Feature request enable use of chars other than [\-0-9a-zA-z] , e.g. IRIs in content variants

[JJ-Author]

Goal: 1. decide whether this feature is needed and makes sense and identify side effects, e.g. unix file name restrictions

At the moment databus throws an error: for invalid databus file id / IRI and distribution resource IRI generation (probably caused by autocompletion) due to missing url-escaping of content variant values in IRI generation

This is also kind of a bug, because SHACL should catch that.

i get following error message

{
  "isSuccess": false,
  "message": "{\"message\":\"[line: 0, col: 0 ] Wrong IRI 'https://databus.dev.dbpedia.link/jj-test/ex-group/exotic-filenames/2025-10-16#exotic-filenames_test=http://url.value/foo#bar'. com.apicatalog.jsonld.deseralization.JsonLdToRdf build\"}"
}

when posting below dataid version

curl -X POST https://databus.dev.dbpedia.link/api/register \
  -H "X-API-KEY: !!!missing!!!" \
  -H "Content-Type: application/json" \
  -d '{
  "@context": "https://databus.dev.dbpedia.link/res/context.jsonld",
  "@graph": [
    {
      "@id": "https://databus.dev.dbpedia.link/jj-test/ex-group/exotic-filenames/2025-10-16",
      "@type": "Version",
      "title": "initial-exotic-content-vars-and-values-test",
      "abstract": "initial test",
      "description": "try out different content variants",
      "license": "http://foo.lic",
      "distribution": [
        {
          "@type": "Part",
          "formatExtension": "none",
          "compression": "none",
          "downloadURL": "https://httpbin.org/anything?test=url-value",
          "dcv:test": "http://url.value/foo#bar"
        }
      ]
    }
  ]
}'

so when the IRIs are generated it should be taken care that appropriate encoding/escaping is applied on the values of the content variants (leaving the original value in this example "dcv:test": "http://url.value/foo#bar" of course untouched) before they will be appended / inserted into the IRIs.