Core Tutorial

(This tutorial is greatly inspired from the SQLAlchemy SQL Expression Language Tutorial, which is recommended reading, eventually.)

This tutorial shows how to use the SQLAlchemy Expression Language (a.k.a. SQLAlchemy Core) with GeoAlchemy. As defined by the SQLAlchemy documentation itself, in contrast to the ORM’s domain-centric mode of usage, the SQL Expression Language provides a schema-centric usage paradigm.

Connect to the DB

For this tutorial we will use a PostGIS 2 database. To connect we use SQLAlchemy’s create_engine() function:

>>> from sqlalchemy import create_engine
>>> engine = create_engine('postgresql://gis:gis@localhost/gis', echo=True)

In this example the name of the database, the database user, and the database password, is gis.

The echo flag is a shortcut to setting up SQLAlchemy logging, which is accomplished via Python’s standard logging module. With it is enabled, we’ll see all the generated SQL produced.

The return value of create_engine is an Engine object, which respresents the core interface to the database.

Define a Table

The very first object that we need to create is a Table. Here we create a lake_table object, which will correspond to the lake table in the database:

>>> from sqlalchemy import Table, Column, Integer, String, MetaData
>>> from geoalchemy2 import Geometry
>>>
>>> metadata = MetaData()
>>> lake_table = Table('lake', metadata,
...     Column('id', Integer, primary_key=True),
...     Column('name', String),
...     Column('geom', Geometry('POLYGON'))
... )

This table is composed of three columns, id, name and geom. The geom column is a geoalchemy2.types.Geometry column whose geometry_type is POLYGON.

Any Table object is added to a MetaData object, which is a catalog of Table objects (and other related objects).

Create the Table

With our Table being defined we’re ready (to have SQLAlchemy) create it in the database:

>>> lake_table.create(engine)

Calling create_all() on metadata would have worked equally well:

>>> metadata.create_all(engine)

In that case every Table that’s referenced to by metadata would be created in the database. The metadata object includes one Table here, our now well-known lake_table object.

Reflecting tables

The reflection system of SQLAlchemy can be used on tables containing geoalchemy2.types.Geometry or geoalchemy2.types.Geography columns. In this case, the type must be imported to be registered into SQLAlchemy, even if it is not used explicitely.

>>> from geoalchemy2 import Geometry  # <= not used but must be imported
>>> from sqlalchemy import create_engine, MetaData
>>> engine = create_engine("postgresql://myuser:mypass@mydb.host.tld/mydbname")
>>> meta = MetaData()
>>> meta.reflect(bind=engine)

Insertions

We want to insert records into the lake table. For that we need to create an Insert object. SQLAlchemy provides multiple constructs for creating an Insert object, here’s one:

>>> ins = lake_table.insert()
>>> str(ins)
INSERT INTO lake (id, name, geom) VALUES (:id, :name, ST_GeomFromEWKT(:geom))

The geom column being a Geometry column, the :geom bind value is wrapped in a ST_GeomFromEWKT call.

To limit the columns named in the INSERT query the values() method can be used:

>>> ins = lake_table.insert().values(name='Majeur',
...                                  geom='POLYGON((0 0,1 0,1 1,0 1,0 0))')
...
>>> str(ins)
INSERT INTO lake (name, geom) VALUES (:name, ST_GeomFromEWKT(:geom))

Tip

The string representation of the SQL expression does not include the data placed in values. We got named bind parameters instead. To view the data we can get a compiled form of the expression, and ask for its params:

>>> ins.compile.params()
{'geom': 'POLYGON((0 0,1 0,1 1,0 1,0 0))', 'name': 'Majeur'}

Up to now we’ve created an INSERT query but we haven’t sent this query to the database yet. Before being able to send it to the database we need a database Connection. We can get a Connection from the Engine object we created earlier:

>>> conn = engine.connect()

We’re now ready to execute our INSERT statement:

>>> result = conn.execute(ins)

This is what the logging system should output:

INSERT INTO lake (name, geom) VALUES (%(name)s, ST_GeomFromEWKT(%(geom)s)) RETURNING lake.id
{'geom': 'POLYGON((0 0,1 0,1 1,0 1,0 0))', 'name': 'Majeur'}
COMMIT

The value returned by conn.execute(), stored in result, is a sqlalchemy.engine.ResultProxy object. In the case of an INSERT we can get the primary key value which was generated from our statement:

>>> result.inserted_primary_key
[1]

Instead of using values() to specify our INSERT data, we can send the data to the execute() method on Connection. So we could rewrite things as follows:

>>> conn.execute(lake_table.insert(),
...              name='Majeur', geom='POLYGON((0 0,1 0,1 1,0 1,0 0))')

Now let’s use another form, allowing to insert multiple rows at once:

>>> conn.execute(lake_table.insert(), [
...     {'name': 'Garde', 'geom': 'POLYGON((1 0,3 0,3 2,1 2,1 0))'},
...     {'name': 'Orta', 'geom': 'POLYGON((3 0,6 0,6 3,3 3,3 0))'}
...     ])
...

Tip

In the above examples the geometries are specified as WKT strings. Specifying them as EWKT strings is also supported.

