Add the title and the short title to the metadata set.

- Tested for perf. hit, 100 iterations add .03s total time.
- Added the -m flag to the cmd line client to get all metadata output.
- Added test for making sure title/short title come back as well.

src/readability_lxml/client.py

@@ -14,7 +14,12 @@ def parse_args():
     parser.add_argument('-v', '--verbose',
         action='store_true',
         default=False,
-        help="Increase logging verbosity to DEBUG.")
+        help='Increase logging verbosity to DEBUG.')
+
+    parser.add_argument('-m', '--metadata',
+        action='store_true',
+        default=False,
+        help='print all metadata as well as content for the content')
 
     parser.add_argument('path', metavar='P', type=str, nargs=1,
         help="The url or file path to process in readable form.")
@@ -47,7 +52,14 @@ def main():
         doc = Document(target.read(),
             debug=args.verbose,
             url=url)
-        print doc.summary().encode(enc, 'replace')
+        if args.metadata:
+            m = doc.summary_with_metadata()
+            print m.title()
+            print m.short_title()
+            print m.confidence
+            print m.html.encode(enc, 'replace')
+        else:
+            print doc.summary().encode(enc, 'replace')
 
     finally:
         target.close()

src/readability_lxml/readability.py

@@ -52,6 +52,11 @@ class Unparseable(ValueError):
     pass
 
 
+# We want to change over the Summary to a nametuple to be more memory
+# effecient and because it doesn't need to be mutable.
+Summary = namedtuple('Summary', ['html', 'confidence', 'title', 'short_title'])
+
+
 def describe(node, depth=1):
     if not hasattr(node, 'tag'):
         return "[%s]" % type(node)
@@ -88,9 +93,7 @@ def text_length(i):
     return len(clean(i.text_content() or ""))
 
 
-# We want to change over the Summary to a nametuple to be more memory
-# effecient and because it doesn't need to be mutable.
-Summary = namedtuple('Summary', ['html', 'confidence'])
+
 
 
 class Document:
@@ -221,7 +224,10 @@ class Document:
                     # Loop through and try again.
                     continue
                 else:
-                    return Summary(confidence=confidence, html=cleaned_article)
+                    return Summary(confidence=confidence,
+                        html=cleaned_article,
+                        short_title=self.short_title,
+                        title=self.title)
         except StandardError, e:
             log.exception('error getting summary: ')
             raise Unparseable(str(e)), None, sys.exc_info()[2]

src/tests/test_article_only.py

@@ -42,8 +42,14 @@ class TestArticleOnly(unittest.TestCase):
         sample = load_sample('si-game.sample.html')
         doc = Document(sample, url='http://sportsillustrated.cnn.com/baseball/mlb/gameflash/2012/04/16/40630_preview.html')
         res = doc.summary_with_metadata(enclose_with_html_tag=False)
-        self.assertTrue(hasattr(res, 'html'), 'res should have an html attrib')
-        self.assertTrue(hasattr(res, 'confidence'), 'res should have an html attrib')
+        self.assertTrue(hasattr(res, 'html'),
+            'res should have an html attrib')
+        self.assertTrue(hasattr(res, 'confidence'),
+            'res should have an html attrib')
+        self.assertTrue(hasattr(res, 'title'),
+                'res should have an titile attrib')
+        self.assertTrue(hasattr(res, 'short_title'),
+            'res should have an short_title attrib')
         self.assertEqual('<div><div class="', res.html[0:17])
         self.assertTrue(res.confidence > 50,
             'The confidence score should be larger than 50: ' + str(res.confidence))