added support for create type as table

CHANGELOG.txt

@@ -8,7 +8,10 @@
         2. ON [PRIMARY] after CREATE TABLE statement (sample in test files test_mssql_specific.py)
         3. WITH statement for TABLE properties
         4. TEXTIMAGE_ON statement
+        5. DEFAULT NEXT VALUE FOR in COLUMN DEFAULT
+
     2. Added support for separating tables DDL by 'GO' statement as in output of MSSQL
+    3. Added support for CREATE TYPE as TABLE
 
 **v0.20.0**
 ### New Features:

README.md

@@ -282,7 +282,7 @@ You also can provide a path where you want to have a dumps with schema with argu
 
 - CREATE SEQUENCE with words: INCREMENT, START, MINVALUE, MAXVALUE, CACHE
 
-- CREATE TYPE statement:  AS ENUM, AS OBJECT, INTERNALLENGTH, INPUT, OUTPUT
+- CREATE TYPE statement:  AS TABLE, AS ENUM, AS OBJECT, INTERNALLENGTH, INPUT, OUTPUT
 
 - LIKE statement (in this and only in this case to output will be added 'like' keyword with information about table from that we did like - 'like': {'schema': None, 'table_name': 'Old_Users'}).
 
@@ -318,6 +318,11 @@ You also can provide a path where you want to have a dumps with schema with argu
 
 - CONSTRAINT [CLUSTERED]... PRIMARY KEY
 - CONSTRAINT ... WITH statement
+- PERIOD FOR SYSTEM_TIME in CREATE TABLE statement
+- ON [PRIMARY] after CREATE TABLE statement (sample in test files test_mssql_specific.py)
+- WITH statement for TABLE properties
+- TEXTIMAGE_ON statement
+- DEFAULT NEXT VALUE FOR in COLUMN DEFAULT
 
 ### MSSQL / MySQL/ Oracle
 
@@ -372,6 +377,21 @@ Big thanks for the involving & contribution with test cases with DDL samples & o
 
 
 ## Changelog
+**v0.21.0**
+### New Features:
+
+    ## MSSQL:
+
+    1. Added support for statements: 
+        1. PERIOD FOR SYSTEM_TIME in CREATE TABLE statement
+        2. ON [PRIMARY] after CREATE TABLE statement (sample in test files test_mssql_specific.py)
+        3. WITH statement for TABLE properties
+        4. TEXTIMAGE_ON statement
+        5. DEFAULT NEXT VALUE FOR in COLUMN DEFAULT
+
+    2. Added support for separating tables DDL by 'GO' statement as in output of MSSQL
+    3. Added support for CREATE TYPE as TABLE
+
 **v0.20.0**
 ### New Features:
 

docs/README.rst

@@ -307,7 +307,7 @@ Supported Statements
   CREATE SEQUENCE with words: INCREMENT, START, MINVALUE, MAXVALUE, CACHE
 
 * 
-  CREATE TYPE statement:  AS ENUM, AS OBJECT, INTERNALLENGTH, INPUT, OUTPUT
+  CREATE TYPE statement:  AS TABLE, AS ENUM, AS OBJECT, INTERNALLENGTH, INPUT, OUTPUT
 
 * 
   LIKE statement (in this and only in this case to output will be added 'like' keyword with information about table from that we did like - 'like': {'schema': None, 'table_name': 'Old_Users'}).
@@ -356,6 +356,11 @@ MSSQL
 
 * CONSTRAINT [CLUSTERED]... PRIMARY KEY
 * CONSTRAINT ... WITH statement
+* PERIOD FOR SYSTEM_TIME in CREATE TABLE statement
+* ON [PRIMARY] after CREATE TABLE statement (sample in test files test_mssql_specific.py)
+* WITH statement for TABLE properties
+* TEXTIMAGE_ON statement
+* DEFAULT NEXT VALUE FOR in COLUMN DEFAULT
 
 MSSQL / MySQL/ Oracle
 ^^^^^^^^^^^^^^^^^^^^^
@@ -426,6 +431,26 @@ Big thanks for the involving & contribution with test cases with DDL samples & o
 Changelog
 ---------
 
+**v0.21.0**
+
+New Features:
+^^^^^^^^^^^^^
+
+.. code-block::
+
+   ## MSSQL:
+
+   1. Added support for statements: 
+       1. PERIOD FOR SYSTEM_TIME in CREATE TABLE statement
+       2. ON [PRIMARY] after CREATE TABLE statement (sample in test files test_mssql_specific.py)
+       3. WITH statement for TABLE properties
+       4. TEXTIMAGE_ON statement
+       5. DEFAULT NEXT VALUE FOR in COLUMN DEFAULT
+
+   2. Added support for separating tables DDL by 'GO' statement as in output of MSSQL
+   3. Added support for CREATE TYPE as TABLE
+
+
 **v0.20.0**
 
