Bug #201

Accept-Charset HTTP header not honored

Added by Gregg - about 9 years ago. Updated about 9 years ago.

Status:New Start date:
Priority:High Due date:
Assignee:Paula Gearon % Done:

0%

Category:Mulgara
Target version:General development
Resolution:

Description

(This ticket obsoletes #197)

Currently there doesn't seem to be a way to request a specific character encoding for query results. I think the way to do this is via content negotiation; at least I haven't seen any way to make such a request in SPARQL, but I might be wrong about that. In any case, I need a way to definitely indicate that I want utf-8 for the results.

The Mulgara HTTP interfaces (version 2.1.3) don't seem to honor the Accept-Charset header. Ideally it should be possible to use the header to stipulate any encoding for XML results. For JSON results, it should be possible to stipulate any UTF (utf-8, utf-16, utf-32) or UCS (UCS-4; I understand UCS-2 is obsolete.)

History

Updated by Gregg - about 9 years ago

Replying to gar:

(This ticket obsoletes #197)

The Mulgara HTTP interfaces (version 2.1.3) don't seem to honor the Accept-Charset header. Ideally it should be possible to use the header to stipulate any encoding for XML results. For JSON results, it should be possible to stipulate any UTF (utf-8, utf-16, utf-32) or UCS (UCS-4; I understand UCS-2 is obsolete.)

Clarification: Accept-Charset can (should?) only be set by the User Agent, not the User. See Section 4.6.2 of the XMLHttpRequest draft; in fact the UA is obliged to terminate an AJAX request if the user tries to set it. So I guess the only way to indicate a preferred results encoding is to use the charset= subfield of the Accept header.

Also available in: Atom PDF