- From: Richard Light <richard@light.demon.co.uk>
- Date: Fri, 2 Mar 2018 19:44:32 +0000
- To: SW-forum <semantic-web@w3.org>
- Message-ID: <9955e807-b359-0673-3c4f-49927aaf9fe1@light.demon.co.uk>
Hi, I have set up a simple Linked Data application which uses an Active Server Page to parse the initial URL and decide what resource to return and in what format. It uses '303 See Other' responses to redirect to the required serialization, in the approved Linked Data manner. I have always used curl to test low-level responses, and Firefox as my browser, and it all works happily in those environments. Recently I tried out my application with other browsers: Edge, Chrome and Opera. To my surprise, the redirection of the default URL gave different results from Firefox. On two of the browsers, it returned the XML version of the data, and on the other it attempted to return an image (yes, I have stretched the definition of 'linked data' to suit my own ends :-)). So in one case the original URL http://wt-modes/Object/WTcoll/id/GRMDC.B199 was converted to http://wt-modes/Object/WTcoll/id/xml/GRMDC.B199 and in the others to http://wt-modes/Object/WTcoll/id/jpeg/GRMDC.B199 My initial reaction was to assume that the problem was with the '303 See Other' redirection, so I rewrote the ASP code to deliver the HTML response directly when no serialization was specified in the initial URL. However, this had no effect on the above results. Given that ASP generates code server-side, and curl reports this: C:\Program Files\curl\bin>curl -v "http://wt-modes/Object/WTcoll/id/GRMDC.B199" * Trying fe80::6dd9:15e0:b7e0:91a... * TCP_NODELAY set * Connected to wt-modes (fe80::6dd9:15e0:b7e0:91a) port 80 (#0) > GET /Object/WTcoll/id/GRMDC.B199 HTTP/1.1 > Host: wt-modes > User-Agent: curl/7.58.0 > Accept: */* > < HTTP/1.1 200 OK < Cache-Control: private < Content-Type: text/html < Expires: Fri, 02 Mar 2018 19:27:01 GMT < Server: Microsoft-IIS/10.0 < Set-Cookie: ASPSESSIONIDAQQDBQAS=OGENELKAMKODCMJBONNGGHIK; path=/ < X-Powered-By: ASP.NET < Date: Fri, 02 Mar 2018 19:27:01 GMT < Content-Length: 1439 < I fail to see how these browsers could generate such a different outcome from Firefox. Does this chime with anyone else's experience, and does anyone have a clue to offer me? Many thanks, Richard -- *Richard Light*
Received on Friday, 2 March 2018 19:45:03 UTC