Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/kayak/pypika
PyPika is a python SQL query builder that exposes the full richness of the SQL language using a syntax that reflects the resulting query. PyPika excels at all sorts of SQL queries but is especially useful for data analysis.
https://github.com/kayak/pypika
builder data functional python python3 pythonic query sql
Last synced: 4 days ago
JSON representation
PyPika is a python SQL query builder that exposes the full richness of the SQL language using a syntax that reflects the resulting query. PyPika excels at all sorts of SQL queries but is especially useful for data analysis.
- Host: GitHub
- URL: https://github.com/kayak/pypika
- Owner: kayak
- License: apache-2.0
- Created: 2016-07-06T14:08:50.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2024-10-21T15:43:38.000Z (about 2 months ago)
- Last Synced: 2024-10-29T09:52:29.507Z (about 1 month ago)
- Topics: builder, data, functional, python, python3, pythonic, query, sql
- Language: Python
- Homepage: http://pypika.readthedocs.io/en/latest/
- Size: 1.2 MB
- Stars: 2,511
- Watchers: 36
- Forks: 296
- Open Issues: 221
-
Metadata Files:
- Readme: README.rst
- Contributing: CONTRIBUTING.rst
- License: LICENSE.txt
- Codeowners: CODEOWNERS
Awesome Lists containing this project
- best-of-python - GitHub - 44% open · ⏱️ 26.04.2024): (Database Clients)
- awesome-rainmana - kayak/pypika - PyPika is a python SQL query builder that exposes the full richness of the SQL language using a syntax that reflects the resulting query. PyPika excels at all sorts of SQL queries but is especially us (Python)
- starred-awesome - pypika - PyPika is a SQL query builder with a pythonic syntax that doesn't limit the expressiveness of SQL (Python)
- jimsghstars - kayak/pypika - PyPika is a python SQL query builder that exposes the full richness of the SQL language using a syntax that reflects the resulting query. PyPika excels at all sorts of SQL queries but is especially us (Python)
README
PyPika - Python Query Builder
=============================.. _intro_start:
|BuildStatus| |CoverageStatus| |Codacy| |Docs| |PyPi| |License|
Abstract
--------What is |Brand|?
|Brand| is a Python API for building SQL queries. The motivation behind |Brand| is to provide a simple interface for
building SQL queries without limiting the flexibility of handwritten SQL. Designed with data analysis in mind, |Brand|
leverages the builder design pattern to construct queries to avoid messy string formatting and concatenation. It is also
easily extended to take full advantage of specific features of SQL database vendors.What are the design goals for |Brand|?
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^|Brand| is a fast, expressive and flexible way to replace handwritten SQL (or even ORM for the courageous souls amongst you).
Validation of SQL correctness is not an explicit goal of |Brand|. With such a large number of
SQL database vendors providing a robust validation of input data is difficult. Instead you are encouraged to check inputs you provide to |Brand| or appropriately handle errors raised from
your SQL database - just as you would have if you were writing SQL yourself... _intro_end:
Read the docs: http://pypika.readthedocs.io/en/latest/
Installation
------------.. _installation_start:
|Brand| supports python ``3.6+``. It may also work on pypy, cython, and jython, but is not being tested for these versions.
To install |Brand| run the following command:
.. code-block:: bash
pip install pypika
.. _installation_end:
Tutorial
--------.. _tutorial_start:
The main classes in pypika are ``pypika.Query``, ``pypika.Table``, and ``pypika.Field``.
.. code-block:: python
from pypika import Query, Table, Field
Selecting Data
^^^^^^^^^^^^^^The entry point for building queries is ``pypika.Query``. In order to select columns from a table, the table must
first be added to the query. For simple queries with only one table, tables and columns can be references using
strings. For more sophisticated queries a ``pypika.Table`` must be used... code-block:: python
q = Query.from_('customers').select('id', 'fname', 'lname', 'phone')
To convert the query into raw SQL, it can be cast to a string.
