Remove hard-wired use of NC_MAX_DIMS

[gdsjaar]

Replace static array dimensioned to NC_MAX_DIMS with dynamic memory allocation of correctly sized array.

[DennisHeimbigner]

I was in the process of making the same change, but this is equivalent, so we can
use it.

[gdsjaar]

Thanks, hope I didn't waste too much of your time. I saw this while looking at the NX_MAX_DIMS issue and decided that it was probably a non-controversial change and might buy me some karma to use later ;-)
..Greg

"A supercomputer is a device for turning compute-bound problems into I/O-bound problems"

From: DennisHeimbigner <[email protected]mailto:[email protected]>
Reply-To: Unidata/netcdf-c <[email protected]mailto:[email protected]>
Date: Wednesday, November 11, 2015 at 2:08 PM
To: Unidata/netcdf-c <[email protected]mailto:[email protected]>
Cc: "Sjaardema, Gregory D" <[email protected]mailto:[email protected]>
Subject: [EXTERNAL] Re: [netcdf-c] Remove hard-wired use of NC_MAX_DIMS (#150)

I was in the process of making the same change, but this is equivalent, so we can
use it.

Reply to this email directly or view it on GitHubhttps://github.com//pull/150#issuecomment-155881056.

[WardF]

Looks good so far; running regression tests via Docker unidata/nctest images to rule out any unforeseen impacts on the fortran, C++ or python interfaces.