Ed Hartnett wrote:
Jeff Whitaker <jswhit@xxxxxxxxxxx> writes:
Jeff Whitaker wrote:
Ed et al: Here's a simple C program that produces an incorrect
netcdf file with netcdf-4.1
Can anyone confirm or deny that this is a real bug? I noticed that
netcdf-4.1 was pulled from the ftp site, so I guess that means someone
is working on it ....
-Jeff
Howdy Jeff!
Yes, indeed, this is a real bug. I have taken down 4.1 as a result, and
we are preparing a release with a fix.
The bug itself manifests itself under the following circumstances: if
you define dimensions, then you define a coordinate variable, and write
to it, and then another coordinate variable, and if the coordinate
variables are in a different order than the dimensions, on re-reading
the file, netcdf-4.1 will be confused.
A key part of activating the bug is that you write some of the
coordinate variable data before defining all the coordinate
variables. That is, if you call all the nc_def_var calls for the
coordinate variables before writing any of their data, things will work
fine. You must also be writing the dimensions and the coordinate
variables in different orders to see this bug. If you defined the
coordinate variables in the same order as you defined the dimensions,
writing them before defining all of them would not be a problem.
The way to check for this problem is to do an ncdump -h of the file and
see if the dimension information in the ncdump is correct. In C it can
be detected by closing the file, re-opening it, and checking the
dimension ids of the variables.
And as for your reference to "working on it," well, aren't we always?
;-)
Thanks Ed - I had no doubts about that. I just wanted confirmation that
the problem wasn't just me doing something stupid.
I will announce when we have a new release. The data files produced with
4.1 will be readable after I get the fix in place - the problem is in
reading the file, it is written out just fine.
Unfortunately this happens at a time when I have scheduled a short trip
to see some family. However you can be sure this is my top programming
priority.
My apologies for any problems or inconvenience this may cause to any
users. With extensive automated testing we hope to ensure that each
release of netCDF is bug-free, but sometimes one slips through the
net...
If I might be so presumptious as to offer a suggestion for avoiding this
in the future (it also happened with the 4.0 release if I remember
correctly).
Let the release candidate become the actual release after some period of
the time with no bug reports (say 2 weeks?). A lot more people actually
exercise the release candidates and beta versions than the snapshots In
both 4.0 and 4.1 a bug was introduced in between the last release
candidate and the final release, and could only be detected if you were
running the snapshots. Real users will always do things that your
automated testing won't.
-Jeff
--
Jeffrey S. Whitaker Phone : (303)497-6313
Meteorologist FAX : (303)497-6449
NOAA/OAR/PSD R/PSD1 Email : Jeffrey.S.Whitaker@xxxxxxxx
325 Broadway Office : Skaggs Research Cntr 1D-113
Boulder, CO, USA 80303-3328 Web : http://tinyurl.com/5telg
