LDBC SNB Documentation

[latest stable build (PDF)] [latest snapshot build (PDF)]

Query specifications

The query specifications are defined in YAML format.

Results are categorized according to their source of origin:

• raw (R), if the result is returned with an unmodified value and type. This is the default category.
• calculated (C), if the result is calculated from other values and conditions.
• aggregated (A), if the result is an aggregated value, e.g. a count or a sum of another value. If a result is both calculated and aggregated (e.g. count(x) + count(y) or avg(x + y)), it is considered an aggregated result.

Generating query cards

1. Install the following dependencies:

   • Python3:

     sudo apt-get install -y python3 python3-pip python3-setuptools
     

   • Pandoc:

     sudo apt-get install -y pandoc
     

2. To generate the TeX files for query cards, run the following command:

   ./generate-tex.py
   

Building the document

To build the document, run make or make texfot. The latter requires Perl but gives you a cleaner output.

If you are using a Linux-based system, you can also check the .travis.yml file as it provides a precise documentation of what you should do in order to build the docs.

Naming conventions

• if the attribute exists in the graph as part of an entity (vertex/edge) and is returned unchanged, name it entity.attributeName
• if it is computed or aggregated, use camelCase notation