Merge pull request #13 from ValentinHerrmann/support-limit

Support LIMIT clause in SQL queries

README.md

@@ -22,3 +22,6 @@ A library that normalizes simple SQL queries and compares them first by equality
 - FROM: one or more table from DB; no queries as tables!
 - WHERE: single conditions; no Paranthesises!
 - GROUP BY one or more columns
+
+### V 0.2.1
+- Support LIMIT clause

sql_testing_tools/Helper.py

@@ -41,7 +41,7 @@ def normalizeSQLQuery(query, baseDict):
 
     formatted_query = []
 
-    # Second pass to process SELECT, WHERE, GROUP BY, and ORDER BY clauses
+    # Second pass to process SELECT, WHERE, GROUP BY, ORDER BY, and LIMIT clauses
     for token in parsed.tokens:
         if token.is_whitespace:
             continue
@@ -53,6 +53,8 @@ def normalizeSQLQuery(query, baseDict):
             formatted_query.append('GROUP BY')
         elif token.ttype is Keyword and token.value.upper() == 'ORDER BY':
             formatted_query.append('ORDER BY')
+        elif token.ttype is Keyword and token.value.upper() == 'LIMIT':
+            formatted_query.append('LIMIT')
         elif (isinstance(token, IdentifierList) or isinstance(token, Identifier)) and formatted_query and formatted_query[-1] == 'FROM':
             formatted_query.append(Tp._from(token, alias_map, baseDict))
         elif (isinstance(token, IdentifierList) or isinstance(token, Identifier)) and formatted_query and formatted_query[-1] == 'GROUP BY':
@@ -66,6 +68,8 @@ def normalizeSQLQuery(query, baseDict):
             if isinstance(token, Function):
                 token = IdentifierList([token])
             formatted_query.append(Tp._select(token, alias_map, baseDict))
+        elif formatted_query and formatted_query[-1] == 'LIMIT':
+            formatted_query.append(Tp._limit(token))
         else:
             formatted_query.append(str(token))
 

sql_testing_tools/TokenProcessing.py

@@ -341,29 +341,5 @@ def _where_twoConditionsOnTopLevel(where, alias_map, baseDict: dict, keyword, is
             return keyword.join(current_condition)
 
 
-
-
-# def _where_xx(where, alias_map, baseDict: dict):
-#     where_tokens = []
-#     for token in where.tokens:
-
-#         if isinstance(token, Comparison):
-#             left, operator, right = [t for t in token.tokens if not t.is_whitespace]
-#             left = _identifier(left, alias_map, baseDict)
-#             right = _identifier(right, alias_map, baseDict)
-#             if left.lower() >= right.lower():
-#                 left, right = right, left
-#                 if operator.value == ">":
-#                     operator = "<"
-#                 elif operator.value == "<":
-#                     operator = ">"
-
-#             where_tokens.append(f"{left.lower()} {operator} {right.lower()}")
-#         elif token.ttype is Keyword:
-#             where_tokens.append(token.value.upper())
-#         elif token.is_whitespace or (token.ttype is Keyword and token.value.upper() == "WHERE"):
-#             continue
-#         else:
-#             where_tokens.append(str(token))
-#     return " ".join(where_tokens)
-
+def _limit(limit):
+    return str(limit).strip()

sql_testing_tools/tests/NormalizeQuery_test.py

@@ -198,3 +198,15 @@ def test_a15_ORDER_BY(self):
             self.fail("\n" + q2 + "\n" + q2)
         if q1 != q2:
             self.fail("\n" + q1 + "\n" + q2)
+
+    def test_a16_LIMIT(self):
+        nr = '16'
+        td = Ba.getTableDict()
+        q1 = He.normalizeSQLQuery(self.readFile("sql_testing_tools/tests/v1/a"+nr+".sql"),td)
+        q2 = He.normalizeSQLQuery(self.readFile("sql_testing_tools/tests/v2/a"+nr+".sql"),td)
+        if q1 != q1:
+            self.fail("\n" + q1 + "\n" + q1)
+        if q2 != q2:
+            self.fail("\n" + q2 + "\n" + q2)
+        if q1 != q2:
+            self.fail("\n" + q1 + "\n" + q2)

sql_testing_tools/tests/v1/a16.sql

@@ -0,0 +1,3 @@
+SELECT name, einwohner_m, einwohner_w 
+FROM Gemeinde 
+LIMIT 10

sql_testing_tools/tests/v2/a16.sql

@@ -0,0 +1,3 @@
+SELECT name, einwohner_m, einwohner_w 
+FROM Gemeinde 
+LIMIT 10