Bonus per 100 chars logic was incorrect

Number of characters was being mod'd by 100 instead of divided, so a paragraph with a character length of 103 would have incorrectly gotten 3 bonus points added to the content score. Add Greg to credits
12 years ago · c8c53b304b
parent ae9208374b
commit c8c53b304b
3 changed files with 33 additions and 2 deletions
--- a/.gitignore
+++ b/.gitignore
@ -1,4 +1,5 @@
 *.pyc
+*.prof

 .installed.cfg
 bin
--- a/src/breadability/scoring.py
+++ b/src/breadability/scoring.py
@ -123,7 +123,7 @@ def score_candidates(nodes):

        # For every 100 characters in this paragraph, add another point. Up to
        # 3 points.
-        length_points = len(innertext) % 100 if innertext else 0
+        length_points = len(innertext) / 100 if innertext else 0
        if length_points > 3:
            content_score += 3
        else:
--- a/src/breadability/tests/test_readable.py
+++ b/src/breadability/tests/test_readable.py
@ -57,7 +57,7 @@ class TestReadableDocument(TestCase):

        """
        doc = Article(load_snippet('document_only_content.html'))
-        
+
        self.assertEqual(doc._readable.tag, 'div')
        self.assertEqual(doc._readable.get('id'), 'readabilityBody')

@ -119,6 +119,7 @@ class TestCleaning(TestCase):
                u'<html><body><p>simple<a href="">link</a></p></body></html>'
        )

+
 class TestCandidateNodes(TestCase):
    """Candidate nodes are scoring containers we use."""

@ -192,6 +193,35 @@ class TestScoringNodes(TestCase):
        scores = sorted([c.content_score for c in candidates.values()])
        self.assertTrue(scores[-1] > 100)

+    def test_bonus_score_per_100_chars_in_p(self):
+        """Nodes get 1pt per 100 characters up to 3 max points"""
+        def build_doc(length):
+            div = '<div id="content" class=""><p>{0}</p></div>'
+            content = 'c' * length
+            test_div = div.format(content)
+            doc = document_fromstring('<html><body>' + test_div + '</body></html>')
+            test_nodes = []
+            for node in doc.getiterator():
+                if node.tag == 'p':
+                    test_nodes.append(node)
+            return test_nodes
+
+        test_nodes = build_doc(400)
+        candidates = score_candidates(test_nodes)
+        pscore_400 = max([c.content_score for c in candidates.values()])
+
+        test_nodes = build_doc(100)
+        candidates = score_candidates(test_nodes)
+        pscore_100 = max([c.content_score for c in candidates.values()])
+
+        test_nodes = build_doc(50)
+        candidates = score_candidates(test_nodes)
+        pscore_50 = max([c.content_score for c in candidates.values()])
+
+        self.assertEqual(pscore_100, pscore_50 + 1)
+        self.assertEqual(pscore_400, pscore_50 + 3)
+
+
 class TestLinkDensityScoring(TestCase):
    """Link density will adjust out candidate scoresself."""