Schemas cleanup: Handle IntNodes properly

kemitche · kemitche · commit ec2f8661db7f · 2012-09-28T11:49:21.000-07:00
IntNodes were getting caught by the StandardAnalyzer for small
values. This could probably be avoided by using whoosh.fields.INT
in the schema declaration as well (and that should get done
eventually). However, a SimpleAnalyzer as the default is better for
l2cs' case anyway, because the CloudSearch server will handle
stop words and query optimizations.

Additionally, IntNodes are now properly generating unicode
elements.
diff --git a/l2cs.py b/l2cs.py
@@ -9,6 +9,7 @@
 
 import sys
 
+import whoosh.analysis
 import whoosh.fields
 import whoosh.qparser.default
 import whoosh.qparser.plugins
@@ -17,7 +18,7 @@
 import whoosh.query
 
 
-__version__ = "2.0.1"
+__version__ = "2.0.2"
 
 
 HANDLERS = {}
@@ -102,7 +103,7 @@ def walk_clause(clause):
 class IntNode(whoosh.qparser.syntax.WordNode):
     def __init__(self, value):
         self.__int_value = int(value)
-        whoosh.qparser.syntax.WordNode.__init__(self, unicode(self.__int_value))
+        whoosh.qparser.syntax.WordNode.__init__(self, value)
     
     def query(self, parser):
         q = whoosh.qparser.syntax.WordNode.query(self, parser)
@@ -144,10 +145,10 @@ def modify_node(self, fieldname, node):
 class YesNoPlugin(PseudoFieldPlugin):
     def modify_node(self, fieldname, node):
         if node.has_text:
-            if node.text in ("yes", "y", "1"):
-                new_node = IntNode(1)
+            if node.text in (u"yes", u"y", u"1"):
+                new_node = IntNode(u'1')
             else:
-                new_node = IntNode(0)
+                new_node = IntNode(u'0')
             new_node.set_fieldname(fieldname)
             return new_node
         else:
@@ -276,11 +277,15 @@ def make_schema(fields, datefields=()):
     additionally create DATETIME fields with those names
     
     '''
-    fields = dict.fromkeys(fields, whoosh.fields.TEXT)
+    text_field = whoosh.fields.TEXT(analyzer=whoosh.analysis.SimpleAnalyzer())
+    fields = dict.fromkeys(fields, text_field)
     if datefields:
         datefields = dict.fromkeys(datefields, whoosh.fields.DATETIME)
         fields.update(datefields)
-    return whoosh.fields.Schema(**fields)
+    schema = whoosh.fields.Schema()
+    for fieldname in fields:
+        schema.add(fieldname, fields[fieldname])
+    return schema
 
 
 def convert(query, parser):
diff --git a/setup.py b/setup.py
@@ -7,7 +7,7 @@
     use_setuptools()
     from setuptools import setup
 
-version = "2.0.1"
+version = "2.0.2"
 
 setup(
     name='l2cs',
diff --git a/test_l2cs.py b/test_l2cs.py
@@ -165,6 +165,9 @@ def test_unicode_intnodes1(self):
             self._run_test(u"count:1", u"count:1", self.schema_parser)
         except AssertionError as e:
             self.fail(e)
+    
+    def test_intnode_empty1(self):
+        self._run_test(u"count:''", u"")
 
 
 if __name__ == '__main__':
