Hi Rob,

Does moving the timezone struct in H5win32defs.h into the Visual Studio block 
just above it solve the problem?

In the very near future I hope that the struct will go away entirely. That 
feature has never worked properly across POSIX. It's probably too late for me 
to get that change into 1.10.0, though, so it'll have to wait another revision.

Cheers,

Dana

-----Original Message-----
From: Hdf-forum [mailto:[email protected]] On Behalf Of Dana 
Robinson
Sent: Sunday, March 27, 2016 4:36 AM
To: HDF Users Discussion List <[email protected]>
Subject: Re: [Hdf-forum] Release of HDF5-1.10.0-pre2

Hi Rob,

- I'll investigate and possibly enter a bug report for make install working 
after make check fails.

- I've updated the h5watch files in the revise_chunks branch.

- I'm not sure why err_compat is failing. I'd have to take a closer look into 
the output. I wonder if it's due to CR/LF differences? I'll see if I can get a 
basic MinGW system set up here to investigate, though that might not be 
possible before our code freeze on Monday. That issue might have to wait for 
1.10.1.

- Still working on that timezone struct problem.

Cheers,

Dana

-----Original Message-----
From: Hdf-forum [mailto:[email protected]] On Behalf Of 
[email protected]
Sent: Saturday, March 26, 2016 2:41 AM
To: HDF Users Discussion List <[email protected]>
Subject: Re: [Hdf-forum] Release of HDF5-1.10.0-pre2

-----Original Message-----
From: Dana Robinson
Sent: Saturday, March 26, 2016 12:33 PM
To: HDF Users Discussion List
Subject: Re: [Hdf-forum] Release of HDF5-1.10.0-pre2

> Can you copy the pertinent files from there into your 1.10.0 
> distribution and let me know if they build ok and don't interfere with the 
> tests?
> test_usecases.sh should pass and the output should indicate that it 
> was skipped.
>
> The changed files are:
>
>   M /branches/revise_chunks/test/atomic_reader.c
>   M /branches/revise_chunks/test/atomic_writer.c
>   M /branches/revise_chunks/test/test_usecases.sh.in
>   M /branches/revise_chunks/test/twriteorder.c
>   M /branches/revise_chunks/test/use.h
>   M /branches/revise_chunks/test/use_append_chunk.c
>   M /branches/revise_chunks/test/use_append_mchunks.c
>   M /branches/revise_chunks/test/use_common.c
>   M /branches/revise_chunks/test/use_disable_mdc_flushes.c

Those changes look pretty good, Dana.

Running 'make' now gets as far as:

###########################################
$ make
Making all in src
make[1]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/src'
make  all-am
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/src'
make[2]: Nothing to be done for `all-am'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/src'
make[1]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/src'
Making all in test
make[1]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/test'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/test'
Making all in tools
make[1]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools'
Making all in lib
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/lib'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/lib'
Making all in h5diff
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5diff'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5diff'
Making all in h5ls
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5ls'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5ls'
Making all in h5dump
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5dump'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5dump'
Making all in misc
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc'
Making all in vds
make[3]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc/vds'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc/vds'
make[3]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc'
make[3]: Nothing to be done for `all-am'.
make[3]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc'
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/misc'
Making all in h5import
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5import'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5import'
Making all in h5repack
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5repack'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5repack'
Making all in h5jam
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5jam'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5jam'
Making all in h5copy
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5copy'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5copy'
Making all in h5stat
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5stat'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/h5stat'
Making all in h5format_convert
make[2]: Entering directory
`/c/_32/comp/hdf5-1.10.0-pre2/tools/h5format_convert
'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory
`/c/_32/comp/hdf5-1.10.0-pre2/tools/h5format_convert'

Making all in perform
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/perform'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools/perform'
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/tools'
make[2]: Nothing to be done for `all-am'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools'
make[1]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/tools'
Making all in .
make[1]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2'
make[1]: Nothing to be done for `all-am'.
make[1]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2'
Making all in hl
make[1]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl'
Making all in src
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/src'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/src'
Making all in test
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/test'
make[2]: Nothing to be done for `all'.
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/test'
Making all in tools
make[2]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools'
Making all in gif2h5
make[3]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools/gif2h5'
make[3]: Nothing to be done for `all'.
make[3]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools/gif2h5'
Making all in h5watch
make[3]: Entering directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools/h5watch'
  CC       h5watchgentest.o
h5watchgentest.c:22:26: fatal error: sys/resource.h: No such file or directory 
#include <sys/resource.h>
                          ^
compilation terminated.
make[3]: *** [h5watchgentest.o] Error 1
make[3]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools/h5watch'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl/tools'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/c/_32/comp/hdf5-1.10.0-pre2/hl'
make: *** [all-recursive] Error 1
###########################################


At that stage I ran 'make check', and that ran fine (for quite a while)
until:

###########################################
.....
Finished testing swmr.exe
============================
============================
Testing testerror.sh
============================
testerror.sh  Test Log
============================
Testing err_compat
*FAILED*
    Expected result differs from actual result
    *** ./testfiles/err_compat_1        Wed Mar 23 06:01:06 2016
    --- ./err_compat.out        Sat Mar 26 16:11:28 2016
    ***************
    *** 1,4 ****
    ! Testing error API H5Eset/get_auto
Testing error API based on data I/O                                   All 
error API tests passed.
         This program tests the Error API compatible with HDF5 version (number).
  There're supposed to be some error messages
      ********* Print error stack in HDF5 default way *********
      HDF5-DIAG: Error detected in HDF5 (version (number)) thread (IDs):
    --- 1,4 ----
    ! Testing error API H5Eset/get_auto Testing error API based on data I/O
                                   All error API tests passed.
         This program tests the Error API compatible with HDF5 version (number).
  There're supposed to be some error messages
      ********* Print error stack in HDF5 default way *********
      HDF5-DIAG: Error detected in HDF5 (version (number)) thread (IDs):
Testing error_test
PASSED
make[4]: *** [testerror.sh.chkexe_] Error 1

###########################################

I'm not *too* worried about failing tests. My interest in this library is that 
it works with the PDL::IO::HDF5 (perl) module on Windows  - which has been the 
case to date (despite some failing tests in the past).

However, I do think it would be good to get 'make' to actually run to 
completion.
I was surprised to find that I can actually run 'make check' and 'make install' 
despite 'make' having terminated in error. That's a rather unusual behaviour in 
my experience.

Cheers,
Rob 


_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org
Twitter: https://twitter.com/hdf5

Reply via email to