Thank you, Jeff. Indeed I request people on this thread to review Ticket 145. Several people expressed a strong opinion that we should not veer from a narrow interpretation of netCDF files being "self-contained". The exception was allowed for cell_measures because of the impact on filesize and performance of 3D and 4D volume fields which would have to be repeated in every file that used them as cell measures.
On Thu, Apr 6, 2017 at 12:18 PM, Painter, Jeff <[email protected]> wrote: > The sentence at issue comes from Ticket 145 on the Trac system, > https://cf-trac.llnl.gov/trac/ticket/145. The final version of this > ticket's changes is in comment 47. It would be helpful if any revisions > could get approved (in a re-opened version of this ticket, or in a new > ticket) in time for Tanya and I to incorporate it in version 1.7. > > - Jeff Painter > > On 4/5/17 4:16 AM, Bärring Lars wrote: > > Hello Martin, David, all, > > Martin raises an relevant point, even though it took me minute or two to see > the difference. > > Now :"The only CF standard attribute which is allowed to refer to external > variables is cell_measures." > meaning that that only the CF standard attribute cell_measures is allowed to > make reference external variables (as well as any non-CF standard attributes > of course). > > Martins suggestion: "The only attribute which is allowed to make a CF > standard reference to external variables is cell_measures" > allows any of the 'free text' CF standard attributes to refer to external > variables, but such reference will not be understood by software (as anything > else in the free text attributes) > > In both cases the intended meaning is preserved that a cell_method reference > to an external variable will be understood by software. > > I support Martin's suggestion. > > Kind regards, > Lars > > > > > -----Original Message----- > From: CF-metadata [mailto:[email protected] > <[email protected]>] On Behalf Of [email protected] > Sent: den 5 april 2017 12:57 > To: [email protected] > Cc: [email protected] > Subject: Re: [CF-metadata] Overstated restriction on references to external > variables in version 1.7 > > Hi David, > > my point is indeed that you should be able to put anything in them, so have a > standards document which says that you can't put references to external > variables in them would be a mistake ..... > > cheers, > Martin > > ________________________________ > From: David Hassell [[email protected]] > Sent: 05 April 2017 11:35 > To: Juckes, Martin (STFC,RAL,RALSP) > Cc: CF Metadata > Subject: Re: [CF-metadata] Overstated restriction on references to external > variables in version 1.7 > > Hello Martin, > > I don't quite see what you mean - such attributes are non-standardised, so > you can already put anything you like in them, but software won't be able to > do anything about it. Could you perhaps give a CDL example? > > Many thanks, > > David > > On 5 April 2017 at 09:36, > <[email protected]<mailto:[email protected]> > <[email protected]>> wrote: > Hello All, > > Section 2.6.3, External Variables, of version 1.7 of the standard contains > the statement that:"The only CF standard attribute which is allowed to refer > to external variables is cell_measures." > > I believe this is stronger than intended: it should be possible, for > instance, to refer to external variables in the comment and history > attributes, and any other free text attribute, or the comment part of the > cell_methods string. > > Should it say something along the lines of: > "The only attribute which is allowed to make a CF standard reference to > external variables is cell_measures"? > > regards, > Martin > > _______________________________________________ > CF-metadata mailing > [email protected]<mailto:[email protected]> > <[email protected]>http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > > > > -- > David Hassell > National Centre for Atmospheric Science > Department of Meteorology, University of Reading, Earley Gate, PO Box 243, > Reading RG6 6BB > Tel: +44 118 378 5613 <+44%20118%20378%205613>http://www.met.reading.ac.uk/ > _______________________________________________ > CF-metadata mailing > [email protected]http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > _______________________________________________ > CF-metadata mailing > [email protected]http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > > > > _______________________________________________ > CF-metadata mailing list > [email protected] > http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata > > -- V. Balaji Office: +1-609-452-6516 Head, Modeling Systems Group, GFDL Mobile: +1-917-273-9824 Princeton University Email: [email protected]
_______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
