Re: 950523: netcdf build problem on DEC Alpha 2100

John,

I can't reproduce the problem you are seeing with the patches.  Using
version 2.1, which is the latest version of patch from the GNU archives that
I just built, everything succeeds, as the appended patch output shows.

It's hard to believe that this might be OS-dependent, but I ran the above on
a SPARCstation 10/Solaris 2.3 platform.

To get around this problem with patch, I've just made the patched and
compressed tar file available as

    ftp://ftp.unidata.ucar.edu/pub/netcdf/netcdf-232pl4.tar.Z

Please let me know if this works better.

--Russ




buddy% /usr/local/gnu/bin/patch --version
Patch version 2.1
buddy% /usr/local/gnu/bin/patch < ~ftp/pub/netcdf/2.3.2-patch3
Hmm...  Looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|This is a patch for the experimental C++ interface to netCDF in version
|2.3.2pl2 (April 1994).
|
|This patch accomplishes the following:
|
|BUG FIXES
|
|    A bug in netcdf.h that prevented correct compilation of the experimental
|    C++ interface is fixed by removing typedefs for ncchar, ncbyte, ncshort,
|    ncfloat, and ncdouble.
|
|    Fixed off-by-one problem with strings reported by Tom Lefebvre.
|
|    Minor changes to work with g++.
|
|    Added put and get functions for ncbyte arrays.
|
|    Fixed a bug in NcVar::get_att(NcToken) member, to make it return 0 for
|    attributes that don't exist.    
|
|    NcAtt::is_valid() had been returning FALSE for global attributes, but this
|    fix makes it return TRUE when appropriate.
|
|This patch-file is designed to be applied from the top-level netcdf source
|directory via Larry Wall's patch(1) utility, e.g.
|
|       $ cd /usr/local/src/netcdf-2.3.2
|       $ patch < this_file
|
|After applying this patch, you *must* remake the netCDF library from scratch
|(from the `configure' step), because the libsrc/netcdf.h file is regenerated
|by the `configure' script from one of the files changed by this patch.
|
|===================================================================
|diff -c1 -r libsrc/netcdf.h.in
|*** oldlibsrc/netcdf.h.in      Tue Jun  8 13:20:36 1993
|--- libsrc/netcdf.h.in Wed Jul 21 12:36:23 1993
--------------------------
Patching file libsrc/netcdf.h.in using Plan A...
Hunk #1 succeeded at 16.
Hunk #2 succeeded at 278.
Hunk #3 succeeded at 288.
Hmm...  The next patch looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|diff -c1 -r c++/ncvalues.cc
|*** oldc++/ncvalues.cc Thu Apr 15 15:41:03 1993
|--- c++/ncvalues.cc    Fri Aug  6 11:13:55 1993
--------------------------
Patching file c++/ncvalues.cc using Plan A...
Hunk #1 succeeded at 6.
Hunk #2 succeeded at 156.
Hunk #3 succeeded at 164.
Hmm...  The next patch looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|diff -c1 -r c++/netcdf.cc
|*** oldc++/netcdf.cc   Sun Apr 25 10:23:53 1993
|--- c++/netcdf.cc      Tue Feb 22 11:09:42 1994
--------------------------
Patching file c++/netcdf.cc using Plan A...
Hunk #1 succeeded at 6.
Hunk #2 succeeded at 208.
Hunk #3 succeeded at 493.
Hunk #4 succeeded at 566.
Hunk #5 succeeded at 587.
Hunk #6 succeeded at 627.
Hunk #7 succeeded at 648.
Hunk #8 succeeded at 680.
Hunk #9 succeeded at 858.
Hmm...  The next patch looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|diff -c1 -r c++/netcdf.hh
|*** oldc++/netcdf.hh   Sun Apr 25 10:23:11 1993
|--- c++/netcdf.hh      Tue Dec  7 09:17:48 1993
--------------------------
Patching file c++/netcdf.hh using Plan A...
Hunk #1 succeeded at 6.
