Edit on GitHub

SQLGlot logo

SQLGlot is a no-dependency SQL parser, transpiler, optimizer, and engine. It can be used to format SQL or translate between 20 different dialects like DuckDB, Presto / Trino, Spark / Databricks, Snowflake, and BigQuery. It aims to read a wide variety of SQL inputs and output syntactically and semantically correct SQL in the targeted dialects.

It is a very comprehensive generic SQL parser with a robust test suite. It is also quite performant, while being written purely in Python.

You can easily customize the parser, analyze queries, traverse expression trees, and programmatically build SQL.

Syntax errors are highlighted and dialect incompatibilities can warn or raise depending on configurations. However, it should be noted that SQL validation is not SQLGlot’s goal, so some syntax errors may go unnoticed.

Learn more about SQLGlot in the API documentation and the expression tree primer.

Contributions are very welcome in SQLGlot; read the contribution guide to get started!

Table of Contents

Install

From PyPI:

pip3 install "sqlglot[rs]"

# Without Rust tokenizer (slower):
# pip3 install sqlglot

Or with a local checkout:

make install

Requirements for development (optional):

make install-dev

Versioning

Given a version number MAJOR.MINOR.PATCH, SQLGlot uses the following versioning strategy:

  • The PATCH version is incremented when there are backwards-compatible fixes or feature additions.
  • The MINOR version is incremented when there are backwards-incompatible fixes or feature additions.
  • The MAJOR version is incremented when there are significant backwards-incompatible fixes or feature additions.

Get in Touch

We'd love to hear from you. Join our community Slack channel!

FAQ

I tried to parse SQL that should be valid but it failed, why did that happen?

  • You need to specify the dialect to read the SQL properly, by default it is SQLGlot's dialect which is designed to be a superset of all dialects parse_one(sql, dialect="spark"). If you tried specifying the dialect and it still doesn't work, please file an issue.

I tried to output SQL but it's not in the correct dialect!

  • You need to specify the dialect to write the sql properly, by default it is in SQLGlot's dialect parse_one(sql, dialect="spark").sql(dialect="spark").

I tried to parse invalid SQL and it should raise an error but it worked! Why didn't it validate my SQL.

  • SQLGlot is not a validator and designed to be very forgiving, handling things like trailing commas.

Examples

Formatting and Transpiling

Easily translate from one dialect to another. For example, date/time functions vary between dialects and can be hard to deal with:

import sqlglot
transpile("SELECT EPOCH_MS(1618088028295)", read="duckdb", write="hive")[0]
'SELECT FROM_UNIXTIME(1618088028295 / 1000)'

SQLGlot can even translate custom time formats:

import sqlglot
transpile("SELECT STRFTIME(x, '%y-%-m-%S')", read="duckdb", write="hive")[0]
"SELECT DATE_FORMAT(x, 'yy-M-ss')"

As another example, let's suppose that we want to read in a SQL query that contains a CTE and a cast to REAL, and then transpile it to Spark, which uses backticks for identifiers and FLOAT instead of REAL:

import sqlglot

sql = """WITH baz AS (SELECT a, c FROM foo WHERE a = 1) SELECT f.a, b.b, baz.c, CAST("b"."a" AS REAL) d FROM foo f JOIN bar b ON f.a = b.a LEFT JOIN baz ON f.a = baz.a"""
print(transpile(sql, write="spark", identify=True, pretty=True)[0])
WITH `baz` AS (
  SELECT
    `a`,
    `c`
  FROM `foo`
  WHERE
    `a` = 1
)
SELECT
  `f`.`a`,
  `b`.`b`,
  `baz`.`c`,
  CAST(`b`.`a` AS FLOAT) AS `d`
FROM `foo` AS `f`
JOIN `bar` AS `b`
  ON `f`.`a` = `b`.`a`
LEFT JOIN `baz`
  ON `f`.`a` = `baz`.`a`

Comments are also preserved on a best-effort basis when transpiling SQL code:

sql = """
/* multi
   line
   comment
*/
SELECT
  tbl.cola /* comment 1 */ + tbl.colb /* comment 2 */,
  CAST(x AS INT), # comment 3
  y               -- comment 4
FROM
  bar /* comment 5 */,
  tbl #          comment 6
"""

print(transpile(sql, read='mysql', pretty=True)[0])
/* multi
   line
   comment
*/
SELECT
  tbl.cola /* comment 1 */ + tbl.colb /* comment 2 */,
  CAST(x AS INT), /* comment 3 */
  y /* comment 4 */
FROM bar /* comment 5 */, tbl /*          comment 6 */

Metadata

You can explore SQL with expression helpers to do things like find columns and tables:

from sqlglot import parse_one, exp

# print all column references (a and b)
for column in parse_one("SELECT a, b + 1 AS c FROM d").find_all(exp.Column):
    print(column.alias_or_name)

# find all projections in select statements (a and c)
for select in parse_one("SELECT a, b + 1 AS c FROM d").find_all(exp.Select):
    for projection in select.expressions:
        print(projection.alias_or_name)

# find all tables (x, y, z)
for table in parse_one("SELECT * FROM x JOIN y JOIN z").find_all(exp.Table):
    print(table.name)

Read the ast primer to learn more about SQLGlot's internals.

