LUX Middle Tier code for Qlever

Installation

Install Qlever

See the Qlever installation instructions.

For MacOS:

• Install homebrew: see https://brew.sh/
• brew tap qlever-dev/qlever
• brew install qlever

For Linux, from source:

• See installing qlever

Install the Qlever UI: (optional but very useful)

• With docker: just wait and do qlever ui

• Locally:

  • git clone https://github.com/qlever-dev/qlever-ui.git
  • cd qlever-ui
  • npm install
  • npm run build
  • python3 -m venv QLEVERUI
  • source QLEVERUI/bin/activate
  • Check pyproject.toml for any changes to dependencies but...
  • pip install django==5.2.12 django-environ==0.13.0 djangorestframework django-import-export==4.4.0 gunicorn==25.1.0 markdown requests==2.32.5 whitenoise[brotli]==6.12.0 pyyaml
  • python manage.py makemigrations --merge && python manage.py migrate
  • ./manage.py createsuperuser
  • ./manage.py runserver localhost:8146

• Login to the admin console and configure the lux endpoint to use localhost:7010

Index and Serve the LUX data

Create a directory for the Qlever data:

• mkdir qlever

Copy the configuration files from the repo:

• cp files/Qleverfile files/Qleverfile-ui.yml lux.settings.json qlever/

Move the data somewhere accessible (and edit Qleverfile to point to it)

• mkdir data
• cp /path/to/triples/data/*gz data/

Index the data, and create the materialized views:

• cd qlever
• qlever index
• ... wait ...

Serve the data via SPARQL at :7010:

• qlever start

Start the UI locally, rather than via docker:

• ???

Running qleverlux

Install luxql: https://github.com/project-lux/luxql/ And pip install -e .

python -m qleverlux.server --help

Notes ... ignore from here

PREFIX lux: <https://lux.collections.yale.edu/ns/>
PREFIX ogc: <http://www.opengis.net/rdf#>
PREFIX osmrel: <https://www.openstreetmap.org/relation/>
PREFIX geo: <http://www.opengis.net/ont/geosparql#>
PREFIX osmkey: <https://www.openstreetmap.org/wiki/Key:>
PREFIX geof: <http://www.opengis.net/def/function/geosparql/>
PREFIX qlss: <https://qlever.cs.uni-freiburg.de/spatialSearch/>

SELECT ?where ?coords WHERE {
  BIND( "POINT(174.763336 -36.848461)"^^geo:wktLiteral AS ?akl )

  SERVICE qlss: {
    _:config  qlss:algorithm qlss:s2 ;
              qlss:left ?akl ;
              qlss:right ?coords ;
              qlss:numNearestNeighbors 20 ;
              qlss:maxDistance 5000 ;
              qlss:bindDistance ?dist_left_right ;
              qlss:payload ?where  .
    {
      ?where lux:placeDefinedBy ?coords .
    }
  }
}

Query to generate a materialized view for item words:

PREFIX lux: <https://lux.collections.yale.edu/ns/> 
SELECT ?word ?uri ?score ?tf WHERE { 
  { ?uri lux:itemPrimaryName ?text BIND (14 AS ?weight) } 
  UNION 
  { ?uri lux:recordText ?text BIND (5 AS ?weight) } 
  UNION 
  { ?uri lux:itemAny/lux:primaryName ?text BIND (1 AS ?weight) } 
  ?uri a lux:Item . 
  GRAPH ?tf { ?text ql:has-word ?word } 
  BIND (?tf * ?weight AS ?score) }

And then the query:

PREFIX view: <https://qlever.cs.uni-freiburg.de/materializedView/>
PREFIX lux: <https://lux.collections.yale.edu/ns/>

SELECT ?subject (SUM(?s1 + ?s2 + ?s3) AS ?score) WHERE {
  SERVICE view:itemWords { [ view:column-word "dort" ; view:column-uri ?subject; view:column-score ?s1 ] }
  SERVICE view:itemWords { [ view:column-word "turner" ; view:column-uri ?subject; view:column-score ?s2 ] }
  SERVICE view:itemWords { [ view:column-word "painting" ; view:column-uri ?subject; view:column-score ?s3 ] }

} GROUP BY ?subject ORDER BY DESC(?score)