Hunk #2 succeeded at 243.
Hunk #3 succeeded at 260.
Hunk #4 succeeded at 272.
Hunk #5 succeeded at 289.
done
buddy% /usr/local/gnu/bin/patch < ~ftp/pub/netcdf/2.3.2-patch4
Hmm...  Looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|This is a patch number 4 for netCDF version 2.3.2.  It updates the software
|to version 2.3.2pl4.  Before applying this patch, you should first apply
|patch 3, available from
|
|    ftp://ftp.unidata.ucar.edu/pub/netcdf/2.3.2-patch3
|
|Patches 3 and 4 only correct problems in the experimental netCDF C++
|interface; they are not needed if you are not using the C++ interface.
|
|Note that after applying patch3, you *must* remake the netCDF library from
|scratch (from the `configure' step), because the libsrc/netcdf.h file is
|regenerated by the `configure' script from one of the files changed by this
|patch.
|
|This patch accomplishes the following:
|
|BUG FIXES
|
|    Fixed bug reported by Tom Glaess (and others) with NcVar::set_cur() not
|    incrementing the record dimension properly.
|
|    Minor changes to make the software compile with g++.
|    
|    Made virtual destructor for NcFile protected instead of =0 since it's
|    not really pure.
|
|    Revised the Makefile to create a library, not just an object file.
|
|    Changed the type of the last argument in the interface of
|    NcFile::add_var(), so an array of dimensions could be used when a
|    variable has more than five dimensions.
|
|    Fixed a memory leak in the NcFile::sync() member function.
|
|    Changed NcVar::edges() to return a non-const, since it's sometimes
|    useful to be able to modify the returned array for later use.
|
|This patch-file is designed to be applied from the top-level netcdf source
|directory via Larry Wall's patch(1) utility.  So, after you have applied
|patch 3, for example, use something like the following:
|
|       $ cd /usr/local/src/netcdf-2.3.2
|       $ patch < this_file
|
|===================================================================
|diff -c1 -r oldc++/Makefile c++/Makefile
|*** oldc++/Makefile    Sun Apr 25 10:17:39 1993
|--- c++/Makefile       Tue Jan 31 16:23:38 1995
--------------------------
Patching file c++/Makefile using Plan A...
Hunk #1 succeeded at 1.
Hunk #2 succeeded at 5.
Hunk #3 succeeded at 15.
Hunk #4 succeeded at 56.
Hunk #5 succeeded at 60.
Hunk #6 succeeded at 69.
Hunk #7 succeeded at 73.
Hmm...  The next patch looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|diff -c1 -r oldc++/netcdf.cc c++/netcdf.cc
|*** oldc++/netcdf.cc   Tue Jan 31 14:08:35 1995
|--- c++/netcdf.cc      Tue Jan 31 12:48:59 1995
--------------------------
Patching file c++/netcdf.cc using Plan A...
Hunk #1 succeeded at 6.
Hunk #2 succeeded at 162.
Hunk #3 succeeded at 168.
Hunk #4 succeeded at 220.
Hunk #5 succeeded at 469.
Hunk #6 succeeded at 573.
Hunk #7 succeeded at 634.
Hunk #8 succeeded at 669.
Hunk #9 succeeded at 684.
Hmm...  The next patch looks like a new-style context diff to me...
The text leading up to this was:
--------------------------
|diff -c1 -r oldc++/netcdf.hh c++/netcdf.hh
|*** oldc++/netcdf.hh   Tue Jan 31 14:08:35 1995
|--- c++/netcdf.hh      Tue Jan 31 12:48:59 1995
--------------------------
Patching file c++/netcdf.hh using Plan A...
Hunk #1 succeeded at 6.
Hunk #2 succeeded at 12.
Hunk #3 succeeded at 35.
Hunk #4 succeeded at 62.
Hunk #5 succeeded at 97.
Hunk #6 succeeded at 234.
Hunk #7 succeeded at 337.
Hunk #8 succeeded at 347.
Hunk #9 succeeded at 378.
done