state == 'checked'
+    end
+
   end
 end

data/lib/commonmarker/version.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module CommonMarker
-  VERSION = '0.18.2'.freeze
+  VERSION = '0.19.0'.freeze
 end

data/test/benchmark.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'commonmarker'
 require 'github/markdown'
 require 'redcarpet'
@@ -9,7 +11,7 @@ def dobench(name, &blk)
   puts Benchmark.measure(&blk)
 end
 
-benchinput = File.open('test/benchinput.md', 'r').read()
+benchinput = File.open('test/benchinput.md', 'r').read
 
 printf("input size = %d bytes\n\n", benchinput.bytesize)
 

data/test/test_attributes.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestAttributes < Minitest::Test
@@ -15,7 +17,7 @@ class TestAttributes < Minitest::Test
 
     sourcepos.delete_if { |h| h.values.all? { |v| v == 0 } }
 
-    result = [{:start_line=>1, :start_column=>1, :end_line=>10, :end_column=>12}, {:start_line=>1, :start_column=>1, :end_line=>1, :end_column=>17}, {:start_line=>1, :start_column=>4, :end_line=>1, :end_column=>17}, {:start_line=>3, :start_column=>1, :end_line=>5, :end_column=>36}, {:start_line=>3, :start_column=>1, :end_line=>3, :end_column=>55}, {:start_line=>4, :start_column=>1, :end_line=>4, :end_column=>53}, {:start_line=>4, :start_column=>2, :end_line=>4, :end_column=>14}, {:start_line=>4, :start_column=>54, :end_line=>4, :end_column=>58}, {:start_line=>5, :start_column=>1, :end_line=>5, :end_column=>36}, {:start_line=>7, :start_column=>1, :end_line=>10, :end_column=>12}, {:start_line=>7, :start_column=>1, :end_line=>7, :end_column=>11}, {:start_line=>7, :start_column=>4, :end_line=>7, :end_column=>11}, {:start_line=>7, :start_column=>4, :end_line=>7, :end_column=>11}, {:start_line=>8, :start_column=>1, :end_line=>10, :end_column=>12}, {:start_line=>8, :start_column=>4, :end_line=>8, :end_column=>11}, {:start_line=>8, :start_column=>4, :end_line=>8, :end_column=>11}, {:start_line=>9, :start_column=>4, :end_line=>10, :end_column=>12}, {:start_line=>9, :start_column=>4, :end_line=>9, :end_column=>12}, {:start_line=>9, :start_column=>6, :end_line=>9, :end_column=>12}, {:start_line=>9, :start_column=>6, :end_line=>9, :end_column=>12}, {:start_line=>10, :start_column=>4, :end_line=>10, :end_column=>12}, {:start_line=>10, :start_column=>6, :end_line=>10, :end_column=>12}, {:start_line=>10, :start_column=>6, :end_line=>10, :end_column=>12}]
+    result = [{start_line: 1, start_column: 1, end_line: 10, end_column: 12}, {start_line: 1, start_column: 1, end_line: 1, end_column: 17}, {start_line: 1, start_column: 4, end_line: 1, end_column: 17}, {start_line: 3, start_column: 1, end_line: 5, end_column: 36}, {start_line: 3, start_column: 1, end_line: 3, end_column: 55}, {start_line: 4, start_column: 1, end_line: 4, end_column: 53}, {start_line: 4, start_column: 2, end_line: 4, end_column: 14}, {start_line: 4, start_column: 54, end_line: 4, end_column: 58}, {start_line: 5, start_column: 1, end_line: 5, end_column: 36}, {start_line: 7, start_column: 1, end_line: 10, end_column: 12}, {start_line: 7, start_column: 1, end_line: 7, end_column: 11}, {start_line: 7, start_column: 4, end_line: 7, end_column: 11}, {start_line: 7, start_column: 4, end_line: 7, end_column: 11}, {start_line: 8, start_column: 1, end_line: 10, end_column: 12}, {start_line: 8, start_column: 4, end_line: 8, end_column: 11}, {start_line: 8, start_column: 4, end_line: 8, end_column: 11}, {start_line: 9, start_column: 4, end_line: 10, end_column: 12}, {start_line: 9, start_column: 4, end_line: 9, end_column: 12}, {start_line: 9, start_column: 6, end_line: 9, end_column: 12}, {start_line: 9, start_column: 6, end_line: 9, end_column: 12}, {start_line: 10, start_column: 4, end_line: 10, end_column: 12}, {start_line: 10, start_column: 6, end_line: 10, end_column: 12}, {start_line: 10, start_column: 6, end_line: 10, end_column: 12}]
 
