Re: [Virtuoso-users] Support for "float" literals

2008-01-21 Thread Christian Becker
Rajeev, in this case we're using properties from the W3C Geo Vocabulary (http://www.w3.org/2003/01/geo/ ) which implicitly uses floats. As the data will be published for others to use, I'd prefer to obey the ontology :) However, maybe you're right and we *should* be using double instead. I ju

Re: [Virtuoso-users] Support for "float" literals

2008-01-20 Thread Ivan Mikhailov
Hello Christian and RŠ°jeev, It is technically possible to enable xsd:float and check if it works really good at startup on given OS/compiler/runtime and dump a warning to the server log. I'll try and if it works for some of my boxes then I'll put it to next VOS release (that is expected soon). Be

Re: [Virtuoso-users] Support for "float" literals

2008-01-20 Thread Rajeev J Sebastian
Hi Christian, What Ivan is trying to say is that we should use xsd:double instead since it matches better with the conception of decimal-point numbers like geo-coordinates. Regards Rajeev J Sebastian On Jan 20, 2008 5:09 PM, Christian Becker wrote: > Hi Ivan, > I see that xsd:floats are not ret

Re: [Virtuoso-users] Support for "float" literals

2008-01-20 Thread Christian Becker
Hi Ivan, I see that xsd:floats are not returned for SPARQL queries, either. Note that one can't always choose another datatype - for DBpedia 3.0 we just changed all geocoordinates to xsd:float to match the related ontology. I'd just like to cast my vote for xsd:float support - apart from the

Re: [Virtuoso-users] Support for "float" literals

2008-01-19 Thread Rajeev J Sebastian
Hello Ivan, Thank you for your response. I really agree with your assessment there :) There was no specific reason for using xsd:float other than because of Javascript and JSON. I will try to find some workaround to use doublePrecision instead. Also, I would like your suggestions on how to imp

Re: [Virtuoso-users] Support for "float" literals

2008-01-19 Thread Ivan Mikhailov
Hello Rajeev, The strongly advised workaround is to use xsd:doublePrecision instead of xsd:float. The problem with float is that it is not enough precise. Depending on hardware platform and C library used by compiler same literal can be parsed and then printed as, say, "4.12" or "4.13", so

Re: [Virtuoso-users] Support for "float" literals

2008-01-19 Thread Rajeev J Sebastian
Hello Hugh, Thanks for the response! For a few days I thought I was going in the wrong direction :) Especially since the mailing list was so quiet. Regards Rajeev J Sebastian On Jan 20, 2008 2:20 AM, Hugh Williams wrote: > Hi Rajeev, > > This is the correct mailing list to use. Once a response

Re: [Virtuoso-users] Support for "float" literals

2008-01-19 Thread Hugh Williams
Hi Rajeev, This is the correct mailing list to use. Once a response is available from development a response will be posted, please bear with us ... Best Regards Hugh Williams OpenLink Software On 19/1/08 19:30, "Rajeev J Sebastian" wrote: > Hello all, > > I had sent the below attached email

Re: [Virtuoso-users] Support for "float" literals

2008-01-19 Thread Rajeev J Sebastian
Hello all, I had sent the below attached email to the list regarding an error I had come across, but there was no response. Is the virtuoso-users@lists.sourceforge.net the correct email address for such questions ? Regards Rajeev J Sebastian On Jan 19, 2008 5:51 AM, Rajeev J Sebastian wrote: >

[Virtuoso-users] Support for "float" literals

2008-01-18 Thread Rajeev J Sebastian
Hello all, I am running Virtuoso Version 05.00.3023-pthreads. When running a sparul something like: insert { "21.0396039604"^^. } I get an error, "37000 Error SP031: SPARQL compiler: Current implementation of SPARQL does not supports literals of type RE