.. code-block:: python
str(q)
Alternatively, you can use the `Query.get_sql()` function:
.. code-block:: python
q.get_sql()
Tables, Columns, Schemas, and Databases
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^In simple queries like the above example, columns in the "from" table can be referenced by passing string names into
the ``select`` query builder function. In more complex examples, the ``pypika.Table`` class should be used. Columns can be
referenced as attributes on instances of ``pypika.Table``... code-block:: python
from pypika import Table, Query
customers = Table('customers')
q = Query.from_(customers).select(customers.id, customers.fname, customers.lname, customers.phone)Both of the above examples result in the following SQL:
.. code-block:: sql
SELECT id,fname,lname,phone FROM customers
An alias for the table can be given using the ``.as_`` function on ``pypika.Table``
.. code-block:: sql
customers = Table('x_view_customers').as_('customers')
q = Query.from_(customers).select(customers.id, customers.phone).. code-block:: sql
SELECT id,phone FROM x_view_customers customers
A schema can also be specified. Tables can be referenced as attributes on the schema.
.. code-block:: sql
from pypika import Table, Query, Schema
views = Schema('views')
q = Query.from_(views.customers).select(customers.id, customers.phone).. code-block:: sql
SELECT id,phone FROM views.customers
Also references to databases can be used. Schemas can be referenced as attributes on the database.
.. code-block:: sql
from pypika import Table, Query, Database
my_db = Database('my_db')
q = Query.from_(my_db.analytics.customers).select(customers.id, customers.phone).. code-block:: sql
SELECT id,phone FROM my_db.analytics.customers
Results can be ordered by using the following syntax:
.. code-block:: python
from pypika import Order
Query.from_('customers').select('id', 'fname', 'lname', 'phone').orderby('id', order=Order.desc)This results in the following SQL:
.. code-block:: sql
SELECT "id","fname","lname","phone" FROM "customers" ORDER BY "id" DESC
Arithmetic
""""""""""Arithmetic expressions can also be constructed using pypika. Operators such as `+`, `-`, `*`, and `/` are implemented
by ``pypika.Field`` which can be used simply with a ``pypika.Table`` or directly... code-block:: python
from pypika import Field
q = Query.from_('account').select(
Field('revenue') - Field('cost')
).. code-block:: sql
SELECT revenue-cost FROM accounts
Using ``pypika.Table``
.. code-block:: python
accounts = Table('accounts')
q = Query.from_(accounts).select(
accounts.revenue - accounts.cost
).. code-block:: sql
SELECT revenue-cost FROM accounts
An alias can also be used for fields and expressions.
.. code-block:: sql
q = Query.from_(accounts).select(
(accounts.revenue - accounts.cost).as_('profit')
).. code-block:: sql
SELECT revenue-cost profit FROM accounts
More arithmetic examples
.. code-block:: python
table = Table('table')
q = Query.from_(table).select(
table.foo + table.bar,
table.foo - table.bar,
table.foo * table.bar,
table.foo / table.bar,
(table.foo+table.bar) / table.fiz,
).. code-block:: sql
SELECT foo+bar,foo-bar,foo*bar,foo/bar,(foo+bar)/fiz FROM table
Filtering
"""""""""Queries can be filtered with ``pypika.Criterion`` by using equality or inequality operators
.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id, customers.fname, customers.lname, customers.phone
).where(
customers.lname == 'Mustermann'
).. code-block:: sql
SELECT id,fname,lname,phone FROM customers WHERE lname='Mustermann'
Query methods such as select, where, groupby, and orderby can be called multiple times. Multiple calls to the where
method will add additional conditions as.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id, customers.fname, customers.lname, customers.phone
).where(
customers.fname == 'Max'
).where(
customers.lname == 'Mustermann'
).. code-block:: sql
SELECT id,fname,lname,phone FROM customers WHERE fname='Max' AND lname='Mustermann'
Filters such as IN and BETWEEN are also supported
.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id,customers.fname
).where(
customers.age[18:65] & customers.status.isin(['new', 'active'])
).. code-block:: sql
SELECT id,fname FROM customers WHERE age BETWEEN 18 AND 65 AND status IN ('new','active')
Filtering with complex criteria can be created using boolean symbols ``&``, ``|``, and ``^``.
