Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/Wikimedia-Sverige/DCAT

A project aimed at generating a DCAT-AP for Wikibase installations in general and Wikidata in particular.
https://github.com/Wikimedia-Sverige/DCAT

Last synced: 5 days ago
JSON representation

A project aimed at generating a DCAT-AP for Wikibase installations in general and Wikidata in particular.

Awesome Lists containing this project

README 

        
DCAT-AP for Wikibase

=================



# Note

This repo stems from the time before the tool was integrated into the Wikimedia dumping infrastructure. The up-to-date repository can be found at [wikimedia/operations-dumps-dcat](https://github.com/wikimedia/operations-dumps-dcat).



---

A project aimed at generating a [DCAT-AP](https://joinup.ec.europa.eu/system/files/project/c3/22/18/DCAT-AP_Final_v1.00.html)

document for [Wikibase](http://wikiba.se) installations

in general and [Wikidata](http://wikidata.org) in particular.



Takes into account access through:



*   Content negotiation (various formats)

*   MediaWiki api (various formats)

*   Entity dumps e.g. json, ttl (assumes that these are compressed)



An example result can be found at [lokal-profil / dcatap.rdf](https://gist.github.com/lokal-profil/8086dc6bf2398d84a311).

The live DCAT-AP description of Wikidata can be found [here](https://dumps.wikimedia.org/wikidatawiki/entities/dcatap.rdf).



## To use



1.  Copy `config.example.json` to `config.json` and change the contents

    to match your installation. Refer to the *Config* section below for

    an explanation of the individual configuration parameters.

2.  Copy `catalog.example.json` to a suitable place (e.g. on-wiki) and

    update the translations to fit your wikibase installation. Set this

    value as `catalog-i18n` in the config file.

3.  Create the dcatap.rdf file by running `php DCAT.php` or

    `php DCAT.php --config="" --dumpDir="" --outputDir=""`

    where each of the options is optional and can be left out.

    The options are:

    1.  `--config` is the relative path to the json file containing the

        configurations, defaults to `./config.json`

    2.  `--dumpDir` is the relative path to the directory containing the

        dumps (if any), defaults to the `directory` parameter in the

        config file

    3.  `--outputDir` is the relative path to the directory where the

        `dcatap.rdf` file should be created, defaults to the `directory`

        parameter in the config file



## Translations



*   Translations which are generic to the tool are handled by [Intuition](https://github.com/Krinkle/intuition)

    and should be translated through [translatewiki.net](https://translatewiki.net).

*   Translations which are specific to a project/catalog are added to

    the location specified in the `catalog-i18n` parameter of the config

    file.



## Config



Below follows a key by key explanation of the config file.



*   `directory`: Relative path to the directory containing the dump

    subcategories (if any) and for the final dcat file.

*   `api-enabled`: (`Boolean`) Is API access activated for the MediaWiki

    installation?

*   `dumps-enabled`: (`Boolean`) Is JSON dump generation activated for the

    WikiBase installation?

*   `uri`: URL used as basis for rdf identifiers,

    e.g. *http://www.example.org/about*

*   `catalog-homepage`: URL for the homepage of the WikiBase installation,

    e.g. *http://www.example.org*

*   `catalog-issued`: ISO date at which the WikiBase installation was

    first issued, e.g. *2000-12-24*

*   `catalog-license`: License of the catalog, i.e. of the dcat file

    itself (not the contents of the WikiBase installation),

    e.g. *http://creativecommons.org/publicdomain/zero/1.0/*

*   `catalog-i18n`: URL or path to json file containing i18n strings for

    catalog title and description. Can be an on-wiki page,

    e.g. *https://www.example.org/w/index.php?title=MediaWiki:DCAT.json&action=raw*

*   `keywords`: (`array`) List of keywords applicable to all of the datasets

*   `themes`: (`array`) List of thematic ids in accordance with

    [Eurovoc](http://eurovoc.europa.eu/), e.g. *2191* for

    http://eurovoc.europa.eu/2191

*   `publisher`:

    *   `name`: Name of the publisher

    *   `homepage`: URL for or the homepage of the publisher

    *   `email`: Contact e-mail for the publisher, should be a function

        address, e.g. *[email protected]*

    *   `publisherType`: Publisher type according to [ADMS](http://purl.org/adms/publishertype/1.0),

        e.g. *NonProfitOrganisation*

*   `contactPoint`:

    *   `name`: Name of the contact point

    *   `email`: E-mail for the contact point, should ideally be a

        function address, e.g. *[email protected]*

    *   `vcardType`: Type of contact point, either `Organization` or

        `Individual`

*   `ld-info`:

    *   `accessURL`: URL to the content negotiation endpoint of the

        WikiBase installation, e.g. *http://www.example.org/entity/*

    *   `mediatype`: (`object`) List of [IANA media types](http://www.iana.org/assignments/media-types/)

        available through content negotiation in the format *file-ending:media-type*

    *   `license`: License of the data in the distribution, e.g.

        *http://creativecommons.org/publicdomain/zero/1.0/*

*   `api-info`:

    *   `accessURL`: URL to the MediaWiki API endpoint of the wiki,

        e.g. *http://www.example.org/w/api.php*

    *   `mediatype`: (`object`) List of non-deprecated formats available

        thorough the API, see ld-info:mediatype above for formatting

    *   `license`: See ld-info:license above

*   `dump-info`:

    *   `accessURL`: URL to the directory where the *.json.gz* files

        reside (`$1` is replaced on the fly by the actual filename),

        e.g. *http://example.org/dumps/$1*

    *   `mediatype`: (`object`) List of media types. e.g.

        `{"json": "application/json"}`

    *   `compression`: (`object`) List of compression formats, in the

        format *name:file-ending* e.g. `{"gzip": "gz"}`

    *   `license`: See ld-info:license above