Damian Steer wrote: > On 15/03/12 15:33, Paolo Castagna wrote: >> Hi, currently with Apache Jena you can run a SPARQL query and get >> your results, in case of a SELECT query, in text, XML, JSON, CSV or >> TSV format. > >> This is very convenient as people often want to use other tools to >> visualize or analyze their results. > >> Sometimes, people want to load data into a spreadsheet, either >> Microsoft Excel or OpenOffice/LibreOffice Calc. This can be >> currently achieved via CSV and TSV which, as far as I know, are >> supported by a lot of spreadsheet software (including Microsoft >> Excel and OpenOffice/LibreOffice). > >> For data publishers, it's a good idea to provide the same data in >> different formats, however, I am not sure there isn't a need for >> strong proprietary data formats (given we provide CSV and TSV >> already). > >> What do you think? > > I've done this fairly recently using Apache POI. They have a streaming > implementation which is fairly easy on memory.
Hi Damian, thanks for your reply. Is your code anywhere available and/or open source? I can write that, but why not re-using and/or improving what others have done it? (I doubt, I can improve your code though. Certainly, I can use it. And if we put it into Jena many others will do. ;-)) > > Not sure whether this is a good candidate for adding to the core, > however. Maybe include the code (it is, after all, small), but make > POI an optional dependency? Yeah, I though about this. I certainly do not want to add another dependency to jena-core and made it bigger. But, your idea of using an optional dependency should not cause big objections. Thanks, Paolo > > (Sorry, been on a training course all day. Brain not working) > > Damian