AND
.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id, customers.fname, customers.lname, customers.phone
).where(
(customers.age >= 18) & (customers.lname == 'Mustermann')
).. code-block:: sql
SELECT id,fname,lname,phone FROM customers WHERE age>=18 AND lname='Mustermann'
OR
.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id, customers.fname, customers.lname, customers.phone
).where(
(customers.age >= 18) | (customers.lname == 'Mustermann')
).. code-block:: sql
SELECT id,fname,lname,phone FROM customers WHERE age>=18 OR lname='Mustermann'
XOR
.. code-block:: python
customers = Table('customers')
q = Query.from_(customers).select(
customers.id, customers.fname, customers.lname, customers.phone
).where(
(customers.age >= 18) ^ customers.is_registered
).. code-block:: sql
SELECT id,fname,lname,phone FROM customers WHERE age>=18 XOR is_registered
Convenience Methods
"""""""""""""""""""In the `Criterion` class, there are the static methods `any` and `all` that allow building chains AND and OR expressions with a list of terms.
.. code-block:: python
from pypika import Criterion
customers = Table('customers')
q = Query.from_(customers).select(
customers.id,
customers.fname
).where(
Criterion.all([
customers.is_registered,
customers.age >= 18,
customers.lname == "Jones",
])
).. code-block:: sql
SELECT id,fname FROM customers WHERE is_registered AND age>=18 AND lname = "Jones"
Grouping and Aggregating
""""""""""""""""""""""""Grouping allows for aggregated results and works similar to ``SELECT`` clauses.
.. code-block:: python
from pypika import functions as fn
customers = Table('customers')
q = Query \
.from_(customers) \
.where(customers.age >= 18) \
.groupby(customers.id) \
.select(customers.id, fn.Sum(customers.revenue)).. code-block:: sql
SELECT id,SUM("revenue") FROM "customers" WHERE "age">=18 GROUP BY "id"
After adding a ``GROUP BY`` clause to a query, the ``HAVING`` clause becomes available. The method
``Query.having()`` takes a ``Criterion`` parameter similar to the method ``Query.where()``... code-block:: python
from pypika import functions as fn
payments = Table('payments')
q = Query \
.from_(payments) \
.where(payments.transacted[date(2015, 1, 1):date(2016, 1, 1)]) \
.groupby(payments.customer_id) \
.having(fn.Sum(payments.total) >= 1000) \
.select(payments.customer_id, fn.Sum(payments.total)).. code-block:: sql
SELECT customer_id,SUM(total) FROM payments
WHERE transacted BETWEEN '2015-01-01' AND '2016-01-01'
GROUP BY customer_id HAVING SUM(total)>=1000Joining Tables and Subqueries
"""""""""""""""""""""""""""""Tables and subqueries can be joined to any query using the ``Query.join()`` method. Joins can be performed with either
a ``USING`` or ``ON`` clauses. The ``USING`` clause can be used when both tables/subqueries contain the same field and
the ``ON`` clause can be used with a criterion. To perform a join, ``...join()`` can be chained but then must be
followed immediately by ``...on()`` or ``...using(*field)``.Join Types
~~~~~~~~~~All join types are supported by |Brand|.
.. code-block:: python
Query \
.from_(base_table)
...
.join(join_table, JoinType.left)
..... code-block:: python
Query \
.from_(base_table)
...