 New Features:

pyproject.toml

@@ -1,7 +1,7 @@
 [tool.poetry]
 name = "simple-ddl-parser"
-version = "0.20.0"
-description = "Simple DDL Parser to parse SQL & dialects like HQL, TSQL, Oracle, AWS Redshift, Snowflake, 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."
+version = "0.21.0"
+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"
 readme = "docs/README.rst"

simple_ddl_parser/ddl_parser.py

@@ -103,7 +103,6 @@ def t_ID(self, t):
             self.lexer.lp_open += 1
             self.lexer.columns_def = True
             self.lexer.last_token = "LP"
-            print(t.value, t.type)
             return t
 
         elif self.is_token_column_name(t):
@@ -129,7 +128,6 @@ def set_last_token(self, t):
             self.lexer.is_table = False
         elif t.type in ["TABLE", "INDEX"]:
             self.lexer.is_table = True
-        print(t.value, t.type)
         return t
 
     def t_error(self, t):

simple_ddl_parser/dialects/mssql.py

@@ -46,7 +46,6 @@ def p_with_args(self, p):
         | with_args RP
         """
         p_list = list(p)
-        print(p_list, "p_list, with_args")
         if isinstance(p[1], dict):
             p[0] = p[1]
         else:
@@ -59,7 +58,6 @@ def p_with_args(self, p):
 
     def p_period_for(self, p):
         """period_for : PERIOD FOR ID LP pid RP"""
-        print(list(p))
         p[0] = {"period_for_system_time": p[5]}
 
     def p_expression_on_primary(self, p):

simple_ddl_parser/dialects/sql.py

@@ -409,26 +409,44 @@ def p_multiple_column_names(self, p: List) -> None:
             if p_list[-1] != ",":
                 p[0].append(p_list[-1])
 
-    def p_expression_type_as(self, p: List) -> None:
-        """expr : type_name ID LP pid RP
+    def p_type_definition(self, p: List) -> None:  # noqa: C901
+        """type_definition : type_name ID LP pid RP
         | type_name ID LP multiple_column_names RP
         | type_name LP id_equals RP
+        | type_name TABLE LP defcolumn
+        | type_definition COMMA defcolumn
+        | type_definition RP
         """
-        p_list = list(p)
+        p_list = remove_par(list(p))
         p[0] = p[1]
-        p[0]["base_type"] = p[2]
-        p[0]["properties"] = {}
-        base_type = p[0]["base_type"].upper()
-        if base_type == "ENUM":
-            p[0]["properties"]["values"] = p_list[4]
-        elif p[0]["base_type"] == "OBJECT":
-            if "type" in p_list[4][0]:
-                p[0]["properties"]["attributes"] = p_list[4]
+        if not p[0].get("properties"):
+            p[0]["properties"] = {}
+
+        if "TABLE" in p_list or isinstance(p_list[-1], dict) and p_list[-1].get("name"):
+            if not p[0]["properties"].get("columns"):
+                p[0]["properties"]["columns"] = []
+            p[0]["properties"]["columns"].append(p_list[-1])
+
+        if len(p_list) > 3:
+            p[0]["base_type"] = p_list[2]
         else:
-            if isinstance(p_list[-2], list):
-                for item in p_list[-2]:
+            p[0]["base_type"] = None
+        if isinstance(p[0]["base_type"], str):
+            base_type = p[0]["base_type"].upper()
+            if base_type == "ENUM":
+                p[0]["properties"]["values"] = p_list[3]
+            elif p[0]["base_type"] == "OBJECT":
+                if "type" in p_list[3][0]:
+                    p[0]["properties"]["attributes"] = p_list[3]
+        else:
+            if isinstance(p_list[-1], list):
+                for item in p_list[-1]:
                     p[0]["properties"].update(item)
 
+    def p_expression_type_as(self, p: List) -> None:
+        """expr : type_definition"""
+        p[0] = p[1]
+
     def p_type_name(self, p: List) -> None:
         """type_name : type_create ID AS
         | type_create ID DOT ID AS
@@ -573,7 +591,6 @@ def p_expression_table(self, p: List) -> None:
         """
         p[0] = p[1]
         p_list = list(p)
-        print(p_list[-1])
         if p_list[-1] != "," and p_list[-1] != ")":
             if "type" in p_list[-1] and "name" in p_list[-1]:
                 p[0]["columns"].append(p_list[-1])
@@ -843,33 +860,45 @@ def p_funct_expr(self, p: List) -> None:
         else:
             p[0] = p[1]
 
+    def p_dot_id(self, p: List) -> None:
+        """dot_id : ID DOT ID"""
+        p[0] = f"{p[1]}.{p[3]}"
+
     def p_default(self, p: List) -> None:
         """default : DEFAULT ID
         | DEFAULT STRING
         | DEFAULT NULL
+        | default FOR dot_id
         | DEFAULT funct_expr
         | DEFAULT LP pid RP
         | default ID
         | default LP RP
         """
         p_list = list(p)
+
         if len(p_list) == 5 and isinstance(p[3], list):
             default = p[3][0]
+        elif "DEFAULT" in p_list and len(p_list) == 4:
+            default = f"{p[2]} {p[3]}"
         else:
             default = p[2]
 