Parser Errors

When the parser detects an error in the syntax, it raises a ParseError:

import sqlglot
transpile("SELECT foo( FROM bar")
sqlglot.errors.ParseError: Expecting ). Line 1, Col: 13.
  select foo( FROM bar
              ~~~~

Structured syntax errors are accessible for programmatic use:

import sqlglot
try:
    transpile("SELECT foo( FROM bar")
except sqlglot.errors.ParseError as e:
    print(e.errors)
[{
  'description': 'Expecting )',
  'line': 1,
  'col': 16,
  'start_context': 'SELECT foo( ',
  'highlight': 'FROM',
  'end_context': ' bar',
  'into_expression': None,
}]

Unsupported Errors

Presto APPROX_DISTINCT supports the accuracy argument which is not supported in Hive:

import sqlglot
transpile("SELECT APPROX_DISTINCT(a, 0.1) FROM foo", read="presto", write="hive")
APPROX_COUNT_DISTINCT does not support accuracy
'SELECT APPROX_COUNT_DISTINCT(a) FROM foo'

Build and Modify SQL

SQLGlot supports incrementally building sql expressions:

from sqlglot import select, condition

where = condition("x=1").and_("y=1")
select("*").from_("y").where(where).sql()
'SELECT * FROM y WHERE x = 1 AND y = 1'

You can also modify a parsed tree:

from sqlglot import parse_one
parse_one("SELECT x FROM y").from_("z").sql()
'SELECT x FROM z'

There is also a way to recursively transform the parsed tree by applying a mapping function to each tree node:

from sqlglot import exp, parse_one

expression_tree = parse_one("SELECT a FROM x")

def transformer(node):
    if isinstance(node, exp.Column) and node.name == "a":
        return parse_one("FUN(a)")
    return node

transformed_tree = expression_tree.transform(transformer)
transformed_tree.sql()
'SELECT FUN(a) FROM x'

SQL Optimizer

SQLGlot can rewrite queries into an "optimized" form. It performs a variety of techniques to create a new canonical AST. This AST can be used to standardize queries or provide the foundations for implementing an actual engine. For example:

import sqlglot
from sqlglot.optimizer import optimize

print(
    optimize(
        parse_one("""
            SELECT A OR (B OR (C AND D))
            FROM x
            WHERE Z = date '2021-01-01' + INTERVAL '1' month OR 1 = 0
        """),
        schema={"x": {"A": "INT", "B": "INT", "C": "INT", "D": "INT", "Z": "STRING"}}
    ).sql(pretty=True)
)
SELECT
  (
    "x"."a" <> 0 OR "x"."b" <> 0 OR "x"."c" <> 0
  )
  AND (
    "x"."a" <> 0 OR "x"."b" <> 0 OR "x"."d" <> 0
  ) AS "_col_0"
FROM "x" AS "x"
WHERE
  CAST("x"."z" AS DATE) = CAST('2021-02-01' AS DATE)

AST Introspection

You can see the AST version of the sql by calling repr:

from sqlglot import parse_one
print(repr(parse_one("SELECT a + 1 AS z")))
Select(
  expressions=[
    Alias(
      this=Add(
        this=Column(
          this=Identifier(this=a, quoted=False)),
        expression=Literal(this=1, is_string=False)),
      alias=Identifier(this=z, quoted=False))])

AST Diff

SQLGlot can calculate the difference between two expressions and output changes in a form of a sequence of actions needed to transform a source expression into a target one:

from sqlglot import diff, parse_one
diff(parse_one("SELECT a + b, c, d"), parse_one("SELECT c, a - b, d"))
[
  Remove(expression=Add(
    this=Column(
      this=Identifier(this=a, quoted=False)),
    expression=Column(
      this=Identifier(this=b, quoted=False)))),
  Insert(expression=Sub(
    this=Column(
      this=Identifier(this=a, quoted=False)),
    expression=Column(
      this=Identifier(this=b, quoted=False)))),
  Keep(source=Identifier(this=d, quoted=False), target=Identifier(this=d, quoted=False)),
  ...
]

See also: Semantic Diff for SQL.