.left_join(join_table) \
.left_outer_join(join_table) \
.right_join(join_table) \
.right_outer_join(join_table) \
.inner_join(join_table) \
.outer_join(join_table) \
.full_outer_join(join_table) \
.cross_join(join_table) \
.hash_join(join_table) \
...See the list of join types here ``pypika.enums.JoinTypes``
Example of a join using `ON`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. code-block:: python
history, customers = Tables('history', 'customers')
q = Query \
.from_(history) \
.join(customers) \
.on(history.customer_id == customers.id) \
.select(history.star) \
.where(customers.id == 5).. code-block:: sql
SELECT "history".* FROM "history" JOIN "customers" ON "history"."customer_id"="customers"."id" WHERE "customers"."id"=5
As a shortcut, the ``Query.join().on_field()`` function is provided for joining the (first) table in the ``FROM`` clause
with the joined table when the field name(s) are the same in both tables.Example of a join using `ON`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. code-block:: python
history, customers = Tables('history', 'customers')
q = Query \
.from_(history) \
.join(customers) \
.on_field('customer_id', 'group') \
.select(history.star) \
.where(customers.group == 'A').. code-block:: sql
SELECT "history".* FROM "history" JOIN "customers" ON "history"."customer_id"="customers"."customer_id" AND "history"."group"="customers"."group" WHERE "customers"."group"='A'
Example of a join using `USING`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. code-block:: python
history, customers = Tables('history', 'customers')
q = Query \
.from_(history) \
.join(customers) \
.using('customer_id') \
.select(history.star) \
.where(customers.id == 5).. code-block:: sql
SELECT "history".* FROM "history" JOIN "customers" USING "customer_id" WHERE "customers"."id"=5
Example of a correlated subquery in the `SELECT`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.. code-block:: python
history, customers = Tables('history', 'customers')
last_purchase_at = Query.from_(history).select(
history.purchase_at
).where(history.customer_id==customers.customer_id).orderby(
history.purchase_at, order=Order.desc
).limit(1)
q = Query.from_(customers).select(
customers.id, last_purchase_at.as_('last_purchase_at')
).. code-block:: sql
SELECT
"id",
(SELECT "history"."purchase_at"
FROM "history"
WHERE "history"."customer_id" = "customers"."customer_id"
ORDER BY "history"."purchase_at" DESC
LIMIT 1) "last_purchase_at"
FROM "customers"Unions
""""""Both ``UNION`` and ``UNION ALL`` are supported. ``UNION DISTINCT`` is synonymous with ``UNION`` so |Brand| does not
provide a separate function for it. Unions require that queries have the same number of ``SELECT`` clauses so
trying to cast a unioned query to string will throw a ``SetOperationException`` if the column sizes are mismatched.To create a union query, use either the ``Query.union()`` method or `+` operator with two query instances. For a
union all, use ``Query.union_all()`` or the `*` operator... code-block:: python
provider_a, provider_b = Tables('provider_a', 'provider_b')
q = Query.from_(provider_a).select(
provider_a.created_time, provider_a.foo, provider_a.bar
) + Query.from_(provider_b).select(
provider_b.created_time, provider_b.fiz, provider_b.buz
).. code-block:: sql
SELECT "created_time","foo","bar" FROM "provider_a" UNION SELECT "created_time","fiz","buz" FROM "provider_b"
Intersect
"""""""""``INTERSECT`` is supported. Intersects require that queries have the same number of ``SELECT`` clauses so
trying to cast a intersected query to string will throw a ``SetOperationException`` if the column sizes are mismatched.To create a intersect query, use the ``Query.intersect()`` method.
.. code-block:: python
provider_a, provider_b = Tables('provider_a', 'provider_b')
q = Query.from_(provider_a).select(
provider_a.created_time, provider_a.foo, provider_a.bar
)
r = Query.from_(provider_b).select(
provider_b.created_time, provider_b.fiz, provider_b.buz
)
intersected_query = q.intersect(r).. code-block:: sql
SELECT "created_time","foo","bar" FROM "provider_a" INTERSECT SELECT "created_time","fiz","buz" FROM "provider_b"
Minus
"""""``MINUS`` is supported. Minus require that queries have the same number of ``SELECT`` clauses so
trying to cast a minus query to string will throw a ``SetOperationException`` if the column sizes are mismatched.To create a minus query, use either the ``Query.minus()`` method or `-` operator with two query instances.