-        if default.isnumeric():
+        if not isinstance(default, dict) and default.isnumeric():
             default = int(default)
+
         if isinstance(p[1], dict):
             p[0] = p[1]
-            for i in p[2:]:
-                if isinstance(p[2], str):
-                    p[2] = p[2].replace("\\'", "'")
-                    if i == ")" or i == "(":
-                        p[0]["default"] = str(p[0]["default"]) + f"{i}"
-                    else:
-                        p[0]["default"] = str(p[0]["default"]) + f" {i}"
-                    p[0]["default"] = p[0]["default"].replace("))", ")")
+            if "FOR" in default:
+                p[0]["default"] = {"next_value_for": p_list[-1]}
+            else:
+                for i in p[2:]:
+                    if isinstance(p[2], str):
+                        p[2] = p[2].replace("\\'", "'")
+                        if i == ")" or i == "(":
+                            p[0]["default"] = str(p[0]["default"]) + f"{i}"
+                        else:
+                            p[0]["default"] = str(p[0]["default"]) + f" {i}"
+                        p[0]["default"] = p[0]["default"].replace("))", ")")
         else:
             p[0] = {"default": default}
 

simple_ddl_parser/parser.py

@@ -137,7 +137,6 @@ def parse_data(self):
         skip_line_words = ["USE", "GO"]
         set_line = None
         for num, line in enumerate(lines):
-            print(repr(line))
             skip = False
             for word in skip_line_words:
                 if line.startswith(word):
@@ -164,7 +163,6 @@ def parse_data(self):
 
                 self.set_default_flags_in_lexer()
                 if not set_line and statement:
-                    print("statement", statement)
                     self.parse_statement(tables, statement)
 
                 statement = None

tests/test_custom_types_and_domains.py

@@ -145,7 +145,7 @@ def test_create_type_with_input_properties():
         "tables": [],
         "types": [
             {
-                "base_type": "(",
+                "base_type": None,
                 "properties": {
                     "INPUT": "my_box_in_function",
                     "INTERNALLENGTH": "16",
@@ -158,3 +158,54 @@ def test_create_type_with_input_properties():
     }
 
     assert expected == result
+
+
+def test_as_table():
+    ddl = """
+    CREATE TYPE dbo.T_LCT_SLIPS AS TABLE (
+    hashKY varbinary(48),
+    numContratoGF bigint
+    );
+    """
+    result = DDLParser(ddl).run(group_by_type=True)
+    expected = {
+        "ddl_properties": [],
+        "domains": [],
+        "schemas": [],
+        "sequences": [],
+        "tables": [],
+        "types": [
+            {
+                "base_type": None,
+                "properties": {
+                    "columns": [
+                        {
+                            "check": None,
+                            "default": None,
+                            "name": "hashKY",
+                            "nullable": True,
+                            "primary_key": False,
+                            "references": None,
+                            "size": 48,
+                            "type": "varbinary",
+                            "unique": False,
+                        },
+                        {
+                            "check": None,
+                            "default": None,
+                            "name": "numContratoGF",
+                            "nullable": True,
+                            "primary_key": False,
+                            "references": None,
+                            "size": None,
+                            "type": "bigint",
+                            "unique": False,
+                        },
+                    ]
+                },
+                "schema": "dbo",
+                "type_name": "T_LCT_SLIPS",
+            }
+        ],
+    }
+    assert result == expected

tests/test_mssql_specific.py

@@ -2972,3 +2972,43 @@ def test_output_separated_by_go_and_textimage():
     }
 
     assert expected == result
+
+
+def test_next_value_for():
+
+    ddl = """CREATE TABLE [dbo].[SLIPEVENTO] (
+    [cdSLIP] [bigint] NOT NULL
+    DEFAULT NEXT VALUE FOR [dbo].[sqCdSLIPEvt] )"""
+    result = DDLParser(ddl).run(group_by_type=True)
+    expected = {
+        "ddl_properties": [],
+        "domains": [],
+        "schemas": [],
+        "sequences": [],
+        "tables": [
+            {
+                "alter": {},
+                "checks": [],
+                "columns": [
+                    {
+                        "check": None,
+                        "default": {"next_value_for": "[dbo].[sqCdSLIPEvt]"},
+                        "name": "[cdSLIP]",
+                        "nullable": False,
+                        "references": None,
+                        "size": None,
+                        "type": "[bigint]",
+                        "unique": False,
+                    }
+                ],
+                "index": [],
+                "partitioned_by": [],
+                "primary_key": [],
+                "schema": "[dbo]",
+                "table_name": "[SLIPEVENTO]",
+                "tablespace": None,
+            }
+        ],
+        "types": [],
+    }
+    assert expected == result