Custom Dialects

Dialects can be added by subclassing Dialect:

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 = ["'", '"']
        IDENTIFIERS = ["`"]

        KEYWORDS = {
            **Tokenizer.KEYWORDS,
            "INT64": TokenType.BIGINT,
            "FLOAT64": TokenType.DOUBLE,
        }

    class Generator(Generator):
        TRANSFORMS = {exp.Array: lambda self, e: f"[{self.expressions(e)}]"}

        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",
        }

print(Dialect["custom"])
<class '__main__.Custom'>

SQL Execution

One can even interpret SQL queries using SQLGlot, where the tables are represented as Python dictionaries. Although the engine is not very fast (it's not supposed to be) and is in a relatively early stage of development, it can be useful for unit testing and running SQL natively across Python objects. Additionally, the foundation can be easily integrated with fast compute kernels (arrow, pandas). Below is an example showcasing the execution of a SELECT expression that involves aggregations and JOINs:

from sqlglot.executor import execute

tables = {
    "sushi": [
        {"id": 1, "price": 1.0},
        {"id": 2, "price": 2.0},
        {"id": 3, "price": 3.0},
    ],
    "order_items": [
        {"sushi_id": 1, "order_id": 1},
        {"sushi_id": 1, "order_id": 1},
        {"sushi_id": 2, "order_id": 1},
        {"sushi_id": 3, "order_id": 2},
    ],
    "orders": [
        {"id": 1, "user_id": 1},
        {"id": 2, "user_id": 2},
    ],
}

execute(
    """
    SELECT
      o.user_id,
      SUM(s.price) AS price
    FROM orders o
    JOIN order_items i
      ON o.id = i.order_id
    JOIN sushi s
      ON i.sushi_id = s.id
    GROUP BY o.user_id
    """,
    tables=tables
)
user_id price
      1   4.0
      2   3.0

See also: Writing a Python SQL engine from scratch.

Used By

Documentation

SQLGlot uses pdoc to serve its API documentation.

A hosted version is on the SQLGlot website, or you can build locally with:

make docs-serve

Run Tests and Lint

make style  # Only linter checks
make unit   # Only unit tests
make check  # Full test suite & linter checks

Benchmarks

Benchmarks run on Python 3.10.12 in seconds.

Query sqlglot sqlglotrs sqlfluff sqltree sqlparse moz_sql_parser sqloxide
tpch 0.00944 (1.0) 0.00590 (0.625) 0.32116 (33.98) 0.00693 (0.734) 0.02858 (3.025) 0.03337 (3.532) 0.00073 (0.077)
short 0.00065 (1.0) 0.00044 (0.687) 0.03511 (53.82) 0.00049 (0.759) 0.00163 (2.506) 0.00234 (3.601) 0.00005 (0.073)
long 0.00889 (1.0) 0.00572 (0.643) 0.36982 (41.56) 0.00614 (0.690) 0.02530 (2.844) 0.02931 (3.294) 0.00059 (0.066)
crazy 0.02918 (1.0) 0.01991 (0.682) 1.88695 (64.66) 0.02003 (0.686) 7.46894 (255.9) 0.64994 (22.27) 0.00327 (0.112)

Optional Dependencies

SQLGlot uses dateutil to simplify literal timedelta expressions. The optimizer will not simplify expressions like the following if the module cannot be found:

