This subdirectory contains various self-contained projects and snippets that can be used as templates for your own experiments with Splitgraph.

Each example has a README file. Some of these examples get or push data from/to the Splitgraph registry at data.splitgraph.com and so require you to be logged into it.

• import-from-csv: Import data from a CSV file into a Splitgraph image.
• import-from-mongo: Import data from MongoDB into Splitgraph.
• pg-replication: Use Splitgraph as a PostgreSQL logical replication client.
• push-to-other-engine: Share data with other Splitgraph engines.
• push-to-object-storage: Upload data to S3-compatible object storage when pushing to another Splitgraph engine.
• iris: Manipulate and query Splitgraph data from a Jupyter notebook.
• bloom-filter: Showcase using bloom filters to query large datasets with a limited amount of cache.
• splitfiles: Use Splitfiles to build Splitgraph data images, track their provenance and keep them up to date.
• splitgraph-cloud: Publish data on Splitgraph Cloud and try out the REST API provided by PostgREST that gets generated for every dataset on there.
• postgrest: Run PostgREST locally against the Splitgraph engine.
• us-election: A real-world Splitfile example that joins multiple datasets.
• benchmarking: A collection of Jupyter notebooks benchmarking various aspects of Splitgraph against synthetic and real-world datasets.
• dbt: Use the dbt CLI against the Splitgraph engine, enriching your dbt-built datasets with versioning, sharing and packaging capabilities.
• postgis: Use Splitgraph, PostGIS and GeoPandas to plot geospatial data.
• pgadmin: Use pgAdmin with Splitgraph.
• sample_splitfiles: A collection of loose Splitfiles that run against interesting datasets on Splitgraph Cloud.

The template example in templates has a sample example.yaml file. Alternatively, you can copy one of the existing examples. Note that most examples that use the example.yaml format and don't require logging into Splitgraph are tested with a suite in the test subdirectory.

In addition, all examples have Asciinema casts generated for them automatically at release time which are then available to be embedded into the website, see the script.