From: gfawcett <gfawcett@6d9bc8c9-1ec2-4278-b937-99fde70a366f>
Date: Thu, 30 Dec 2010 16:48:25 +0000 (+0000)
Subject: Tune item-sorting to ignore punctuation
X-Git-Url: https://old-git.evergreen-ils.org/?a=commitdiff_plain;h=9100d3da889da04236c3044880e975efed39d0b2;p=Syrup.git

Tune item-sorting to ignore punctuation

Article titles with quotation marks in them were throwing off the sort.

git-svn-id: svn://svn.open-ils.org/ILS-Contrib/servres/trunk@1155 6d9bc8c9-1ec2-4278-b937-99fde70a366f
---

diff --git a/conifer/syrup/models.py b/conifer/syrup/models.py
index 182541b..b8c7e29 100644
--- a/conifer/syrup/models.py
+++ b/conifer/syrup/models.py
@@ -307,10 +307,14 @@ class Site(BaseModel):
         # TODO: internationalize the stopwords list.
         STOPWORDS = set(['a', 'an', 'that', 'there', 'the', 'this'])
 
+        RE_PUNCTUATION = re.compile("""[,'".:;]""")
+
         def sort_title(item):
             """First cut of a stop words routine."""
-            normal_text = [t for t in item.lower().split() if t not in STOPWORDS]
-            return  " ".join(normal_text)
+            text = item.lower()
+            text = RE_PUNCTUATION.sub('', text) # remove common punctuation
+            words = [t for t in text.split() if t not in STOPWORDS]
+            return  " ".join(words)
 
         items = self.items()
         # make a node-lookup table