fix

Author: RobinL

fuzzymatcher/data_getter_sqlite.py

@@ -177,17 +177,29 @@ def _tokens_to_matches(self, tokens, misspelling = False):
             """
 
         # This fails if the special tokens 'and' or 'or' are in fts string!  See issue 35!
-        tokens_to_remove = ["AND", "OR"]
-        tokens = [t for t in tokens if t not in tokens_to_remove]
+        tokens_to_escape = ["AND", "OR", "NEAR"]
+
+        def escape_token(t):
+            # return t
+            if t in tokens_to_escape:
+                return '"' + t + '"'
+            else:
+                return t
+
+
+        tokens = [escape_token(t) for t in tokens]
+
         fts_string = " ".join(tokens)
 
+
         if misspelling:
             table_name = "_concat_all_alternatives"
         else:
             table_name = "_concat_all"
 
         sql = get_records_sql.format(table_name, fts_string, self.return_records_limit)
 
+
         cur = self.con.cursor()
         cur.execute(sql)
         results = cur.fetchall()

tests/data/left_token_escape.csv

@@ -0,0 +1,5 @@
+id,fname,mname,lname,dob,another_field
+1,or,or and,and,20/05/1980,other data
+2,or,or,or smith or,15/06/1990,more data
+3,near,and,near,20/05/1960,another thing
+

tests/data/right_token_escape.csv

@@ -0,0 +1,4 @@
+id,name,middlename,surname,date,other
+1,or,or,or smith or,15/06/1990,more data
+2,near,and,near,20/05/1960,another thing
+3,or,or and,and,20/05/1980,other data

tests/test_misc.py

@@ -17,4 +17,31 @@ def test_nulls_no_errors(self):
 
         on = ["first_name", "surname", "dob", "city"]
 
-        flj = link_table(df_left, df_right, on, on)
+        flj = link_table(df_left, df_right, on, on)
+
+
+class TestNulls(unittest.TestCase):
+    """
+    Test what happens when the user provides input data with
+    fts4 match expression keyworks like AND, OR, NEAR
+    """
+
+    def test_nulls_no_errors(self):
+        """
+
+        """
+
+
+        df_left = pd.read_csv("tests/data/left_token_escape.csv")
+        df_right = pd.read_csv("tests/data/right_token_escape.csv")
+
+        # Columns to match on from df_left
+        left_on = ["fname", "mname", "lname"]
+
+        # Columns to match on from df_right
+        right_on = ["name", "middlename", "surname"]
+
+        on = ["first_name", "surname", ]
+
+        flj = link_table(df_left, df_right, left_on, right_on,
+                         left_id_col="id", right_id_col="id")