Eberhard, Thanks a lot for the information. Especially as member also of JSON JSR EG I might play with the RI (check it out on java.net if you like, at least Open Source projects will NOT have to pay WURFL equivalent anual licence fees to Oracle for the JSON Spec;-) and see, what it can do to consume such service.
Agorava Project is already very JSON capable, that's another place where we hope to look into it. Oh and there's a meeting of Vienna.js, the local JavaScript User Group next week, I'll sure tell those guys and probably even propose it on the agenda:-D Thanks and Happy Holidays, Werner Am 21.12.2012 21:17 schrieb "eberhard speer jr." <[email protected]>: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > As announced here last week I have an "OpenDDR json service" up and > running on one of my servers here. > > It doesn't do anything particularly spectacular : it responds to, in > this case a DDR : Device Description Request, with a JSON formatted > key - value 'dictionary' of all the properties and their respective > values for the resolved device's UA, using the OpenDDR resources and code. > > What I did try to address is the question of how to get the > UA-to-resolve from the requester -- who also has it's own UA -- to the > DDR service. > > In a 'regular' RESTful service a 'query parameter' is part of the URL. > However in this case, an UA as a 'query parameter' in an URL raises a > number of issues : > > - - it's *ugly*, > - - it makes the 'service' hard to read thru it's URL, > - - the URL's length could become an issue, but most importantly : > - - UAs can include character sequences that can cause errors and > legitimate security concerns with the receiving server > > As a result, I proposed an extra HTTP request header field : > Ddr-User-Agent whose value is the UA-to-resolve. > (http://www.ducis.net/Static/rfc_ddr_api.htm) > > Since RESTful requests are constructed 'programatically' putting the > UA in an additional header rather then the URL shouldn't prove to be > much of a coding challenge. > > The service is described here (including code examples) : > http://www.ducis.net/Miri/Ddr > > You can test the service 'code-less' using a Chrome plugin like cRest > as shown in this screen-print. > http://www.ducis.net/Static/ddr_rest.png > > If you think this is a useful proposal, I'll gladly contribute the > documentation and code examples such are they are to the DeviceMap > repository. > > Regards, > > esjr > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.8 (MingW32) > Comment: Using GnuPG with undefined - http://www.enigmail.net/ > > iQEcBAEBAgAGBQJQ1MPPAAoJEOxywXcFLKYcT7QIAMJVHWJFJ4X9DB7fJelgBNW6 > fOCcXc9pezJn8RpQ/0Eg5Jfn5luBUgYLohtELyjx9qgSqD8yw+p8oKFtr0oMLx79 > UlHdkFDAcz+FJZtULzH4Jh+PeKKdFUJFICIy62m5bz6K5Qz/NI9AQiSlv477aLxu > yZk3F3dHCInSzdpUb5j9uCmTTmZe0YOYxvNa8TdK1YoGTVlrD18pY/Oz4L6ohwPW > 8GsHg/6l0IW7HvcO/L4oFVX4YAykwZwwlLuDZztGG5IS39FPcVbcJWewBoABCKju > JjMlRRWwKRdSbDKogNIlzyr1P41ORP7MG/ySkll1YgwNMaLUOIyDagodTRmD9Vw= > =B6rk > -----END PGP SIGNATURE----- >
