Merge pull request #287 from xnuinside/release_v_1.7.0

release version 1.7.0 with adding support for enum and set column types

CHANGELOG.txt

@@ -1,3 +1,11 @@
+**v1.7.0**
+### Fixes
+1. DEFAULT Value with '::' cast parsed correctly now - https://github.com/xnuinside/simple-ddl-parser/issues/286
+
+### Improvements
+1. Added support for ENUM & SET column type - https://github.com/xnuinside/simple-ddl-parser/issues/259 
+
+
 **v1.6.1**
 ### Fixes
 1. #289 CREATE SCHEMA IF NOT EXISTS plus comment fail

README.md

@@ -489,6 +489,14 @@ for help with debugging & testing support for BigQuery dialect DDLs:
 * https://github.com/kalyan939
 
 ## Changelog
+**v1.7.0**
+### Fixes
+1. DEFAULT Value with '::' cast parsed correctly now - https://github.com/xnuinside/simple-ddl-parser/issues/286
+
+### Improvements
+1. Added support for ENUM & SET column type - https://github.com/xnuinside/simple-ddl-parser/issues/259 
+
+
 **v1.6.1**
 ### Fixes
 1. #289 CREATE SCHEMA IF NOT EXISTS plus comment fail

docs/README.rst

@@ -555,6 +555,20 @@ for help with debugging & testing support for BigQuery dialect DDLs:
 Changelog
 ---------
 
+**v1.7.0**
+
+Fixes
+^^^^^
+
+
+#. DEFAULT Value with '::' cast parsed correctly now - https://github.com/xnuinside/simple-ddl-parser/issues/286
+
+Improvements
+^^^^^^^^^^^^
+
+
+#. Added support for ENUM & SET column type - https://github.com/xnuinside/simple-ddl-parser/issues/259 
+
 **v1.6.1**
 
 Fixes

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "simple-ddl-parser"
-version = "1.6.1"
+version = "1.7.0"
 description = "Simple DDL Parser to parse SQL & dialects like HQL, TSQL (MSSQL), Oracle, AWS Redshift, Snowflake, MySQL, PostgreSQL, etc ddl files to json/python dict with full information about columns: types, defaults, primary keys, etc.; sequences, alters, custom types & other entities from ddl."
 authors = ["Iuliia Volkova <[email protected]>"]
 license = "MIT"

simple_ddl_parser/ddl_parser.py

@@ -61,7 +61,7 @@ def after_columns_tokens(self, t: LexToken) -> LexToken:
         t.type = tok.after_columns_tokens.get(t.value.upper(), t.type)
         if t.type != "ID":
             self.lexer.after_columns = True
-        elif self.lexer.columns_def:
+        elif not self.lexer.after_columns and self.lexer.columns_def:
             t.type = tok.columns_definition.get(t.value.upper(), t.type)
         return t
 

simple_ddl_parser/dialects/sql.py

@@ -241,13 +241,18 @@ def p_c_type(self, p: List) -> None:
         | id DOT id
         | tid
         | ARRAY
+        | ENUM LP pid RP
+        | SET LP pid RP
         | c_type ARRAY
         | c_type tid
         """
         p[0] = {}
         p_list = remove_par(list(p))
         _type = None
-        if len(p_list) == 2:
+        if p_list[1] in ("ENUM", "SET"):
+            p[0] = {"property": {"values": p_list[-1]}}
+            _type = p_list[1]
+        elif len(p_list) == 2:
             _type = p_list[-1]
         elif isinstance(p[1], str) and p[1].lower() == "encode":
             p[0] = {"property": {"encode": p[2]}}
@@ -603,6 +608,8 @@ def p_create_schema(self, p: List) -> None:
         """create_schema : c_schema id id
         | c_schema id id id
         | c_schema id
+        | create_schema COMMENT STRING
+        | create_schema COMMENT EQ STRING
         | c_schema id DOT id
         | c_schema IF NOT EXISTS id
         | c_schema IF NOT EXISTS id DOT id
@@ -611,19 +618,19 @@ def p_create_schema(self, p: List) -> None:
         p[0] = {}
         auth_index = None
 
-        if "comment" in p_list[-1]:
-            del p_list[-1]
-
         self.add_if_not_exists(p[0], p_list)
 
         if isinstance(p_list[1], dict):
             p[0] = p_list[1]
-            self.set_properties_for_schema_and_database(p, p_list)
+            if "COMMENT" in p_list:
+                p[0]["comment"] = p_list[-1]
+            else:
+                self.set_properties_for_schema_and_database(p, p_list)
         elif auth in p_list:
             auth_index = p_list.index(auth)
             self.set_auth_property_in_schema(p, p_list)
 
-        if isinstance(p_list[-1], str):
+        if not p[0].get("schema_name") and isinstance(p_list[-1], str):
             if auth_index:
                 schema_name = p_list[auth_index - 1]
                 if schema_name is None:
@@ -869,10 +876,10 @@ def get_column_and_value_from_alter(p: List) -> Tuple:
         return column, value
 
