Feat: treat CHAR[ACTER] VARYING as VARCHAR for all dialects (#5093)

* adds support for parsing CHARACTER VARYING

* char varying

* char varying

* tsql char varying unit tests

* moved char varying to main tokens

* ruff

Author: ewhitley

sqlglot/dialects/duckdb.py

@@ -320,7 +320,6 @@ class Tokenizer(tokens.Tokenizer):
             "BITSTRING": TokenType.BIT,
             "BPCHAR": TokenType.TEXT,
             "CHAR": TokenType.TEXT,
-            "CHARACTER VARYING": TokenType.TEXT,
             "DATETIME": TokenType.TIMESTAMPNTZ,
             "DETACH": TokenType.DETACH,
             "EXCLUDE": TokenType.EXCEPT,

sqlglot/dialects/postgres.py

@@ -321,7 +321,6 @@ class Tokenizer(tokens.Tokenizer):
             "BEGIN": TokenType.COMMAND,
             "BEGIN TRANSACTION": TokenType.BEGIN,
             "BIGSERIAL": TokenType.BIGSERIAL,
-            "CHARACTER VARYING": TokenType.VARCHAR,
             "CONSTRAINT TRIGGER": TokenType.COMMAND,
             "CSTRING": TokenType.PSEUDO_TYPE,
             "DECLARE": TokenType.COMMAND,

sqlglot/dialects/snowflake.py

@@ -949,8 +949,6 @@ class Tokenizer(tokens.Tokenizer):
             **tokens.Tokenizer.KEYWORDS,
             "FILE://": TokenType.URI_START,
             "BYTEINT": TokenType.INT,
-            "CHAR VARYING": TokenType.VARCHAR,
-            "CHARACTER VARYING": TokenType.VARCHAR,
             "EXCLUDE": TokenType.EXCEPT,
             "FILE FORMAT": TokenType.FILE_FORMAT,
             "GET": TokenType.GET,

sqlglot/tokens.py

@@ -907,6 +907,8 @@ class Tokenizer(metaclass=_Tokenizer):
         "JSONB": TokenType.JSONB,
         "CHAR": TokenType.CHAR,
         "CHARACTER": TokenType.CHAR,
+        "CHAR VARYING": TokenType.VARCHAR,
+        "CHARACTER VARYING": TokenType.VARCHAR,
         "NCHAR": TokenType.NCHAR,
         "VARCHAR": TokenType.VARCHAR,
         "VARCHAR2": TokenType.VARCHAR,

tests/dialects/test_dialect.py

@@ -326,6 +326,46 @@ def test_cast(self):
                 "doris": "CAST(a AS VARCHAR(3))",
             },
         )
+        self.validate_all(
+            "CAST(a AS CHARACTER VARYING)",
+            write={
+                "bigquery": "CAST(a AS STRING)",
+                "drill": "CAST(a AS VARCHAR)",
+                "duckdb": "CAST(a AS TEXT)",
+                "materialize": "CAST(a AS VARCHAR)",
+                "mysql": "CAST(a AS CHAR)",
+                "hive": "CAST(a AS STRING)",
+                "oracle": "CAST(a AS VARCHAR2)",
+                "postgres": "CAST(a AS VARCHAR)",
+                "presto": "CAST(a AS VARCHAR)",
+                "redshift": "CAST(a AS VARCHAR)",
+                "snowflake": "CAST(a AS VARCHAR)",
+                "spark": "CAST(a AS STRING)",
+                "starrocks": "CAST(a AS VARCHAR)",
+                "tsql": "CAST(a AS VARCHAR)",
+                "doris": "CAST(a AS VARCHAR)",
+            },
+        )
+        self.validate_all(
+            "CAST(a AS CHARACTER VARYING(3))",
+            write={
+                "bigquery": "CAST(a AS STRING)",
+                "drill": "CAST(a AS VARCHAR(3))",
+                "duckdb": "CAST(a AS TEXT(3))",
+                "materialize": "CAST(a AS VARCHAR(3))",
+                "mysql": "CAST(a AS CHAR(3))",
+                "hive": "CAST(a AS VARCHAR(3))",
+                "oracle": "CAST(a AS VARCHAR2(3))",
+                "postgres": "CAST(a AS VARCHAR(3))",
+                "presto": "CAST(a AS VARCHAR(3))",
+                "redshift": "CAST(a AS VARCHAR(3))",
+                "snowflake": "CAST(a AS VARCHAR(3))",
+                "spark": "CAST(a AS VARCHAR(3))",
+                "starrocks": "CAST(a AS VARCHAR(3))",
+                "tsql": "CAST(a AS VARCHAR(3))",
+                "doris": "CAST(a AS VARCHAR(3))",
+            },
+        )
         self.validate_all(
             "CAST(a AS SMALLINT)",
             write={