     assert_equal result, sourcepos
   end

data/test/test_basics.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestBasics < Minitest::Test

data/test/test_commonmark.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestCommonmark < Minitest::Test

data/test/test_doc.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestDocNode < Minitest::Test

data/test/test_encoding.rb

@@ -1,4 +1,6 @@
 # coding: utf-8
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestEncoding < Minitest::Test

data/test/test_extensions.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'test_helper'
 
 class TestExtensions < Minitest::Test
@@ -18,30 +20,30 @@ Another extension:
 
   def test_uses_specified_extensions
     CommonMarker.render_html(@markdown, :DEFAULT, %i[]).tap do |out|
-      assert out.include?("| a")
-      assert out.include?("| <strong>x</strong>")
-      assert out.include?("~~hi~~")
+      assert out.include?('| a')
+      assert out.include?('| <strong>x</strong>')
+      assert out.include?('~~hi~~')
     end
 
     CommonMarker.render_html(@markdown, :DEFAULT, %i[table]).tap do |out|
-      refute out.include?("| a")
-      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each {|html| assert out.include?(html) }
-      assert out.include?("~~hi~~")
+      refute out.include?('| a')
+      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each { |html| assert out.include?(html) }
+      assert out.include?('~~hi~~')
     end
 
     CommonMarker.render_html(@markdown, :DEFAULT, %i[strikethrough]).tap do |out|
-      assert out.include?("| a")
-      refute out.include?("~~hi~~")
-      assert out.include?("<del>hi</del>")
+      assert out.include?('| a')
+      refute out.include?('~~hi~~')
+      assert out.include?('<del>hi</del>')
     end
 
-    doc = CommonMarker.render_doc("~a~ ~~b~~ ~~~c~~~", :STRIKETHROUGH_DOUBLE_TILDE, [:strikethrough])
+    doc = CommonMarker.render_doc('~a~ ~~b~~ ~~~c~~~', :STRIKETHROUGH_DOUBLE_TILDE, [:strikethrough])
     assert_equal doc.to_html, "<p>~a~ <del>b</del> ~~~c~~~</p>\n"
 
     CommonMarker.render_html(@markdown, :DEFAULT, %i[table strikethrough]).tap do |out|
-      refute out.include?("| a")
-      refute out.include?("| <strong>x</strong>")
-      refute out.include?("~~hi~~")
+      refute out.include?('| a')
+      refute out.include?('| <strong>x</strong>')
+      refute out.include?('~~hi~~')
     end
 
   end
@@ -50,24 +52,24 @@ Another extension:
     doc = CommonMarker.render_doc(@markdown, :DEFAULT, %i[table])
 
     doc.to_html.tap do |out|
-      refute out.include?("| a")
-      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each {|html| assert out.include?(html) }
-      assert out.include?("~~hi~~")
+      refute out.include?('| a')
+      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each { |html| assert out.include?(html) }
+      assert out.include?('~~hi~~')
     end
 
     HtmlRenderer.new.render(doc).tap do |out|
-      refute out.include?("| a")
-      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each {|html| assert out.include?(html) }
-      assert out.include?("~~hi~~")
+      refute out.include?('| a')
+      %w(<table> <tr> <th> a </th> <td> c </td> <strong>x</strong>).each { |html| assert out.include?(html) }
+      assert out.include?('~~hi~~')
     end
 
-    doc = CommonMarker.render_doc("~a~ ~~b~~ ~~~c~~~", :STRIKETHROUGH_DOUBLE_TILDE, [:strikethrough])
+    doc = CommonMarker.render_doc('~a~ ~~b~~ ~~~c~~~', :STRIKETHROUGH_DOUBLE_TILDE, [:strikethrough])
     assert_equal HtmlRenderer.new.render(doc), "<p>~a~ <del>b</del> ~~~c~~~</p>\n"
   end
 
   def test_bad_extension_specifications
-    assert_raises(TypeError) { CommonMarker.render_html(@markdown, :DEFAULT, "nope") }
-    assert_raises(TypeError) { CommonMarker.render_html(@markdown, :DEFAULT, ["table"]) }
+    assert_raises(TypeError) { CommonMarker.render_html(@markdown, :DEFAULT, 'nope') }
+    assert_raises(TypeError) { CommonMarker.render_html(@markdown, :DEFAULT, ['table']) }
     assert_raises(ArgumentError) { CommonMarker.render_html(@markdown, :DEFAULT, %i[table bad]) }
   end