Re: [Geotools-devel] [netcdf-java] runtime aggregation

Hello Daniele, In the meantime, I have gotten feedback from the netcdf-java community that it is up to the user to put the aggregated files in the proper order. That still leaves us with the lack of support for decreasing order in geotools. I made a patch and added it to my aggregation

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Hi Daniele, Did you have any more comments on my last email? Would it make sense to convert all Strings to Dates, and sort them , in order to find the minimum and maximum? Isn't that a bit expensive (and repetitive) for those two methods (you'd be doing it twice while doing it once provides

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Yeah, still that implies that the order of the files specified in the aggregation must be the same as the order of values for the aggregated dimension inside those files themselves. I can find no such specification though, but maybe I am missing it. It seems that in the examples on

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Hi Niels, On Wed, Sep 28, 2016 at 11:05 AM, Niels Charlier wrote: > On 23-09-16 17:03, Daniele Romagnoli wrote: > > I think that once the aggregation is in place, times can be in any order > as you reported, so I think it's time to revisit the timeVariable min/max > computation

Re: [Geotools-devel] [netcdf-java] runtime aggregation

On 23-09-16 17:03, Daniele Romagnoli wrote: I think that once the aggregation is in place, times can be in any order as you reported, so I think it's time to revisit the timeVariable min/max computation (not sure there are other methods around based on that supposition. I didn't check it yet).

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Hi Niels, On Fri, Sep 23, 2016 at 4:23 PM, Niels Charlier wrote: > HI Daniele, > > Thank you for your quick response. > > Two remarks though > > - I do seem to get from your explanation that it can also be decreasing > rather than increasing, which means it should at least be

Re: [Geotools-devel] [netcdf-java] runtime aggregation

HI Daniele, Thank you for your quick response. Two remarks though - I do seem to get from your explanation that it can also be decreasing rather than increasing, which means it should at least be possible to work the other way around, which was the case here. There are only two files, the

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Hi Niels, On Fri, Sep 23, 2016 at 2:40 PM, Niels Charlier wrote: > Simone, Daniele, > > Do you have an opinion on point (3) of this email? > > It seems that one of you have written the code that expects a > chronological order of listed aggregation files. > > The logic is in

Re: [Geotools-devel] [netcdf-java] runtime aggregation

Ben, Sean, I am also including the geotools list in the discussion, because it regards both libraries. I got the aggregation file to work. Apart from some obvious bug fixes and additions to support the "runtime" dimension in geotools, I had to make a few changes to the xml file. What is