SPARQL jsonld endpoint still returns incorrect ISO-8859-1 charset in response content-type header

[Barry Pearce]

Hi folks,

[ Using the endpoint: http://vocab.getty.edu/sparql.jsonld ]

Although reported back in 2016 (https://groups.google.com/g/gettyvocablod/c/8jvZ-2ZhRj0/m/8mVjNEDiCAAJ ) and the thread suggesting this had been fixed, the endpoint still states the content-type is "application/ld+json;charset=ISO-8859-1".

The content is actually UTF-8.

Specifying UTF-8 in the Accept header does not alter the response.

With many clients (certainly in Java the standard Java HttpResponse BodyHandlers and the Spring Frfamework RestTemplate) use the content-type header to provide automatic charset conversion to the unicode internal string representation. In this case I worked around it by forcing the charset on the BodyHandler.

The 2018 fix seems to have reverted. The workaround works for me, but it is probably good to resolve.

Cheers
Barry

[Getty Vocabularies LOD]

Thanks, Barry.

We will fix this issue in the next version of web site.

Gregg