     def p_alter_default(self, p: List) -> None:
-        """alter_default : alt_table id id
-        | alt_table ADD constraint id id
-        | alt_table ADD id STRING
-        | alt_table ADD constraint id STRING
+        """alter_default : alt_table DEFAULT id
+        | alt_table ADD constraint DEFAULT id
+        | alt_table ADD DEFAULT STRING
+        | alt_table ADD constraint DEFAULT STRING
         | alter_default id
         | alter_default FOR pid
         """
@@ -1555,9 +1562,9 @@ def p_default(self, p: List) -> None:
         | default id
         | DEFAULT ID EQ id_or_string
         | DEFAULT funct_expr
+        | default dot_id
         """
         p_list = remove_par(list(p))
-
         default = self.pre_process_default(p_list)
         if "DEFAULT" in p_list:
             index_default = p_list.index("DEFAULT")
@@ -1567,6 +1574,7 @@ def p_default(self, p: List) -> None:
             default = " ".join(p_list[1:])
         if default.isnumeric():
             default = int(default)
+
         if isinstance(p[1], dict):
             p[0] = self.process_dict_default_value(p_list, default)
         else:
@@ -1596,10 +1604,11 @@ def process_dict_default_value(p_list: List, default: Any) -> Dict:
             for i in p_list[2:]:
                 if isinstance(p_list[2], str):
                     p_list[2] = p_list[2].replace("\\'", "'")
-                    if i == ")" or i == "(":
-                        data["default"] = str(data["default"]) + f"{i}"
+                    if i == ")" or i == "(" or "::" in p_list[-1]:
+                        item_to_append = f"{i}"
                     else:
-                        data["default"] = str(data["default"]) + f" {i}"
+                        item_to_append = f" {i}"
+                    data["default"] = str(data["default"]) + item_to_append
                     data["default"] = data["default"].replace("))", ")")
         return data
 
@@ -1770,6 +1779,8 @@ def p_ref(self, p: List) -> None:
         | ref LP pid RP
         | ref ON DELETE id
         | ref ON UPDATE id
+        | ref ON DELETE SET
+        | ref ON UPDATE SET
         | ref DEFERRABLE INITIALLY id
         | ref NOT DEFERRABLE
         """

simple_ddl_parser/parser.py

@@ -179,13 +179,20 @@ def pre_process_data(self, data):
             data = self.process_regex_input(data)
         quote_before = r"((?!\'[\w]*[\\']*[\w]*)"
         quote_after = r"((?![\w]*[\\']*[\w]*\')))"
+        num = 0
         # add space everywhere except strings
         for symbol, replace_to in [
             (r"(,)+", " , "),
             (r"((\()){1}", " ( "),
             (r"((\))){1}", " ) "),
         ]:
-            data = re.sub(quote_before + symbol + quote_after, replace_to, data)
+            num += 1
+            if num == 2:
+                # need for correct work with `(`` but not need in other symbols
+                quote_after_use = quote_after.replace(")))", "))*)")
+            else:
+                quote_after_use = quote_after
+            data = re.sub(quote_before + symbol + quote_after_use, replace_to, data)
 
         if data.count("'") % 2 != 0:
             data = data.replace("\\'", "pars_m_single")

simple_ddl_parser/tokens.py

@@ -37,6 +37,10 @@
     "SALT",
     "NO",
     "USING",
+    "DELETE",
+    "UPDATE",
+    "DEFAULT",
+    "COMMENT",
     # bigquery
     "OPTIONS",
     # snoflake
@@ -47,16 +51,14 @@
 
 
 columns_definition = {
-    "DELETE",
-    "UPDATE",
+    "ENUM",
+    "SET",
     "NULL",
     "ARRAY",
-    "DEFAULT",
     "COLLATE",
     "ENFORCED",
     "ENCODE",
     "GENERATED",
-    "COMMENT",
     "POLICY",
     "MASKING",
     "WITH",

tests/dialects/test_mysql.py

@@ -631,3 +631,94 @@ def test_table_properties():
         }
     ]
     assert result == expected
+
+
+def test_enum_column_type():
+    expected = {
+        "ddl_properties": [],
+        "domains": [],
+        "schemas": [],
+        "sequences": [],
+        "tables": [
+            {
+                "alter": {},
+                "checks": [],
+                "columns": [
+                    {
+                        "check": None,
+                        "default": "'enabled'",
+                        "name": "cancellation_type",
+                        "nullable": False,
+                        "references": None,
+                        "size": None,
+                        "type": "ENUM",
+                        "unique": False,
+                        "values": ["'enabled'", "'disabled'"],
+                    }
+                ],
+                "index": [],
+                "partitioned_by": [],
+                "primary_key": [],
+                "schema": None,
+                "table_name": "myset",
+                "tablespace": None,
+            }
+        ],
+        "types": [],
+    }
+    ddl = """
+CREATE TABLE myset (
+     cancellation_type enum('enabled','disabled') NOT NULL DEFAULT 'enabled'
+);
+"""
+    result = DDLParser(ddl, debug=True).run(
+        group_by_type=True,
+        output_mode="mysql",
+    )
+    assert result == expected
+
+
+def test_set_type():
+    expected = {
+        "ddl_properties": [],
+        "domains": [],
+        "schemas": [],
+        "sequences": [],
+        "tables": [
+            {
+                "alter": {},
+                "checks": [],
+                "columns": [
+                    {
+                        "check": None,
+                        "default": None,
+                        "name": "randomcolumn",
+                        "nullable": True,
+                        "references": None,
+                        "size": None,
+                        "type": "SET",
+                        "unique": False,
+                        "values": ["'a'", "'b'", "'c'", "'d'"],
+                    }
+                ],
+                "index": [],
+                "partitioned_by": [],
+                "primary_key": [],
+                "schema": None,
+                "table_name": "myset",
+                "tablespace": None,
+            }
+        ],
+        "types": [],
+    }
+
+    ddl = """
+    CREATE TABLE myset (
+        randomcolumn SET('a', 'b', 'c', 'd')
+    );
+    """
+    result = DDLParser(ddl, debug=True).run(
+        group_by_type=True,
+        output_mode="mysql",
+    )
+    assert expected == result