Add query_by_tags which returns complete tag records.

simpletag.py

@@ -2,7 +2,7 @@
 import sqlite3
 
 __author__ = 'Acer.Yang <yangacer@gmail.com>'
-__version__ = '0.1.4'
+__version__ = '0.1.5'
 
 
 def get_token(text):
@@ -82,6 +82,15 @@ def query_ids(self, query_str):
         for row in csr.execute(sql, (query_str,)):
             yield row[0]
 
+    def query_by_tags(self, query_str, tokenize=True):
+        sql = self.__sql__['SQL_QUERY_BY_TAGS']
+        csr = self.conn.cursor()
+        for row in csr.execute(sql, (query_str, )):
+            if tokenize is True:
+                yield row[0], [tok for tok in get_token(row[1])]
+            else:
+                yield row[0], row[1]
+
     def query_tags(self, docid):
         sql = self.__sql__['SQL_QUERY_TAGS']
         csr = self.conn.cursor()
@@ -154,10 +163,14 @@ def __init__(self, name):
                 SELECT rowid FROM {0}_text_id WHERE textid=?);
             ''',
             SQL_QUERY_IDS='''
-            SELECT * FROM {0}_text_id as lhs
+            SELECT * FROM {0}_text_id AS lhs
                 JOIN (SELECT docid FROM {0} WHERE tags MATCH ?) AS rhs
                 ON (lhs.rowid=rhs.docid);
             ''',
+            SQL_QUERY_BY_TAGS='''
+            SELECT * FROM {0}_text_id, {0} WHERE {0}.tags MATCH ? AND
+                {0}_text_id.rowid = {0}.docid;
+            ''',
             SQL_QUERY_TAGS='''
             SELECT tags FROM {0} WHERE docid=(
                 SELECT rowid FROM {0}_text_id WHERE textid=?);
@@ -238,6 +251,9 @@ def __init__(self, name, conn=None):
             SQL_PURGE_TBL='DELETE FROM {0};',
             SQL_DEL='DELETE FROM {} WHERE docid=?;',
             SQL_QUERY_IDS='SELECT docid FROM {0} WHERE tags MATCH ?;',
+            SQL_QUERY_BY_TAGS='''
+            SELECT docid, tags FROM {0} WHERE tags MATCH ?;
+            ''',
             SQL_QUERY_TAGS='SELECT tags FROM {} WHERE docid=?;',
             SQL_STATS='''
             SELECT term, documents, occurrences FROM {}_terms WHERE col=0;

tests.py

@@ -117,6 +117,38 @@ def test_update_then_query_tags(self):
         self.assertEqual([u'中文', u'行不行'], [i for i in self.ns.query_tags(789)])
         pass
 
+    def test_update_then_query_by_tags_tokenize(self):
+        self.ns.update(123, 'simpletag is awsome!')
+        self.ns.update(456, 'test is a MUST!')
+        self.assertEqual([(123, ['simpletag', 'is', 'awsome']),
+                          (456, ['test', 'is', 'a', 'MUST']), ],
+                         [i for i in self.ns.query_by_tags('is')])
+        pass
+
+    def test_update_then_query_by_tags_tokenize_str(self):
+        self.ns_str.update('/a/b', 'simpletag is awsome!')
+        self.ns_str.update('/b/a', 'test is a MUST!')
+        self.assertEqual([('/a/b', ['simpletag', 'is', 'awsome']),
+                          ('/b/a', ['test', 'is', 'a', 'MUST']), ],
+                         [i for i in self.ns_str.query_by_tags('is')])
+        pass
+
+    def test_update_then_query_by_tags_none_tokenize(self):
+        self.ns.update(123, 'simpletag is awsome!')
+        self.ns.update(456, 'test is a MUST!')
+        self.assertEqual([(123, 'simpletag is awsome!'),
+                          (456, 'test is a MUST!'), ],
+                         [i for i in self.ns.query_by_tags('is', False)])
+        pass
+
+    def test_update_then_query_by_tags_none_tokenize_str(self):
+        self.ns_str.update('/a/b', 'simpletag is awsome!')
+        self.ns_str.update('/b/a', 'test is a MUST!')
+        self.assertEqual([('/a/b', 'simpletag is awsome!'),
+                          ('/b/a', 'test is a MUST!'), ],
+                         [i for i in self.ns_str.query_by_tags('is', False)])
+        pass
+
     def test_stats(self):
         self.ns.update(123, 'simpletag is awsome!')
         self.ns.update(456, 'test is a MUST!')