x + interval '1' month

  1# ruff: noqa: F401
  2"""
  3.. include:: ../README.md
  4
  5----
  6"""
  7
  8from __future__ import annotations
  9
 10import logging
 11import typing as t
 12
 13from sqlglot import expressions as exp
 14from sqlglot.dialects.dialect import Dialect as Dialect, Dialects as Dialects
 15from sqlglot.diff import diff as diff
 16from sqlglot.errors import (
 17    ErrorLevel as ErrorLevel,
 18    ParseError as ParseError,
 19    TokenError as TokenError,
 20    UnsupportedError as UnsupportedError,
 21)
 22from sqlglot.expressions import (
 23    Expression as Expression,
 24    alias_ as alias,
 25    and_ as and_,
 26    case as case,
 27    cast as cast,
 28    column as column,
 29    condition as condition,
 30    except_ as except_,
 31    from_ as from_,
 32    func as func,
 33    intersect as intersect,
 34    maybe_parse as maybe_parse,
 35    not_ as not_,
 36    or_ as or_,
 37    select as select,
 38    subquery as subquery,
 39    table_ as table,
 40    to_column as to_column,
 41    to_identifier as to_identifier,
 42    to_table as to_table,
 43    union as union,
 44)
 45from sqlglot.generator import Generator as Generator
 46from sqlglot.parser import Parser as Parser
 47from sqlglot.schema import MappingSchema as MappingSchema, Schema as Schema
 48from sqlglot.tokens import Tokenizer as Tokenizer, TokenType as TokenType
 49
 50if t.TYPE_CHECKING:
 51    from sqlglot._typing import E
 52    from sqlglot.dialects.dialect import DialectType as DialectType
 53
 54logger = logging.getLogger("sqlglot")
 55
 56
 57try:
 58    from sqlglot._version import __version__, __version_tuple__
 59except ImportError:
 60    logger.error(
 61        "Unable to set __version__, run `pip install -e .` or `python setup.py develop` first."
 62    )
 63
 64
 65pretty = False
 66"""Whether to format generated SQL by default."""
 67
 68schema = MappingSchema()
 69"""The default schema used by SQLGlot (e.g. in the optimizer)."""
 70
 71
 72def parse(
 73    sql: str, read: DialectType = None, dialect: DialectType = None, **opts
 74) -> t.List[t.Optional[Expression]]:
 75    """
 76    Parses the given SQL string into a collection of syntax trees, one per parsed SQL statement.
 77
 78    Args:
 79        sql: the SQL code string to parse.
 80        read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
 81        dialect: the SQL dialect (alias for read).
 82        **opts: other `sqlglot.parser.Parser` options.
 83
 84    Returns:
 85        The resulting syntax tree collection.
 86    """
 87    return Dialect.get_or_raise(read or dialect).parse(sql, **opts)
 88
 89
 90@t.overload
 91def parse_one(sql: str, *, into: t.Type[E], **opts) -> E:
 92    ...
 93
 94
 95@t.overload
 96def parse_one(sql: str, **opts) -> Expression:
 97    ...
 98
 99
100def parse_one(
101    sql: str,
102    read: DialectType = None,
103    dialect: DialectType = None,
104    into: t.Optional[exp.IntoType] = None,
105    **opts,
106) -> Expression:
107    """
108    Parses the given SQL string and returns a syntax tree for the first parsed SQL statement.
109
110    Args:
111        sql: the SQL code string to parse.
112        read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
113        dialect: the SQL dialect (alias for read)
114        into: the SQLGlot Expression to parse into.
115        **opts: other `sqlglot.parser.Parser` options.
116
117    Returns:
118        The syntax tree for the first parsed statement.
119    """
120
121    dialect = Dialect.get_or_raise(read or dialect)
122
123    if into:
124        result = dialect.parse_into(into, sql, **opts)
125    else:
126        result = dialect.parse(sql, **opts)
127
128    for expression in result:
129        if not expression:
130            raise ParseError(f"No expression was parsed from '{sql}'")
131        return expression
132    else:
133        raise ParseError(f"No expression was parsed from '{sql}'")
134
135
136def transpile(
137    sql: str,
138    read: DialectType = None,
139    write: DialectType = None,
140    identity: bool = True,
141    error_level: t.Optional[ErrorLevel] = None,
142    **opts,
143) -> t.List[str]:
144    """
145    Parses the given SQL string in accordance with the source dialect and returns a list of SQL strings transformed
146    to conform to the target dialect. Each string in the returned list represents a single transformed SQL statement.
147
148    Args:
149        sql: the SQL code string to transpile.
150        read: the source dialect used to parse the input string (eg. "spark", "hive", "presto", "mysql").
151        write: the target dialect into which the input should be transformed (eg. "spark", "hive", "presto", "mysql").
152        identity: if set to `True` and if the target dialect is not specified the source dialect will be used as both:
153            the source and the target dialect.
154        error_level: the desired error level of the parser.
155        **opts: other `sqlglot.generator.Generator` options.
156
157    Returns:
158        The list of transpiled SQL statements.
159    """
160    write = (read if write is None else write) if identity else write
161    write = Dialect.get_or_raise(write)
162    return [
163        write.generate(expression, copy=False, **opts) if expression else ""
164        for expression in parse(sql, read, error_level=error_level)
165    ]
logger = <Logger sqlglot (WARNING)>
pretty = False

Whether to format generated SQL by default.

schema = <sqlglot.schema.MappingSchema object>

The default schema used by SQLGlot (e.g. in the optimizer).

def parse( sql: str, read: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, **opts) -> List[Optional[sqlglot.expressions.Expression]]:
73def parse(
74    sql: str, read: DialectType = None, dialect: DialectType = None, **opts
75) -> t.List[t.Optional[Expression]]:
76    """
77    Parses the given SQL string into a collection of syntax trees, one per parsed SQL statement.
78
79    Args:
80        sql: the SQL code string to parse.
81        read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
82        dialect: the SQL dialect (alias for read).
83        **opts: other `sqlglot.parser.Parser` options.
84
85    Returns:
86        The resulting syntax tree collection.
87    """
88    return Dialect.get_or_raise(read or dialect).parse(sql, **opts)

Parses the given SQL string into a collection of syntax trees, one per parsed SQL statement.

