Support Tag2Link rules to get URLs for a tag

[pyrog]

See Tag2Link and JOSM ticket #21988.

Also an appropriate SPARQL query could return a format constraint as a regular expression.
See SIRET number on wikidata.

See also #140 (comment).

[Zverik]

No. There will be no wikidata, sparql or any other obscure things in this editor. The editor is for collecting data, not for in-depth maintenance of mapped objects. The less fields the better.

[pyrog]

Could we discuss this point ?

The editor is for collecting data,
👍

One could check data on the field, like Facebook or official identifiers like SIRET.

SPARQL is difficult to learn but that's not the point.

• Tag2Link is a universal way to add links, open them (and check identifiers values).
  It is used at least by JOSM and Osmose Q&A.
• You could get a simple json from this project.
• Or query SPARQL servers only when you rebuild the project.

The less fields the better.
👍

One could use the future profile system to display only what he need.
iD templates suggest other tags that aren't useful in my country for example.

• gnis:feature_id for a power substation is useless (USA only). We use ref:FR:gdo instead.

Tag2Links rules could be specific by country or by language (not used yet but wikidata and SPARQL allow this).

[Zverik]

Thing is, this editor is intended to be used in the field. You tap a button, fill as much info as possible in a short timeframe, and move to the next amenity. This proposal introduces a dependency on a web service (of which the app has just two: imagery and OSM, both optional when surveying). I do not want that, since it rises a question, why package all the presets and imagery lists when I could download them when needed.

Also this is a working plugin for JOSM, so people who are interested (and I believe there are less than 100 of them in the world) could use it to mass-add needed information.

[pyrog]

This proposal introduces a dependency on a web service

No if rules are packaged with the app: a script run SPARQL queries and store data in a json file.
The script could be run manually or run automatically by continuous integration workflows.

(JOSM do SPARQL queries in the past, now it use webjar. A script run manually "build" the index.json file stored in this webjar).

But if Every Door will have custom presets (see #119) then tag2link is not mandatory 😃