Improve parsing of readme (extraction of title and description as plain text and not markdown)

1 files changed, 14 insertions, 3 deletions

diff --git a/lib/parse/readme.js b/lib/parse/readme.js
index aab094d..e3788af 100644
--- a/lib/parse/readme.js
+++ b/lib/parse/readme.js
@@ -1,5 +1,6 @@
 var _ = require('lodash');
 var marked = require('marked');
+var textRenderer = require('marked-text-renderer');
 
 function extractFirstNode(nodes, nType) {
     return _.chain(nodes)
@@ -14,16 +15,26 @@ function extractFirstNode(nodes, nType) {
 
 function parseReadme(src) {
     var nodes, title, description;
+    var renderer = textRenderer();
 
     // Parse content
     nodes = marked.lexer(src);
-    
+
     var title = extractFirstNode(nodes, "heading");
     var description = extractFirstNode(nodes, "paragraph");
 
+    var convert = _.compose(
+        function(text) {
+            return _.unescape(text.replace(/(\r\n|\n|\r)/gm, ""));
+        },
+        _.partialRight(marked, _.extend({}, marked.defaults, {
+            renderer: renderer
+        }))
+    );
+
     return {
-        title: title,
-        description: description
+        title: convert(title),
+        description: convert(description)
     };
 }