Merge pull request #274 from github/feature/add-xmpp-support

Added xmpp and mailto support to the autoprefixer extension

CMakeLists.txt

@@ -4,7 +4,7 @@ project(cmark-gfm)
 set(PROJECT_VERSION_MAJOR 0)
 set(PROJECT_VERSION_MINOR 29)
 set(PROJECT_VERSION_PATCH 0)
-set(PROJECT_VERSION_GFM 4)
+set(PROJECT_VERSION_GFM 5)
 set(PROJECT_VERSION ${PROJECT_VERSION_MAJOR}.${PROJECT_VERSION_MINOR}.${PROJECT_VERSION_PATCH}.gfm.${PROJECT_VERSION_GFM})
 
 include("FindAsan.cmake")

changelog.txt

@@ -1,3 +1,6 @@
+[0.29.0.gfm.5]
+  * Added xmpp: and mailto: support to the autolink extension
+
 [0.29.0.gfm.4]
   * Remove `source` from list of HTML blocks
 

extensions/autolink.c

@@ -269,6 +269,22 @@ static cmark_node *match(cmark_syntax_extension *ext, cmark_parser *parser,
   // inline was finished in inlines.c.
 }
 
+static bool validate_protocol(char protocol[], uint8_t *data, int rewind) {
+  size_t len = strlen(protocol);
+
+  // Check that the protocol matches
+  for (int i = 1; i <= len; i++) {
+    if (data[-rewind - i] != protocol[len - i]) {
+      return false;
+    }
+  }
+
+  char prev_char = data[-rewind - len - 1];
+
+  // Make sure the character before the protocol is non-alphanumeric
+  return !cmark_isalnum(prev_char);
+}
+
 static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset, int depth) {
   // postprocess_text can recurse very deeply if there is a very long line of
   // '@' only.  Stop at a reasonable depth to ensure it cannot crash.
@@ -278,6 +294,8 @@ static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset,
   uint8_t *data = text->as.literal.data,
     *at;
   size_t size = text->as.literal.len;
+  bool auto_mailto = true;
+  bool is_xmpp = false;
   int rewind, max_rewind,
       nb = 0, np = 0, ns = 0;
 
@@ -304,8 +322,18 @@ static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset,
     if (strchr(".+-_", c) != NULL)
       continue;
 
-    if (c == '/')
-      ns++;
+    if (strchr(":", c) != NULL) {
+      if (validate_protocol("mailto:", data, rewind)) {
+        auto_mailto = false;
+        continue;
+      }
+
+      if (validate_protocol("xmpp:", data, rewind)) {
+        auto_mailto = false;
+        is_xmpp = true;
+        continue;
+      }
+    }
 
     break;
   }
@@ -325,6 +353,8 @@ static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset,
       nb++;
     else if (c == '.' && link_end < size - 1 && cmark_isalnum(data[link_end + 1]))
       np++;
+    else if (c == '/' && is_xmpp)
+      continue;
     else if (c != '-' && c != '_')
       break;
   }
@@ -347,7 +377,8 @@ static void postprocess_text(cmark_parser *parser, cmark_node *text, int offset,
   cmark_node *link_node = cmark_node_new_with_mem(CMARK_NODE_LINK, parser->mem);
   cmark_strbuf buf;
   cmark_strbuf_init(parser->mem, &buf, 10);
-  cmark_strbuf_puts(&buf, "mailto:");
+  if (auto_mailto)
+    cmark_strbuf_puts(&buf, "mailto:");
   cmark_strbuf_put(&buf, data - rewind, (bufsize_t)(link_end + rewind));
   link_node->as.link.url = cmark_chunk_buf_detach(&buf);
 

test/extensions.txt

@@ -545,6 +545,38 @@ size (100) for parsing delimiters in inlines.c
 
 scyther@pokemon.com
 
+scy.the_rbe-edr+ill@pokemon.com
+
+scyther@pokemon.com.
+
+scyther@pokemon.com/
+
+scyther@pokemon.com/beedrill@pokemon.com
+
+mailto:scyther@pokemon.com
+
+This is a mailto:scyther@pokemon.com
+
+mailto:scyther@pokemon.com.
+
+mmmmailto:scyther@pokemon.com
+
+mailto:scyther@pokemon.com/
+
+mailto:scyther@pokemon.com/message
+
+mailto:scyther@pokemon.com/mailto:beedrill@pokemon.com
+
+xmpp:scyther@pokemon.com
+
+xmpp:scyther@pokemon.com.
+
+xmpp:scyther@pokemon.com/message
+
+xmpp:scyther@pokemon.com/message.
+
+Email me at:scyther@pokemon.com
+
 www.github.com www.github.com/á
 
 www.google.com/a_b
@@ -568,6 +600,22 @@ http://🍄.ga/ http://x🍄.ga/
 <p>: <a href="http://google.com">http://google.com</a> <a href="https://google.com">https://google.com</a></p>
 <p><a href="http://google.com/%C3%A5">http://google.com/å</a> <a href="http://google.com/%C3%A5">http://google.com/å</a></p>
 <p><a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a></p>
+<p><a href="mailto:scy.the_rbe-edr+ill@pokemon.com">scy.the_rbe-edr+ill@pokemon.com</a></p>
+<p><a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a>.</p>
+<p><a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a>/</p>
+<p><a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a>/<a href="mailto:beedrill@pokemon.com">beedrill@pokemon.com</a></p>
+<p><a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a></p>
+<p>This is a <a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a></p>
+<p><a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a>.</p>
+<p>mmmmailto:<a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a></p>
+<p><a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a>/</p>
+<p><a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a>/message</p>
+<p><a href="mailto:scyther@pokemon.com">mailto:scyther@pokemon.com</a>/<a href="mailto:beedrill@pokemon.com">mailto:beedrill@pokemon.com</a></p>
+<p><a href="xmpp:scyther@pokemon.com">xmpp:scyther@pokemon.com</a></p>
+<p><a href="xmpp:scyther@pokemon.com">xmpp:scyther@pokemon.com</a>.</p>
+<p><a href="xmpp:scyther@pokemon.com/message">xmpp:scyther@pokemon.com/message</a></p>
+<p><a href="xmpp:scyther@pokemon.com/message">xmpp:scyther@pokemon.com/message</a>.</p>
+<p>Email me at:<a href="mailto:scyther@pokemon.com">scyther@pokemon.com</a></p>
 <p><a href="http://www.github.com">www.github.com</a> <a href="http://www.github.com/%C3%A1">www.github.com/á</a></p>
 <p><a href="http://www.google.com/a_b">www.google.com/a_b</a></p>
 <p><strong>Autolink and <a href="http://inlines">http://inlines</a></strong></p>