Support xsd:int type coercion

[ebremer]

Describe the bug
Using titanium 1.1.0, importing triples with, RdfDocument.of with setUseNativeTypes to true is being ignored.

To Reproduce
_:B58e8ccf8X2D76bcX2D46caX2D9912X2Dec48f4fe6d83 http://iiif.io/api/image/2#height "4652"^^http://www.w3.org/2001/XMLSchema#int .
_:B58e8ccf8X2D76bcX2D46caX2D9912X2Dec48f4fe6d83 http://iiif.io/api/image/2#width "6324"^^http://www.w3.org/2001/XMLSchema#int .

with above yields:
"http://iiif.io/api/image/2#height": [
{
"@value": "4652",
"@type": "http://www.w3.org/2001/XMLSchema#int"
}
],
"http://iiif.io/api/image/2#width": [
{
"@value": "6324",
"@type": "http://www.w3.org/2001/XMLSchema#int"
}
]
Expected behavior
should have yielded:
"http://iiif.io/api/image/2#height": 4652,
"http://iiif.io/api/image/2#width": "6324"

[filip26]

Hi,
thank you for reporting that.xsd:int is not supported now as the spec lists only xsd:boolean, xsd:double, xsd:string and xsd:integer.

TODO

• print warning about an unsupported XSD type
• implement xsd:int support (signed 32-bit integer)

[ebremer]

The specification listed at https://www.w3.org/TR/json-ld11/, says "...all other numbers are interpreted as typed values with type xsd:integer". As no specific number types are listed, I would think it would be safe to assume that long, int, short, byte, nonNegativeInteger, etc, which are all subclassed from integer, could be supported without breaking any rules.