Postgres - Tag - Simon Jakubowskihttps://sijakubo.github.io/info/tags/postgres/Postgres - Tag - Simon JakubowskiHugo -- gohugo.ioensijakubo@gmail.com ( Simon Jakubowski)sijakubo@gmail.com ( Simon Jakubowski)Tue, 03 Mar 2026 00:00:00 +0000Forget CSV or Excel files. Share your Data with Apache Parquethttps://sijakubo.github.io/info/posts/post-12/Tue, 03 Mar 2026 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-12/What is Apache Parquet?

Apache Parquet is an open source, column-oriented data file format designed for efficient data storage and retrieval.

It provides:

  • High-performance compression
  • Efficient encoding schemes
  • Support for complex data types
  • Broad adoption across analytics tools and programming languages

Parquet is commonly used as a table replacement for analytical workloads.

How to Create Parquet Files

There are several ways to create Parquet files:

  • From CSV files
  • Directly from PostgreSQL using the pg_parquet extension
  • Using DuckDB
  • Using web-based CSV to Parquet converters

In this post, DuckDB is used as a lightweight and practical tool to generate and query Parquet files.

]]>Create a Feature Collection from Postgis Geometries using Postgis >= 3.0.0https://sijakubo.github.io/info/posts/post-8/Thu, 27 Apr 2023 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-8/Generating a Feature and Feature Collection form postgis is rather simple. If you convert a simple GEOMETRY with the st_asgeojson postgis will generate a GeoJSON “geometry”.

SELECT st_asgeojson(f.geometry)
FROM field f;

will result in a single geometry:

{
  "type" : "Polygon",
  "coordinates" : [...]
}

If you provide a RECORD (since Postgis 3.0.0), postgis will generate a Feature with all the fields as Properties:

SELECT st_asgeojson(f.*)
FROM field f;

will result in:

{
  "type" : "Feature",
  "geometry" : {
    "type" : "Polygon",
    "coordinates" : [
      ...
    ]
  },
  "properties" : {
    "id" : "052ff36a-4ea7-4307-9bc3-414f49a62163",
    "area_square_meters" : 29434,
    "altitude" : 40,
    "granule_code" : "31UGT",
    "creation_date" : "2021-10-21T23:32:25.974059",
    "name" : "Simple test field",
    ...
  }
}

Where the field table having the Columns: id, area_square_meters, …

]]>PostgreSQL Composite Types to reduce boilerplate code from you database schemahttps://sijakubo.github.io/info/posts/post-7/Thu, 30 Mar 2023 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-7/PostgreSQL is a powerful and flexible open-source relational database management system. One of its most useful features is the ability to define custom composite types (Since Postgres 11), which are user-defined data types that combine multiple fields of different types into a single logical entity. In this article, we’ll explore the benefits of using composite types in PostgreSQL and how to use them in your database schema.

Create Composite Type

To define a composite type in PostgreSQL, you use the CREATE TYPE statement, followed by the name of the type and a list of its component fields and types. For example, here’s how you might define a audit type:

]]>Zero downtime deployment with breaking DB-Schema change - by examplehttps://sijakubo.github.io/info/posts/post-6/Thu, 12 Jan 2023 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-6/When running multiple Server nodes, sharing the same Database, it becomes more and more important to handle backwards compatibility on the database especially during Deployments.

When we start da Deployment, a new Server node is being deployed to a cluster. This node will then migrate the Database to it’s designated schema version. If this new version is not backwards compatible, the currently running Server instances will run into several problems, reading or writing to the database.

]]>Testcontainers | Postgres: Postgis with Spring and Spockhttps://sijakubo.github.io/info/posts/post-5/Tue, 18 Oct 2022 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-5/A great way to include all the required resources for integration tests, Testcontainers provides an easy way to start docker containers from within your testcode. This removes the necessity to create test docker-compose.yml files and write README.md code on how to even run a test.

Here is a short example using testcontainers with Spring and Spock on a Postgis / Postgres database.

@SpringBootTest
@AutoConfigureMockMvc
@AutoConfigureTestDatabase(replace = AutoConfigureTestDatabase.Replace.NONE)
@ContextConfiguration(initializers = [Initializer.class])
class FieldControllerTest extends IntegrationTest {

  @Shared
  public static JdbcDatabaseContainer postgreSQLContainer = new PostgisContainerProvider()
    .newInstance()
    .withDatabaseName("integration-tests-db")
    .withUsername("sa")
    .withPassword("sa");


  static class Initializer implements ApplicationContextInitializer<ConfigurableApplicationContext> {

    void initialize(ConfigurableApplicationContext configurableApplicationContext) {
      postgreSQLContainer.start()
      TestPropertyValues.of(
        "spring.datasource.url=" + postgreSQLContainer.getJdbcUrl(),
        "spring.datasource.username=" + postgreSQLContainer.getUsername(),
        "spring.datasource.password=" + postgreSQLContainer.getPassword()
      ).applyTo(configurableApplicationContext.getEnvironment());
    }
  }

  @Autowired
  MockMvc mockMvc

  @Autowired
  FieldRepository fieldRepository

  void setup() {
    fieldRepository.save(new Field(
      UUID.fromString('ad68f894-c16b-4953-b577-7cddb3e85ae5'), "initSampleField",
      new Polygon(
        PositionSequenceBuilders.variableSized(G2D.class)
          .add(5.8208837124389, 51.0596004663904)
          .add(5.83490292265498, 51.0571257015788)
          .add(5.87078646658134, 51.0451607414904)
          .add(5.79146302423308, 51.0612386272784)
          .add(5.8208837124389, 51.0596004663904)
          .toPositionSequence(),
        CoordinateReferenceSystems.WGS84
      )
    ))
  }

  ...
}

To integrate testcontainers into your project, you’ll need the following dependencies:

]]>Postgres JSON beginner Classhttps://sijakubo.github.io/info/posts/post-4/Sat, 10 Jul 2021 00:00:00 +0000sijakubo@gmail.com ( Simon Jakubowski)https://sijakubo.github.io/info/posts/post-4/Operators see: https://www.postgresql.org/docs/9.5/functions-json.html

Create column using jsonb

CREATE TABLE article (
    id      UUID PRIMARY KEY,
    properties JSONB NOT NULL
);

Insert values

INSERT INTO article
VALUES (gen_random_uuid(), '{
  "price": 12.1,
  "name": "Kugelschreiber",
  "tags": {
    "manufacturer": "Siemens",
    "discounted": true
  }
}')

Select value from JSON

select
  properties -> 'price'
  properties ->> 'price'
from article
  • -> will extract the value as JSONB
  • ->> will extract the value as text

Compare values

SELECT *
FROM article
WHERE CAST(properties ->> 'price' AS NUMERIC) > 10

Check where value is contained

SELECT *
FROM article
WHERE properties -> 'tags' ? 'discounted'

or

]]>