.. code-block:: python
provider_a, provider_b = Tables('provider_a', 'provider_b')
q = Query.from_(provider_a).select(
provider_a.created_time, provider_a.foo, provider_a.bar
)
r = Query.from_(provider_b).select(
provider_b.created_time, provider_b.fiz, provider_b.buz
)
minus_query = q.minus(r)(or)
minus_query = Query.from_(provider_a).select(
provider_a.created_time, provider_a.foo, provider_a.bar
) - Query.from_(provider_b).select(
provider_b.created_time, provider_b.fiz, provider_b.buz
).. code-block:: sql
SELECT "created_time","foo","bar" FROM "provider_a" MINUS SELECT "created_time","fiz","buz" FROM "provider_b"
EXCEPT
""""""``EXCEPT`` is supported. Minus require that queries have the same number of ``SELECT`` clauses so
trying to cast a except query to string will throw a ``SetOperationException`` if the column sizes are mismatched.To create a except query, use the ``Query.except_of()`` method.
.. code-block:: python
provider_a, provider_b = Tables('provider_a', 'provider_b')
q = Query.from_(provider_a).select(
provider_a.created_time, provider_a.foo, provider_a.bar
)
r = Query.from_(provider_b).select(
provider_b.created_time, provider_b.fiz, provider_b.buz
)
minus_query = q.except_of(r).. code-block:: sql
SELECT "created_time","foo","bar" FROM "provider_a" EXCEPT SELECT "created_time","fiz","buz" FROM "provider_b"
Date, Time, and Intervals
"""""""""""""""""""""""""Using ``pypika.Interval``, queries can be constructed with date arithmetic. Any combination of intervals can be
used except for weeks and quarters, which must be used separately and will ignore any other values if selected... code-block:: python
from pypika import functions as fn
fruits = Tables('fruits')
q = Query.from_(fruits) \
.select(fruits.id, fruits.name) \
.where(fruits.harvest_date + Interval(months=1) < fn.Now()).. code-block:: sql
SELECT id,name FROM fruits WHERE harvest_date+INTERVAL 1 MONTH QueryBuilder:
if isinstance(col, str):
col = Field(col)return query.where(col > fn.Now() - num_days)
def count_groups(query: QueryBuilder, *groups) -> QueryBuilder:
return query.groupby(*groups).select(*groups, fn.Count("*").as_("n_rows"))base_query = Query.from_("table")
query = (
base_query
.pipe(filter_days, "date", num_days=7)
.pipe(count_groups, "col1", "col2")
)This produces:
.. code-block:: sql
SELECT "col1","col2",COUNT(*) n_rows
FROM "table"
WHERE "date">NOW()-7
GROUP BY "col1","col2".. _tutorial_end:
.. _contributing_start:
Contributing
------------We welcome community contributions to |Brand|. Please see the `contributing guide <6_contributing.html>`_ to more info.
.. _contributing_end:
.. _license_start:
License
-------Copyright 2020 KAYAK Germany, GmbH
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License athttp://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.Crafted with ♥ in Berlin.
.. _license_end:
.. _appendix_start:
.. |Brand| replace:: *PyPika*
.. _appendix_end:
.. _available_badges_start:
.. |BuildStatus| image:: https://github.com/kayak/pypika/workflows/Unit%20Tests/badge.svg
:target: https://github.com/kayak/pypika/actions
.. |CoverageStatus| image:: https://coveralls.io/repos/kayak/pypika/badge.svg?branch=master
:target: https://coveralls.io/github/kayak/pypika?branch=master
.. |Codacy| image:: https://api.codacy.com/project/badge/Grade/6d7e44e5628b4839a23da0bd82eaafcf
:target: https://www.codacy.com/app/twheys/pypika
.. |Docs| image:: https://readthedocs.org/projects/pypika/badge/?version=latest
:target: http://pypika.readthedocs.io/en/latest/
.. |PyPi| image:: https://img.shields.io/pypi/v/pypika.svg?style=flat
:target: https://pypi.python.org/pypi/pypika
.. |License| image:: https://img.shields.io/hexpm/l/plug.svg?maxAge=2592000
:target: http://www.apache.org/licenses/LICENSE-2.0.. _available_badges_end: