Edit on GitHub

Dialects

While there is a SQL standard, most SQL engines support a variation of that standard. This makes it difficult to write portable SQL code. SQLGlot bridges all the different variations, called "dialects", with an extensible SQL transpilation framework.

The base sqlglot.dialects.dialect.Dialect class implements a generic dialect that aims to be as universal as possible.

Each SQL variation has its own Dialect subclass, extending the corresponding Tokenizer, Parser and Generator classes as needed.

Implementing a custom Dialect

Creating a new SQL dialect may seem complicated at first, but it is actually quite simple in SQLGlot:

from sqlglot import exp
from sqlglot.dialects.dialect import Dialect
from sqlglot.generator import Generator
from sqlglot.tokens import Tokenizer, TokenType


class Custom(Dialect):
    class Tokenizer(Tokenizer):
        QUOTES = ["'", '"']  # Strings can be delimited by either single or double quotes
        IDENTIFIERS = ["`"]  # Identifiers can be delimited by backticks

        # Associates certain meaningful words with tokens that capture their intent
        KEYWORDS = {
            **Tokenizer.KEYWORDS,
            "INT64": TokenType.BIGINT,
            "FLOAT64": TokenType.DOUBLE,
        }

    class Generator(Generator):
        # Specifies how AST nodes, i.e. subclasses of exp.Expression, should be converted into SQL
        TRANSFORMS = {
            exp.Array: lambda self, e: f"[{self.expressions(e)}]",
        }

        # Specifies how AST nodes representing data types should be converted into SQL
        TYPE_MAPPING = {
            exp.DataType.Type.TINYINT: "INT64",
            exp.DataType.Type.SMALLINT: "INT64",
            exp.DataType.Type.INT: "INT64",
            exp.DataType.Type.BIGINT: "INT64",
            exp.DataType.Type.DECIMAL: "NUMERIC",
            exp.DataType.Type.FLOAT: "FLOAT64",
            exp.DataType.Type.DOUBLE: "FLOAT64",
            exp.DataType.Type.BOOLEAN: "BOOL",
            exp.DataType.Type.TEXT: "STRING",
        }

The above example demonstrates how certain parts of the base Dialect class can be overridden to match a different specification. Even though it is a fairly realistic starting point, we strongly encourage the reader to study existing dialect implementations in order to understand how their various components can be modified, depending on the use-case.

 1# ruff: noqa: F401
 2"""
 3## Dialects
 4
 5While there is a SQL standard, most SQL engines support a variation of that standard. This makes it difficult
 6to write portable SQL code. SQLGlot bridges all the different variations, called "dialects", with an extensible
 7SQL transpilation framework.
 8
 9The base `sqlglot.dialects.dialect.Dialect` class implements a generic dialect that aims to be as universal as possible.
10
11Each SQL variation has its own `Dialect` subclass, extending the corresponding `Tokenizer`, `Parser` and `Generator`
12classes as needed.
13
14### Implementing a custom Dialect
15
16Creating a new SQL dialect may seem complicated at first, but it is actually quite simple in SQLGlot:
17
18```python
19from sqlglot import exp
20from sqlglot.dialects.dialect import Dialect
21from sqlglot.generator import Generator
22from sqlglot.tokens import Tokenizer, TokenType
23
24
25class Custom(Dialect):
26    class Tokenizer(Tokenizer):
27        QUOTES = ["'", '"']  # Strings can be delimited by either single or double quotes
28        IDENTIFIERS = ["`"]  # Identifiers can be delimited by backticks
29
30        # Associates certain meaningful words with tokens that capture their intent
31        KEYWORDS = {
32            **Tokenizer.KEYWORDS,
33            "INT64": TokenType.BIGINT,
34            "FLOAT64": TokenType.DOUBLE,
35        }
36
37    class Generator(Generator):
38        # Specifies how AST nodes, i.e. subclasses of exp.Expression, should be converted into SQL
39        TRANSFORMS = {
40            exp.Array: lambda self, e: f"[{self.expressions(e)}]",
41        }
42
43        # Specifies how AST nodes representing data types should be converted into SQL
44        TYPE_MAPPING = {
45            exp.DataType.Type.TINYINT: "INT64",
46            exp.DataType.Type.SMALLINT: "INT64",
47            exp.DataType.Type.INT: "INT64",
48            exp.DataType.Type.BIGINT: "INT64",
49            exp.DataType.Type.DECIMAL: "NUMERIC",
50            exp.DataType.Type.FLOAT: "FLOAT64",
51            exp.DataType.Type.DOUBLE: "FLOAT64",
52            exp.DataType.Type.BOOLEAN: "BOOL",
53            exp.DataType.Type.TEXT: "STRING",
54        }
55```
56
57The above example demonstrates how certain parts of the base `Dialect` class can be overridden to match a different
58specification. Even though it is a fairly realistic starting point, we strongly encourage the reader to study existing
59dialect implementations in order to understand how their various components can be modified, depending on the use-case.
60
61----
62"""
63
64from sqlglot.dialects.bigquery import BigQuery
65from sqlglot.dialects.clickhouse import ClickHouse
66from sqlglot.dialects.databricks import Databricks
67from sqlglot.dialects.dialect import Dialect, Dialects
68from sqlglot.dialects.doris import Doris
69from sqlglot.dialects.drill import Drill
70from sqlglot.dialects.duckdb import DuckDB
71from sqlglot.dialects.hive import Hive
72from sqlglot.dialects.mysql import MySQL
73from sqlglot.dialects.oracle import Oracle
74from sqlglot.dialects.postgres import Postgres
75from sqlglot.dialects.presto import Presto
76from sqlglot.dialects.redshift import Redshift
77from sqlglot.dialects.snowflake import Snowflake
78from sqlglot.dialects.spark import Spark
79from sqlglot.dialects.spark2 import Spark2
80from sqlglot.dialects.sqlite import SQLite
81from sqlglot.dialects.starrocks import StarRocks
82from sqlglot.dialects.tableau import Tableau
83from sqlglot.dialects.teradata import Teradata
84from sqlglot.dialects.trino import Trino
85from sqlglot.dialects.tsql import TSQL