Thor SPARQL Editor

A platform-agnostic, configurable, and brandable SPARQL editor and visualization interface.

Features

• Autocompletion of your entire ontology
• Result visualizations modes including; a table, an image grid, a pie chart, and a map
• Integration with your query library
• Shareable queries / URLs
• A resizable code editor
• An interactive tour of the GUI
• A configurable color scheme
• ...and plenty more...

Usage

Thor is a set of static files and all you need to get started is to serve it and a configuration file at config/config.json.

To create a configuration file you can start from an existing configuration or create on from scratch using the reference below.

Example configurations:

• FornPunkt
• TORA (National Archives of Sweden)

The FornPunkt project does contain a full configuration for the Caddy webserver showing how to serve Thor and the necessary configuration files. You can also read more about deploying static files in the guides below:

• NGINX: Serving Static Content
• Caddy: Static files quick-start

Not that nothing is stopping you form serving Thor configuration from a non-static source. You can therefore dynamicly change your endpoint or provide an interface for updating the configuration if you want to.

Configuration reference

Struture

{
    "sparql_endpoint": "",
    "query_library_endpoint": "",
    "title": "",
    "demo_tour": {
        "demo_query": "",
        "demo_query_cursor_position": {
            "line": 1,
            "ch": 1
        }
    },
    "map_background": {
        "url_template": "",
        "attribution": ""
    },
    "color_scheme": {
        "main_brand": "",
        "main_brand_darkened": "",
        "secondary_brand": "",
        "secondary_brand_darkened": "",
        "main_text": "",
        "main_text_lighted": "",
        "secondary_text": "",
        "background": "",
        "background_shaded": "",
        "border": ""
    },
    "favicons": {
        "favicon_success": "",
        "favicon_error": "",
        "favicon_progess": ""
    },
    "autocomplete": {
        "classes": [],
        "prefixes": [],
        "properties": [],
        "services": [],
        "uris": []
    }
}

sparql_endpoint

The SPARQL endpoint to use. If left out, the user will be able to input their own endpoint.

query_library_endpoint

The endpoint to use for the query library. If left out, the query library will be disabled.

title

The title of the editor.

demo_tour

An object for the demo tour configuration. If left out, the demo tour will be disabled. Note that the demo_query and demo_query_cursor_position can be used to trigger/demo autocompletion features as if a real user would be typing.

• demo_query: The query to use for the demo tour. Use \n for newlines.
• demo_query_cursor_position: The cursor position to use for the demo tour.
  • line: The line number for the cursor position.
  • ch: The character number for the cursor position.

map_background

Optional map background settings.

• url_template: URL template for map tiles. See URL template at leafletjs.com.
• attribution: String to use for map attribution, can take HTML.

autocomplete

An object for the autocompletion configuration.

• classes: An array of plain-class-URIs to autocomplete, http://schema.org/CreativeWork not schema:CreativeWork.
• prefixes: An array of prefixes to autocomplete, following SPARQL format, rdfs: <http://www.w3.org/2000/01/rdf-schema#>, etc.
• properties: An array of plain-property-URIs to autocomplete, http://www.w3.org/2000/01/rdf-schema#label not rdfs:label.
• services: An array of federated endpoints to autocomplete, http://dbpedia.org/sparql, etc.
• uris: An array of URIs to autocomplete.

color_scheme

An object for the color scheme configuration. These are all exposed as root CSS variables so you can tweak the defaults in the browser's dev tools before adding them to the configuration.

• main_brand
• main_brand_darkened
• secondary_brand
• secondary_brand_darkened
• main_text
• main_text_lighted
• secondary_text
• background
• background_shaded
• border

SPARQL magic-comments reference

Thor uses the same comment system as the Wikidata Query Service GUI to allow users to control the page title and default visualization.

defaultView

The defaultView comment allows users to pick one of the following as the default result visualization: Table, Map, PieChart, and ImageGrid.

Example: #defaultView:Map

Title

The Title comment sets the page title.

Example: #title:My best query

URL parameter reference

config

The config parameter allows a single Thor instance to use multiply configuirations. This parameter is only used if alternative configurations extists in the config directory. Note that the value of the parameter should be the configuration filename minus the file-suffix.

query (URI fragment)

The query URI fragment can be set to populate the editor with text. The text should be URL encoded. Example: #query=Hello%20World

Query libraries

A query library is a JSON file containing an array of queries. Each query is an object with the following properties:

• title: The title of the query.
• tags: An array of tags for the query.
• body: The query body.

It's common to generate this JSON file from a code-snippet library or from induvidual SPARQL query files. Here is an example, generating a library from .rq files using Python.

Example

[
  {
    "title": "List users by the number of monuments registered with a given tag",
    "tags": [
      "monuments",
      "users",
      "tags"
    ],
    "body": "PREFIX schema: <http://schema.org/>\n\nSELECT ?creator (COUNT(?monument) AS ?monument_count) WHERE {\n  BIND(<https://fornpunkt.se/tagg/stensattning> AS ?tag)\n  ?monument schema:keywords ?tag ;\n        a schema:CreativeWork ;\n        schema:creator ?creator .\n}\nGROUP BY ?creator\nORDER BY DESC(?monument_count)\n"
  },
]