Selections

Inserting involved creating an Insert object, so it’d come to no surprise that Selecting involves creating a Select object. The primary construct to generate SELECT statements is SQLAlchemy`s select() function:

>>> from sqlalchemy.sql import select
>>> s = select([lake_table])
>>> str(s)
SELECT lake.id, lake.name, ST_AsEWKB(lake.geom) AS geom FROM lake

The geom column being a Geometry it is wrapped in a ST_AsEWKB call when specified as a column in a SELECT statement.

We can now execute the statement and look at the results:

>>> result = conn.execute(s)
>>> for row in result:
...     print 'name:', row['name'], '; geom:', row['geom'].desc
...
name: Majeur ; geom: 0103...
name: Garde ; geom: 0103...
name: Orta ; geom: 0103...

row['geom'] is a geoalchemy2.types.WKBElement instance. In this example we just get an hexadecimal representation of the geometry’s WKB value using the desc property.

Spatial Query

As spatial database users executing spatial queries is of a great interest to us. There comes GeoAlchemy!

Spatial relationship

Using spatial filters in SQL SELECT queries is very common. Such queries are performed by using spatial relationship functions, or operators, in the WHERE clause of the SQL query.

For example, to find lakes that contain the point POINT(4 1), we can use this:

>>> from sqlalchemy import func
>>> s = select([lake_table],
               func.ST_Contains(lake_table.c.geom, 'POINT(4 1)'))
>>> str(s)
SELECT lake.id, lake.name, ST_AsEWKB(lake.geom) AS geom FROM lake WHERE ST_Contains(lake.geom, :param_1)
>>> result = conn.execute(s)
>>> for row in result:
...     print 'name:', row['name'], '; geom:', row['geom'].desc
...
name: Orta ; geom: 0103...

GeoAlchemy allows rewriting this more concisely:

>>> s = select([lake_table], lake_table.c.geom.ST_Contains('POINT(4 1)'))
>>> str(s)
SELECT lake.id, lake.name, ST_AsEWKB(lake.geom) AS geom FROM lake WHERE ST_Contains(lake.geom, :param_1)

Here the ST_Contains function is applied to lake.c.geom. And the generated SQL the lake.geom column is actually passed to the ST_Contains function as the first argument.

Here’s another spatial query, based on ST_Intersects:

   >>> s = select([lake_table],
   ...            lake_table.c.geom.ST_Intersects('LINESTRING(2 1,4 1)'))
   >>> result = conn.execute(s)
   >>> for row in result:
   ...     print 'name:', row['name'], '; geom:', row['geom'].desc
   ...
   name: Garde ; geom: 0103...
   name: Orta ; geom: 0103...

This query selects lakes whose geometries intersect ``LINESTRING(2 1,4 1)``.

The GeoAlchemy functions all start with ST_. Operators are also called as functions, but the names of operator functions don’t include the ST_ prefix.

As an example let’s use PostGIS’ && operator, which allows testing whether the bounding boxes of geometries intersect. GeoAlchemy provides the intersects function for that:

>>> s = select([lake_table],
...            lake_table.c.geom.intersects('LINESTRING(2 1,4 1)'))
>>> result = conn.execute(s)
>>> for row in result:
...     print 'name:', row['name'], '; geom:', row['geom'].desc
...
name: Garde ; geom: 0103...
name: Orta ; geom: 0103...

Processing and Measurement

Here’s a Select that calculates the areas of buffers for our lakes:

>>> s = select([lake_table.c.name,
                func.ST_Area(
                    lake_table.c.geom.ST_Buffer(2)).label('bufferarea')])
>>> str(s)
SELECT lake.name, ST_Area(ST_Buffer(lake.geom, %(param_1)s)) AS bufferarea FROM lake
>>> result = conn.execute(s)
>>> for row in result:
...     print '%s: %f' % (row['name'], row['bufferarea'])
Majeur: 21.485781
Garde: 32.485781
Orta: 45.485781

Obviously, processing and measurement functions can also be used in WHERE clauses. For example:

>>> s = select([lake_table.c.name],
               lake_table.c.geom.ST_Buffer(2).ST_Area() > 33)
>>> str(s)
SELECT lake.name FROM lake WHERE ST_Area(ST_Buffer(lake.geom, :param_1)) > :ST_Area_1
>>> result = conn.execute(s)
>>> for row in result:
...     print row['name']
Orta

And, like any other functions supported by GeoAlchemy, processing and measurement functions can be applied to geoalchemy2.elements.WKBElement. For example:

>>> s = select([lake_table], lake_table.c.name == 'Majeur')
>>> result = conn.execute(s)
>>> lake = result.fetchone()
>>> bufferarea = conn.scalar(lake[lake_table.c.geom].ST_Buffer(2).ST_Area())
>>> print '%s: %f' % (lake['name'], bufferarea)
Majeur: 21.485781

Use Raster functions

A few functions (like ST_Transform(), ST_Union(), ST_SnapToGrid(), …) can be used on both geoalchemy2.types.Geometry and geoalchemy2.types.Raster types. In GeoAlchemy2, these functions are only defined for Geometry as it can not be defined for several types at the same time. Thus using these functions on Raster requires minor tweaking to enforce the type by passing the type_=Raster argument to the function:

>>> s = select([func.ST_Transform(
                    lake_table.c.raster,
                    2154,
                    type_=Raster)
                .label('transformed_raster')])

Further Reference