| Name | sqlglot JSON |
| Version |
23.13.7
JSON |
| download |
| home_page | https://github.com/tobymao/sqlglot |
| Summary | An easily customizable SQL parser and transpiler |
| upload_time | 2024-05-04 22:01:25 |
| maintainer | None |
| docs_url | None |
| author | Toby Mao |
| requires_python | >=3.7 |
| license | MIT |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
SQLGlot is a no-dependency SQL parser, transpiler, optimizer, and engine. It can be used to format SQL or translate between [21 different dialects](https://github.com/tobymao/sqlglot/blob/main/sqlglot/dialects/__init__.py) like [DuckDB](https://duckdb.org/), [Presto](https://prestodb.io/) / [Trino](https://trino.io/), [Spark](https://spark.apache.org/) / [Databricks](https://www.databricks.com/), [Snowflake](https://www.snowflake.com/en/), and [BigQuery](https://cloud.google.com/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](https://github.com/tobymao/sqlglot/blob/main/tests/). It is also quite [performant](#benchmarks), while being written purely in Python.
You can easily [customize](#custom-dialects) the parser, [analyze](#metadata) queries, traverse expression trees, and programmatically [build](#build-and-modify-sql) SQL.
Syntax [errors](#parser-errors) are highlighted and dialect incompatibilities can warn or raise depending on configurations. However, SQLGlot does not aim to be a SQL validator, so it may fail to detect certain syntax errors.
Learn more about SQLGlot in the API [documentation](https://sqlglot.com/) and the expression tree [primer](https://github.com/tobymao/sqlglot/blob/main/posts/ast_primer.md).
Contributions are very welcome in SQLGlot; read the [contribution guide](https://github.com/tobymao/sqlglot/blob/main/CONTRIBUTING.md) to get started!
## Table of Contents
* [Install](#install)
* [Versioning](#versioning)
* [Get in Touch](#get-in-touch)
* [FAQ](#faq)
* [Examples](#examples)
* [Formatting and Transpiling](#formatting-and-transpiling)
* [Metadata](#metadata)
* [Parser Errors](#parser-errors)
* [Unsupported Errors](#unsupported-errors)
* [Build and Modify SQL](#build-and-modify-sql)
* [SQL Optimizer](#sql-optimizer)
* [AST Introspection](#ast-introspection)
* [AST Diff](#ast-diff)
* [Custom Dialects](#custom-dialects)
* [SQL Execution](#sql-execution)
* [Used By](#used-by)
* [Documentation](#documentation)
* [Run Tests and Lint](#run-tests-and-lint)
* [Benchmarks](#benchmarks)
* [Optional Dependencies](#optional-dependencies)
## Install
From PyPI:
```bash
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](https://tobikodata.com/slack)!
## FAQ
I tried to parse SQL that should be valid but it failed, why did that happen?
* Most of the time, issues like this occur because the "source" dialect is omitted during parsing. For example, this is how to correctly parse a SQL query written in Spark SQL: `parse_one(sql, dialect="spark")` (alternatively: `read="spark"`). If no dialect is specified, `parse_one` will attempt to parse the query according to the "SQLGlot dialect", which is designed to be a superset of all supported dialects. 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!
* Like parsing, generating SQL also requires the target dialect to be specified, otherwise the SQLGlot dialect will be used by default. For example, to transpile a query from Spark SQL to DuckDB, do `parse_one(sql, dialect="spark").sql(dialect="duckdb")` (alternatively: `transpile(sql, read="spark", write="duckdb")`).
I tried to parse invalid SQL and it worked, even though it should raise an error! Why didn't it validate my SQL?
* SQLGlot does not aim to be a SQL validator - it is designed to be very forgiving. This makes the codebase more comprehensive and also gives more flexibility to its users, e.g. by allowing them to include trailing commas in their projection lists.
## 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:
```python
import sqlglot
sqlglot.transpile("SELECT EPOCH_MS(1618088028295)", read="duckdb", write="hive")[0]
```
```sql
'SELECT FROM_UNIXTIME(1618088028295 / POW(10, 3))'
```
SQLGlot can even translate custom time formats:
```python
import sqlglot
sqlglot.transpile("SELECT STRFTIME(x, '%y-%-m-%S')", read="duckdb", write="hive")[0]
```
```sql
"SELECT DATE_FORMAT(x, 'yy-M-ss')"
```
Identifier delimiters and data types can be translated as well:
```python
import sqlglot
# Spark SQL requires backticks (`) for delimited identifiers and uses `FLOAT` over `REAL`
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"""
# Translates the query into Spark SQL, formats it, and delimits all of its identifiers
print(sqlglot.transpile(sql, write="spark", identify=True, pretty=True)[0])
```
```sql
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:
```python
sql = """
/* multi
line
comment
*/
SELECT
tbl.cola /* comment 1 */ + tbl.colb /* comment 2 */,
CAST(x AS SIGNED), # comment 3
y -- comment 4
FROM
bar /* comment 5 */,
tbl # comment 6
"""
# Note: MySQL-specific comments (`#`) are converted into standard syntax
print(sqlglot.transpile(sql, read='mysql', pretty=True)[0])
```
```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 */
```
### Metadata
You can explore SQL with expression helpers to do things like find columns and tables in a query:
```python
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](https://github.com/tobymao/sqlglot/blob/main/posts/ast_primer.md) to learn more about SQLGlot's internals.
### Parser Errors
When the parser detects an error in the syntax, it raises a `ParseError`:
```python
import sqlglot
sqlglot.transpile("SELECT foo FROM (SELECT baz FROM t")
```
```
sqlglot.errors.ParseError: Expecting ). Line 1, Col: 34.
SELECT foo FROM (SELECT baz FROM t
~
```
Structured syntax errors are accessible for programmatic use:
```python
import sqlglot
try:
sqlglot.transpile("SELECT foo FROM (SELECT baz FROM t")
except sqlglot.errors.ParseError as e:
print(e.errors)
```
```python
[{
'description': 'Expecting )',
'line': 1,
'col': 34,
'start_context': 'SELECT foo FROM (SELECT baz FROM ',
'highlight': 't',
'end_context': '',
'into_expression': None
}]
```
### Unsupported Errors
It may not be possible to translate some queries between certain dialects. For these cases, SQLGlot may emit a warning and will proceed to do a best-effort translation by default:
```python
import sqlglot
sqlglot.transpile("SELECT APPROX_DISTINCT(a, 0.1) FROM foo", read="presto", write="hive")
```
```sql
APPROX_COUNT_DISTINCT does not support accuracy
'SELECT APPROX_COUNT_DISTINCT(a) FROM foo'
```
This behavior can be changed by setting the [`unsupported_level`](https://github.com/tobymao/sqlglot/blob/b0e8dc96ba179edb1776647b5bde4e704238b44d/sqlglot/errors.py#L9) attribute. For example, we can set it to either `RAISE` or `IMMEDIATE` to ensure an exception is raised instead:
```python
import sqlglot
sqlglot.transpile("SELECT APPROX_DISTINCT(a, 0.1) FROM foo", read="presto", write="hive", unsupported_level=sqlglot.ErrorLevel.RAISE)
```
```
sqlglot.errors.UnsupportedError: APPROX_COUNT_DISTINCT does not support accuracy
```
There are queries that require additional information to be accurately transpiled, such as the schemas of the tables referenced in them. This is because certain transformations are type-sensitive, meaning that type inference is needed in order to understand their semantics. Even though the `qualify` and `annotate_types` optimizer [rules](https://github.com/tobymao/sqlglot/tree/main/sqlglot/optimizer) can help with this, they are not used by default because they add significant overhead and complexity.
Transpilation is generally a hard problem, so SQLGlot employs an "incremental" approach to solving it. This means that there may be dialect pairs that currently lack support for some inputs, but this is expected to improve over time. We highly appreciate well-documented and tested issues or PRs, so feel free to [reach out](#get-in-touch) if you need guidance!
### Build and Modify SQL
SQLGlot supports incrementally building SQL expressions:
```python
from sqlglot import select, condition
where = condition("x=1").and_("y=1")
select("*").from_("y").where(where).sql()
```
```sql
'SELECT * FROM y WHERE x = 1 AND y = 1'
```
It's possible to modify a parsed tree:
```python
from sqlglot import parse_one
parse_one("SELECT x FROM y").from_("z").sql()
```
```sql
'SELECT x FROM z'
```
Parsed expressions can also be transformed recursively by applying a mapping function to each node in the tree:
```python
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()
```
```sql
'SELECT FUN(a) FROM x'
```
### SQL Optimizer
SQLGlot can rewrite queries into an "optimized" form. It performs a variety of [techniques](https://github.com/tobymao/sqlglot/blob/main/sqlglot/optimizer/optimizer.py) 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:
```python
import sqlglot
from sqlglot.optimizer import optimize
print(
optimize(
sqlglot.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)
)
```
```sql
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 parsed SQL by calling `repr`:
```python
from sqlglot import parse_one
print(repr(parse_one("SELECT a + 1 AS z")))
```
```python
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 semantic 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:
```python
from sqlglot import diff, parse_one
diff(parse_one("SELECT a + b, c, d"), parse_one("SELECT c, a - b, d"))
```
```python
[
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=Column(this=Identifier(this=a, quoted=False)),
target=Column(this=Identifier(this=a, quoted=False))),
...
]
```
See also: [Semantic Diff for SQL](https://github.com/tobymao/sqlglot/blob/main/posts/sql_diff.md).
### Custom Dialects
[Dialects](https://github.com/tobymao/sqlglot/tree/main/sqlglot/dialects) can be added by subclassing `Dialect`:
```python
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
SQLGlot is able to interpret SQL queries, where the tables are represented as Python dictionaries. The engine is not supposed to be fast, but 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, such as [Arrow](https://arrow.apache.org/docs/index.html) and [Pandas](https://pandas.pydata.org/).
The example below showcases the execution of a query that involves aggregations and joins:
```python
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
)
```
```python
user_id price
1 4.0
2 3.0
```
See also: [Writing a Python SQL engine from scratch](https://github.com/tobymao/sqlglot/blob/main/posts/python_sql_engine.md).
## Used By
* [SQLMesh](https://github.com/TobikoData/sqlmesh)
* [Apache Superset](https://github.com/apache/superset)
* [Dagster](https://github.com/dagster-io/dagster)
* [Fugue](https://github.com/fugue-project/fugue)
* [ibis](https://github.com/ibis-project/ibis)
* [mysql-mimic](https://github.com/kelsin/mysql-mimic)
* [Querybook](https://github.com/pinterest/querybook)
* [Quokka](https://github.com/marsupialtail/quokka)
* [Splink](https://github.com/moj-analytical-services/splink)
## Documentation
SQLGlot uses [pdoc](https://pdoc.dev/) to serve its API documentation.
A hosted version is on the [SQLGlot website](https://sqlglot.com/), or you can build locally with:
```
make docs-serve
```
## Run Tests and Lint
```
make style # Only linter checks
make unit # Only unit tests (or unit-rs, to use the Rust tokenizer)
make test # Unit and integration tests (or test-rs, to use the Rust tokenizer)
make check # Full test suite & linter checks
```
## Benchmarks
[Benchmarks](https://github.com/tobymao/sqlglot/blob/main/benchmarks/bench.py) 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](https://github.com/dateutil/dateutil) to simplify literal timedelta expressions. The optimizer will not simplify expressions like the following if the module cannot be found:
```sql
x + interval '1' month
```
Raw data
{
"_id": null,
"home_page": "https://github.com/tobymao/sqlglot",
"name": "sqlglot",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": null,
"author": "Toby Mao",
"author_email": "toby.mao@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/62/34/f34392ce9e6461d41503851fe69ddf436c17b7a8dcb0b9778f0c5aa33f60/sqlglot-23.13.7.tar.gz",
"platform": null,
"description": "\n\nSQLGlot is a no-dependency SQL parser, transpiler, optimizer, and engine. It can be used to format SQL or translate between [21 different dialects](https://github.com/tobymao/sqlglot/blob/main/sqlglot/dialects/__init__.py) like [DuckDB](https://duckdb.org/), [Presto](https://prestodb.io/) / [Trino](https://trino.io/), [Spark](https://spark.apache.org/) / [Databricks](https://www.databricks.com/), [Snowflake](https://www.snowflake.com/en/), and [BigQuery](https://cloud.google.com/bigquery/). It aims to read a wide variety of SQL inputs and output syntactically and semantically correct SQL in the targeted dialects.\n\nIt is a very comprehensive generic SQL parser with a robust [test suite](https://github.com/tobymao/sqlglot/blob/main/tests/). It is also quite [performant](#benchmarks), while being written purely in Python.\n\nYou can easily [customize](#custom-dialects) the parser, [analyze](#metadata) queries, traverse expression trees, and programmatically [build](#build-and-modify-sql) SQL.\n\nSyntax [errors](#parser-errors) are highlighted and dialect incompatibilities can warn or raise depending on configurations. However, SQLGlot does not aim to be a SQL validator, so it may fail to detect certain syntax errors.\n\nLearn more about SQLGlot in the API [documentation](https://sqlglot.com/) and the expression tree [primer](https://github.com/tobymao/sqlglot/blob/main/posts/ast_primer.md).\n\nContributions are very welcome in SQLGlot; read the [contribution guide](https://github.com/tobymao/sqlglot/blob/main/CONTRIBUTING.md) to get started!\n\n## Table of Contents\n\n* [Install](#install)\n* [Versioning](#versioning)\n* [Get in Touch](#get-in-touch)\n* [FAQ](#faq)\n* [Examples](#examples)\n * [Formatting and Transpiling](#formatting-and-transpiling)\n * [Metadata](#metadata)\n * [Parser Errors](#parser-errors)\n * [Unsupported Errors](#unsupported-errors)\n * [Build and Modify SQL](#build-and-modify-sql)\n * [SQL Optimizer](#sql-optimizer)\n * [AST Introspection](#ast-introspection)\n * [AST Diff](#ast-diff)\n * [Custom Dialects](#custom-dialects)\n * [SQL Execution](#sql-execution)\n* [Used By](#used-by)\n* [Documentation](#documentation)\n* [Run Tests and Lint](#run-tests-and-lint)\n* [Benchmarks](#benchmarks)\n* [Optional Dependencies](#optional-dependencies)\n\n## Install\n\nFrom PyPI:\n\n```bash\npip3 install \"sqlglot[rs]\"\n\n# Without Rust tokenizer (slower):\n# pip3 install sqlglot\n```\n\nOr with a local checkout:\n\n```\nmake install\n```\n\nRequirements for development (optional):\n\n```\nmake install-dev\n```\n\n## Versioning\n\nGiven a version number `MAJOR`.`MINOR`.`PATCH`, SQLGlot uses the following versioning strategy:\n\n- The `PATCH` version is incremented when there are backwards-compatible fixes or feature additions.\n- The `MINOR` version is incremented when there are backwards-incompatible fixes or feature additions.\n- The `MAJOR` version is incremented when there are significant backwards-incompatible fixes or feature additions.\n\n## Get in Touch\n\nWe'd love to hear from you. Join our community [Slack channel](https://tobikodata.com/slack)!\n\n## FAQ\n\nI tried to parse SQL that should be valid but it failed, why did that happen?\n\n* Most of the time, issues like this occur because the \"source\" dialect is omitted during parsing. For example, this is how to correctly parse a SQL query written in Spark SQL: `parse_one(sql, dialect=\"spark\")` (alternatively: `read=\"spark\"`). If no dialect is specified, `parse_one` will attempt to parse the query according to the \"SQLGlot dialect\", which is designed to be a superset of all supported dialects. If you tried specifying the dialect and it still doesn't work, please file an issue.\n\nI tried to output SQL but it's not in the correct dialect!\n\n* Like parsing, generating SQL also requires the target dialect to be specified, otherwise the SQLGlot dialect will be used by default. For example, to transpile a query from Spark SQL to DuckDB, do `parse_one(sql, dialect=\"spark\").sql(dialect=\"duckdb\")` (alternatively: `transpile(sql, read=\"spark\", write=\"duckdb\")`).\n\nI tried to parse invalid SQL and it worked, even though it should raise an error! Why didn't it validate my SQL?\n\n* SQLGlot does not aim to be a SQL validator - it is designed to be very forgiving. This makes the codebase more comprehensive and also gives more flexibility to its users, e.g. by allowing them to include trailing commas in their projection lists.\n\n## Examples\n\n### Formatting and Transpiling\n\nEasily translate from one dialect to another. For example, date/time functions vary between dialects and can be hard to deal with:\n\n```python\nimport sqlglot\nsqlglot.transpile(\"SELECT EPOCH_MS(1618088028295)\", read=\"duckdb\", write=\"hive\")[0]\n```\n\n```sql\n'SELECT FROM_UNIXTIME(1618088028295 / POW(10, 3))'\n```\n\nSQLGlot can even translate custom time formats:\n\n```python\nimport sqlglot\nsqlglot.transpile(\"SELECT STRFTIME(x, '%y-%-m-%S')\", read=\"duckdb\", write=\"hive\")[0]\n```\n\n```sql\n\"SELECT DATE_FORMAT(x, 'yy-M-ss')\"\n```\n\nIdentifier delimiters and data types can be translated as well:\n\n```python\nimport sqlglot\n\n# Spark SQL requires backticks (`) for delimited identifiers and uses `FLOAT` over `REAL`\nsql = \"\"\"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\"\"\"\n\n# Translates the query into Spark SQL, formats it, and delimits all of its identifiers\nprint(sqlglot.transpile(sql, write=\"spark\", identify=True, pretty=True)[0])\n```\n\n```sql\nWITH `baz` AS (\n SELECT\n `a`,\n `c`\n FROM `foo`\n WHERE\n `a` = 1\n)\nSELECT\n `f`.`a`,\n `b`.`b`,\n `baz`.`c`,\n CAST(`b`.`a` AS FLOAT) AS `d`\nFROM `foo` AS `f`\nJOIN `bar` AS `b`\n ON `f`.`a` = `b`.`a`\nLEFT JOIN `baz`\n ON `f`.`a` = `baz`.`a`\n```\n\nComments are also preserved on a best-effort basis:\n\n```python\nsql = \"\"\"\n/* multi\n line\n comment\n*/\nSELECT\n tbl.cola /* comment 1 */ + tbl.colb /* comment 2 */,\n CAST(x AS SIGNED), # comment 3\n y -- comment 4\nFROM\n bar /* comment 5 */,\n tbl # comment 6\n\"\"\"\n\n# Note: MySQL-specific comments (`#`) are converted into standard syntax\nprint(sqlglot.transpile(sql, read='mysql', pretty=True)[0])\n```\n\n```sql\n/* multi\n line\n comment\n*/\nSELECT\n tbl.cola /* comment 1 */ + tbl.colb /* comment 2 */,\n CAST(x AS INT), /* comment 3 */\n y /* comment 4 */\nFROM bar /* comment 5 */, tbl /* comment 6 */\n```\n\n\n### Metadata\n\nYou can explore SQL with expression helpers to do things like find columns and tables in a query:\n\n```python\nfrom sqlglot import parse_one, exp\n\n# print all column references (a and b)\nfor column in parse_one(\"SELECT a, b + 1 AS c FROM d\").find_all(exp.Column):\n print(column.alias_or_name)\n\n# find all projections in select statements (a and c)\nfor select in parse_one(\"SELECT a, b + 1 AS c FROM d\").find_all(exp.Select):\n for projection in select.expressions:\n print(projection.alias_or_name)\n\n# find all tables (x, y, z)\nfor table in parse_one(\"SELECT * FROM x JOIN y JOIN z\").find_all(exp.Table):\n print(table.name)\n```\n\nRead the [ast primer](https://github.com/tobymao/sqlglot/blob/main/posts/ast_primer.md) to learn more about SQLGlot's internals.\n\n### Parser Errors\n\nWhen the parser detects an error in the syntax, it raises a `ParseError`:\n\n```python\nimport sqlglot\nsqlglot.transpile(\"SELECT foo FROM (SELECT baz FROM t\")\n```\n\n```\nsqlglot.errors.ParseError: Expecting ). Line 1, Col: 34.\n SELECT foo FROM (SELECT baz FROM t\n ~\n```\n\nStructured syntax errors are accessible for programmatic use:\n\n```python\nimport sqlglot\ntry:\n sqlglot.transpile(\"SELECT foo FROM (SELECT baz FROM t\")\nexcept sqlglot.errors.ParseError as e:\n print(e.errors)\n```\n\n```python\n[{\n 'description': 'Expecting )',\n 'line': 1,\n 'col': 34,\n 'start_context': 'SELECT foo FROM (SELECT baz FROM ',\n 'highlight': 't',\n 'end_context': '',\n 'into_expression': None\n}]\n```\n\n### Unsupported Errors\n\nIt may not be possible to translate some queries between certain dialects. For these cases, SQLGlot may emit a warning and will proceed to do a best-effort translation by default:\n\n```python\nimport sqlglot\nsqlglot.transpile(\"SELECT APPROX_DISTINCT(a, 0.1) FROM foo\", read=\"presto\", write=\"hive\")\n```\n\n```sql\nAPPROX_COUNT_DISTINCT does not support accuracy\n'SELECT APPROX_COUNT_DISTINCT(a) FROM foo'\n```\n\nThis behavior can be changed by setting the [`unsupported_level`](https://github.com/tobymao/sqlglot/blob/b0e8dc96ba179edb1776647b5bde4e704238b44d/sqlglot/errors.py#L9) attribute. For example, we can set it to either `RAISE` or `IMMEDIATE` to ensure an exception is raised instead:\n\n```python\nimport sqlglot\nsqlglot.transpile(\"SELECT APPROX_DISTINCT(a, 0.1) FROM foo\", read=\"presto\", write=\"hive\", unsupported_level=sqlglot.ErrorLevel.RAISE)\n```\n\n```\nsqlglot.errors.UnsupportedError: APPROX_COUNT_DISTINCT does not support accuracy\n```\n\nThere are queries that require additional information to be accurately transpiled, such as the schemas of the tables referenced in them. This is because certain transformations are type-sensitive, meaning that type inference is needed in order to understand their semantics. Even though the `qualify` and `annotate_types` optimizer [rules](https://github.com/tobymao/sqlglot/tree/main/sqlglot/optimizer) can help with this, they are not used by default because they add significant overhead and complexity.\n\nTranspilation is generally a hard problem, so SQLGlot employs an \"incremental\" approach to solving it. This means that there may be dialect pairs that currently lack support for some inputs, but this is expected to improve over time. We highly appreciate well-documented and tested issues or PRs, so feel free to [reach out](#get-in-touch) if you need guidance!\n\n### Build and Modify SQL\n\nSQLGlot supports incrementally building SQL expressions:\n\n```python\nfrom sqlglot import select, condition\n\nwhere = condition(\"x=1\").and_(\"y=1\")\nselect(\"*\").from_(\"y\").where(where).sql()\n```\n\n```sql\n'SELECT * FROM y WHERE x = 1 AND y = 1'\n```\n\nIt's possible to modify a parsed tree:\n\n```python\nfrom sqlglot import parse_one\nparse_one(\"SELECT x FROM y\").from_(\"z\").sql()\n```\n\n```sql\n'SELECT x FROM z'\n```\n\nParsed expressions can also be transformed recursively by applying a mapping function to each node in the tree:\n\n```python\nfrom sqlglot import exp, parse_one\n\nexpression_tree = parse_one(\"SELECT a FROM x\")\n\ndef transformer(node):\n if isinstance(node, exp.Column) and node.name == \"a\":\n return parse_one(\"FUN(a)\")\n return node\n\ntransformed_tree = expression_tree.transform(transformer)\ntransformed_tree.sql()\n```\n\n```sql\n'SELECT FUN(a) FROM x'\n```\n\n### SQL Optimizer\n\nSQLGlot can rewrite queries into an \"optimized\" form. It performs a variety of [techniques](https://github.com/tobymao/sqlglot/blob/main/sqlglot/optimizer/optimizer.py) 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:\n\n```python\nimport sqlglot\nfrom sqlglot.optimizer import optimize\n\nprint(\n optimize(\n sqlglot.parse_one(\"\"\"\n SELECT A OR (B OR (C AND D))\n FROM x\n WHERE Z = date '2021-01-01' + INTERVAL '1' month OR 1 = 0\n \"\"\"),\n schema={\"x\": {\"A\": \"INT\", \"B\": \"INT\", \"C\": \"INT\", \"D\": \"INT\", \"Z\": \"STRING\"}}\n ).sql(pretty=True)\n)\n```\n\n```sql\nSELECT\n (\n \"x\".\"a\" <> 0 OR \"x\".\"b\" <> 0 OR \"x\".\"c\" <> 0\n )\n AND (\n \"x\".\"a\" <> 0 OR \"x\".\"b\" <> 0 OR \"x\".\"d\" <> 0\n ) AS \"_col_0\"\nFROM \"x\" AS \"x\"\nWHERE\n CAST(\"x\".\"z\" AS DATE) = CAST('2021-02-01' AS DATE)\n```\n\n### AST Introspection\n\nYou can see the AST version of the parsed SQL by calling `repr`:\n\n```python\nfrom sqlglot import parse_one\nprint(repr(parse_one(\"SELECT a + 1 AS z\")))\n```\n\n```python\nSelect(\n expressions=[\n Alias(\n this=Add(\n this=Column(\n this=Identifier(this=a, quoted=False)),\n expression=Literal(this=1, is_string=False)),\n alias=Identifier(this=z, quoted=False))])\n```\n\n### AST Diff\n\nSQLGlot can calculate the semantic 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:\n\n```python\nfrom sqlglot import diff, parse_one\ndiff(parse_one(\"SELECT a + b, c, d\"), parse_one(\"SELECT c, a - b, d\"))\n```\n\n```python\n[\n Remove(expression=Add(\n this=Column(\n this=Identifier(this=a, quoted=False)),\n expression=Column(\n this=Identifier(this=b, quoted=False)))),\n Insert(expression=Sub(\n this=Column(\n this=Identifier(this=a, quoted=False)),\n expression=Column(\n this=Identifier(this=b, quoted=False)))),\n Keep(\n source=Column(this=Identifier(this=a, quoted=False)),\n target=Column(this=Identifier(this=a, quoted=False))),\n ...\n]\n```\n\nSee also: [Semantic Diff for SQL](https://github.com/tobymao/sqlglot/blob/main/posts/sql_diff.md).\n\n### Custom Dialects\n\n[Dialects](https://github.com/tobymao/sqlglot/tree/main/sqlglot/dialects) can be added by subclassing `Dialect`:\n\n```python\nfrom sqlglot import exp\nfrom sqlglot.dialects.dialect import Dialect\nfrom sqlglot.generator import Generator\nfrom sqlglot.tokens import Tokenizer, TokenType\n\n\nclass Custom(Dialect):\n class Tokenizer(Tokenizer):\n QUOTES = [\"'\", '\"']\n IDENTIFIERS = [\"`\"]\n\n KEYWORDS = {\n **Tokenizer.KEYWORDS,\n \"INT64\": TokenType.BIGINT,\n \"FLOAT64\": TokenType.DOUBLE,\n }\n\n class Generator(Generator):\n TRANSFORMS = {exp.Array: lambda self, e: f\"[{self.expressions(e)}]\"}\n\n TYPE_MAPPING = {\n exp.DataType.Type.TINYINT: \"INT64\",\n exp.DataType.Type.SMALLINT: \"INT64\",\n exp.DataType.Type.INT: \"INT64\",\n exp.DataType.Type.BIGINT: \"INT64\",\n exp.DataType.Type.DECIMAL: \"NUMERIC\",\n exp.DataType.Type.FLOAT: \"FLOAT64\",\n exp.DataType.Type.DOUBLE: \"FLOAT64\",\n exp.DataType.Type.BOOLEAN: \"BOOL\",\n exp.DataType.Type.TEXT: \"STRING\",\n }\n\nprint(Dialect[\"custom\"])\n```\n\n```\n<class '__main__.Custom'>\n```\n\n### SQL Execution\n\nSQLGlot is able to interpret SQL queries, where the tables are represented as Python dictionaries. The engine is not supposed to be fast, but 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, such as [Arrow](https://arrow.apache.org/docs/index.html) and [Pandas](https://pandas.pydata.org/).\n\nThe example below showcases the execution of a query that involves aggregations and joins:\n\n```python\nfrom sqlglot.executor import execute\n\ntables = {\n \"sushi\": [\n {\"id\": 1, \"price\": 1.0},\n {\"id\": 2, \"price\": 2.0},\n {\"id\": 3, \"price\": 3.0},\n ],\n \"order_items\": [\n {\"sushi_id\": 1, \"order_id\": 1},\n {\"sushi_id\": 1, \"order_id\": 1},\n {\"sushi_id\": 2, \"order_id\": 1},\n {\"sushi_id\": 3, \"order_id\": 2},\n ],\n \"orders\": [\n {\"id\": 1, \"user_id\": 1},\n {\"id\": 2, \"user_id\": 2},\n ],\n}\n\nexecute(\n \"\"\"\n SELECT\n o.user_id,\n SUM(s.price) AS price\n FROM orders o\n JOIN order_items i\n ON o.id = i.order_id\n JOIN sushi s\n ON i.sushi_id = s.id\n GROUP BY o.user_id\n \"\"\",\n tables=tables\n)\n```\n\n```python\nuser_id price\n 1 4.0\n 2 3.0\n```\n\nSee also: [Writing a Python SQL engine from scratch](https://github.com/tobymao/sqlglot/blob/main/posts/python_sql_engine.md).\n\n## Used By\n\n* [SQLMesh](https://github.com/TobikoData/sqlmesh)\n* [Apache Superset](https://github.com/apache/superset)\n* [Dagster](https://github.com/dagster-io/dagster)\n* [Fugue](https://github.com/fugue-project/fugue)\n* [ibis](https://github.com/ibis-project/ibis)\n* [mysql-mimic](https://github.com/kelsin/mysql-mimic)\n* [Querybook](https://github.com/pinterest/querybook)\n* [Quokka](https://github.com/marsupialtail/quokka)\n* [Splink](https://github.com/moj-analytical-services/splink)\n\n## Documentation\n\nSQLGlot uses [pdoc](https://pdoc.dev/) to serve its API documentation.\n\nA hosted version is on the [SQLGlot website](https://sqlglot.com/), or you can build locally with:\n\n```\nmake docs-serve\n```\n\n## Run Tests and Lint\n\n```\nmake style # Only linter checks\nmake unit # Only unit tests (or unit-rs, to use the Rust tokenizer)\nmake test # Unit and integration tests (or test-rs, to use the Rust tokenizer)\nmake check # Full test suite & linter checks\n```\n\n## Benchmarks\n\n[Benchmarks](https://github.com/tobymao/sqlglot/blob/main/benchmarks/bench.py) run on Python 3.10.12 in seconds.\n\n| Query | sqlglot | sqlglotrs | sqlfluff | sqltree | sqlparse | moz_sql_parser | sqloxide |\n| --------------- | --------------- | --------------- | --------------- | --------------- | --------------- | --------------- | --------------- |\n| 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) |\n| 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) |\n| 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) |\n| 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) |\n\n\n## Optional Dependencies\n\nSQLGlot uses [dateutil](https://github.com/dateutil/dateutil) to simplify literal timedelta expressions. The optimizer will not simplify expressions like the following if the module cannot be found:\n\n```sql\nx + interval '1' month\n```\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "An easily customizable SQL parser and transpiler",
"version": "23.13.7",
"project_urls": {
"Homepage": "https://github.com/tobymao/sqlglot"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "8dca76838ca38ac358ba2d18983b2fd119b3a75fe44eead300863a221163337d",
"md5": "58172f1b0b4e8329b382423aba88d566",
"sha256": "58c39dd236e243eedd512c7e84f3d6e9da559e2559f33c425633f813b93dcbd7"
},
"downloads": -1,
"filename": "sqlglot-23.13.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "58172f1b0b4e8329b382423aba88d566",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 392053,
"upload_time": "2024-05-04T22:01:18",
"upload_time_iso_8601": "2024-05-04T22:01:18.692308Z",
"url": "https://files.pythonhosted.org/packages/8d/ca/76838ca38ac358ba2d18983b2fd119b3a75fe44eead300863a221163337d/sqlglot-23.13.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "6234f34392ce9e6461d41503851fe69ddf436c17b7a8dcb0b9778f0c5aa33f60",
"md5": "3c41fb144f6ba2ac8b0e1c67d0a6ee82",
"sha256": "e78b2fe345fd5187b7e895d01f51965a561d234804a0769a5a9ba6745015694b"
},
"downloads": -1,
"filename": "sqlglot-23.13.7.tar.gz",
"has_sig": false,
"md5_digest": "3c41fb144f6ba2ac8b0e1c67d0a6ee82",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 13093895,
"upload_time": "2024-05-04T22:01:25",
"upload_time_iso_8601": "2024-05-04T22:01:25.640290Z",
"url": "https://files.pythonhosted.org/packages/62/34/f34392ce9e6461d41503851fe69ddf436c17b7a8dcb0b9778f0c5aa33f60/sqlglot-23.13.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-05-04 22:01:25",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "tobymao",
"github_project": "sqlglot",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "sqlglot"
}
