[netCDF #YSI-369373]: Problems with 2D-Coords with netCDF4

> > OK, that bug was a lot more trouble than it looked like at first sight!
> >
> > But I believe I have gotten it at last. If you could try the snapshot that
will be produced tonight, than you can see if it works for you. I am adding
more tests, but I already added the one you sent me, and it works now.
> >
> > Thanks,
> >
> > Ed
> >
> > Ticket Details
> > ===================
> > Ticket ID: YSI-369373
> > Department: Support netCDF
> > Priority: Critical
> > Status: Closed
>
> Howdy Ed,
>
>
> i just tested the snapshop netcdf-4.0.1-beta1-snapshot2008121402 and i
> want to tell you about the results. In short form: It fails. :-(
>
> In long form:
> 1. "make checks" fails
> ----------------------
> I din't debug that in detail (IMHO this bug also appears in the last
> version), but the output says:
>
> Testing: ref_nctst_netcdf4_classic
> SUCCEED: ref_nctst_netcdf4_classic
> Testing: ref_const_test
> 35c35
> <  uint64v = 1863462912, 18446744073369944064, 14820165136566439339 ;
> ---
> >  uint64v = 18446744071562067968, 2147483647, 14820165136566439339 ;
> FAIL: ref_const_test
> 35c35
> <  uint64v = 1863462912, 18446744073369944064, 14820165136566439339 ;
> ---
> >  uint64v = 18446744071562067968, 2147483647, 14820165136566439339 ;
> FAIL: runtests.sh
> =========================================
> 1 of 1 tests failed
> Please report to address@hidden
> =========================================


Ah ha!

I do not see this error anywhere. That is very interesting. You say it has
appeared in the last two snapshots for you? Let me put this aside for the
moment and focus on the 2D coordinates bug.

>
> 2. My Test-Program fails again. :-(
> -----------------------------------
>
> I'm not sure if you still have it, so i will append it again to this
> e-mail. The problem is the same as in the last version. Perhaps i
> haven't your fixed version?
>

I note that the netcdf version number in your error message shows that you are
using the snapshot from last Sunday, but I checked in the fix for the
coordinate variables bug Monday night (during a long but productive delay at
the Denver airport on the way to the AGU in San Fran. ;-)

So can you try the current snapshot instead?
ftp://ftp.unidata.ucar.edu/pub/netcdf/snapshot/netcdf-4-daily.tar.gz

I have incorporated your test into libsrc4/tst_vars3.c (the second test). I
just took your code and changed the error handling, and then wrote some extra
code to check results. Everything seems to work fine. When I look at the
resulting file with ncdump, I see:

../ncdump/ncdump tst_vars3.nc
netcdf tst_vars3 {
dimensions:
        time = UNLIMITED ; // (1 currently)
        tl = 15 ;
variables:
        char time(time, tl) ;
data:

 time =
  "20051224.150000" ;
}

In fact I have just added some more tests to tst_vars3 and everything continues
to look good to me!

Finally, I reran your program, and run ncdump on the output to see:
netcdf test4 {
dimensions:
        time = UNLIMITED ; // (1 currently)
        tl = 15 ;
variables:
        char time(time, tl) ;
data:

 time =
  "20051224.150000" ;
}

So please try again, and let me know what happens.

Thanks,

Ed


Ticket Details
===================
Ticket ID: YSI-369373
Department: Support netCDF
Priority: Critical
Status: Open