Arguments:
  • sql: the SQL code string to parse.
  • read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
  • dialect: the SQL dialect (alias for read).
  • **opts: other sqlglot.parser.Parser options.
Returns:

The resulting syntax tree collection.

def parse_one( sql: str, read: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, dialect: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, into: Union[str, Type[sqlglot.expressions.Expression], Collection[Union[str, Type[sqlglot.expressions.Expression]]], NoneType] = None, **opts) -> sqlglot.expressions.Expression:
101def parse_one(
102    sql: str,
103    read: DialectType = None,
104    dialect: DialectType = None,
105    into: t.Optional[exp.IntoType] = None,
106    **opts,
107) -> Expression:
108    """
109    Parses the given SQL string and returns a syntax tree for the first parsed SQL statement.
110
111    Args:
112        sql: the SQL code string to parse.
113        read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
114        dialect: the SQL dialect (alias for read)
115        into: the SQLGlot Expression to parse into.
116        **opts: other `sqlglot.parser.Parser` options.
117
118    Returns:
119        The syntax tree for the first parsed statement.
120    """
121
122    dialect = Dialect.get_or_raise(read or dialect)
123
124    if into:
125        result = dialect.parse_into(into, sql, **opts)
126    else:
127        result = dialect.parse(sql, **opts)
128
129    for expression in result:
130        if not expression:
131            raise ParseError(f"No expression was parsed from '{sql}'")
132        return expression
133    else:
134        raise ParseError(f"No expression was parsed from '{sql}'")

Parses the given SQL string and returns a syntax tree for the first parsed SQL statement.

Arguments:
  • sql: the SQL code string to parse.
  • read: the SQL dialect to apply during parsing (eg. "spark", "hive", "presto", "mysql").
  • dialect: the SQL dialect (alias for read)
  • into: the SQLGlot Expression to parse into.
  • **opts: other sqlglot.parser.Parser options.
Returns:

The syntax tree for the first parsed statement.

def transpile( sql: str, read: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, write: Union[str, sqlglot.dialects.dialect.Dialect, Type[sqlglot.dialects.dialect.Dialect], NoneType] = None, identity: bool = True, error_level: Optional[sqlglot.errors.ErrorLevel] = None, **opts) -> List[str]:
137def transpile(
138    sql: str,
139    read: DialectType = None,
140    write: DialectType = None,
141    identity: bool = True,
142    error_level: t.Optional[ErrorLevel] = None,
143    **opts,
144) -> t.List[str]:
145    """
146    Parses the given SQL string in accordance with the source dialect and returns a list of SQL strings transformed
147    to conform to the target dialect. Each string in the returned list represents a single transformed SQL statement.
148
149    Args:
150        sql: the SQL code string to transpile.
151        read: the source dialect used to parse the input string (eg. "spark", "hive", "presto", "mysql").
152        write: the target dialect into which the input should be transformed (eg. "spark", "hive", "presto", "mysql").
153        identity: if set to `True` and if the target dialect is not specified the source dialect will be used as both:
154            the source and the target dialect.
155        error_level: the desired error level of the parser.
156        **opts: other `sqlglot.generator.Generator` options.
157
158    Returns:
159        The list of transpiled SQL statements.
160    """
161    write = (read if write is None else write) if identity else write
162    write = Dialect.get_or_raise(write)
163    return [
164        write.generate(expression, copy=False, **opts) if expression else ""
165        for expression in parse(sql, read, error_level=error_level)
166    ]

Parses the given SQL string in accordance with the source dialect and returns a list of SQL strings transformed to conform to the target dialect. Each string in the returned list represents a single transformed SQL statement.

Arguments:
  • sql: the SQL code string to transpile.
  • read: the source dialect used to parse the input string (eg. "spark", "hive", "presto", "mysql").
  • write: the target dialect into which the input should be transformed (eg. "spark", "hive", "presto", "mysql").
  • identity: if set to True and if the target dialect is not specified the source dialect will be used as both: the source and the target dialect.
  • error_level: the desired error level of the parser.
  • **opts: other sqlglot.generator.Generator options.
Returns:

The list of transpiled SQL statements.