From steve at sumost.ca  Sat Aug 22 23:20:00 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Sat, 22 Aug 2015 22:20 -0500
Subject: [MINC-development] MINC with NetCDF 4.4?
Message-ID: <2204850.mNumVZCYYR@riemann>

Hi,

Debian just upgraded NetCDF to version 4.4.0, which broke the MINC build.  
Debian was using MINC 2.2.00, so I spent some time today updating MINC.

I can't find any release tarballs more recent than 2.2.00 (from 2012!).  So I 
presume everyone is working from github tree?  I updated to the tip of the 
'master' branch and discovered that if compiled with MINC1 support, some tests 
fail.  They fail also in develop:

The following tests FAILED:
         25 - test_xfm_3 (Failed)
         26 - copy_xfm (Failed)
         27 - verify_xfm_1 (Failed)
         28 - verify_xfm_float (Failed)
         29 - verify_xfm_short (Failed)
         30 - verify_xfm_byte (Failed)
         31 - verify_xfm_2 (Failed)


Anyone have experience with NetCDF 4.4?

Thanks,
-Steve
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150822/48ef8369/attachment.pgp>

From robert.d.vincent at mcgill.ca  Sun Aug 23 08:21:02 2015
From: robert.d.vincent at mcgill.ca (Robert D. Vincent)
Date: Sun, 23 Aug 2015 08:21:02 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <2204850.mNumVZCYYR@riemann>
References: <2204850.mNumVZCYYR@riemann>
Message-ID: <CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>

Hi Steve,

Yes, we are mostly building directly off the github tree these days. I have
not tried building with NetCDF > 4.X. As you probably know, it's surprising
to see these particular test failures, as the general transform code should
not contain many code paths that rely on NetCDF.

How did you add 4.4 to the build - i.e. what is the URL of the NetCDF
tarball you are using?

Which project are you building - libminc, minc-tools, or minc-toolkit (the
all-encompassing meta-project)?

    -bert

On Sat, Aug 22, 2015 at 11:20 PM, Steve M. Robbins <steve at sumost.ca> wrote:

> Hi,
>
> Debian just upgraded NetCDF to version 4.4.0, which broke the MINC build.
> Debian was using MINC 2.2.00, so I spent some time today updating MINC.
>
> I can't find any release tarballs more recent than 2.2.00 (from 2012!).
> So I
> presume everyone is working from github tree?  I updated to the tip of the
> 'master' branch and discovered that if compiled with MINC1 support, some
> tests
> fail.  They fail also in develop:
>
> The following tests FAILED:
>          25 - test_xfm_3 (Failed)
>          26 - copy_xfm (Failed)
>          27 - verify_xfm_1 (Failed)
>          28 - verify_xfm_float (Failed)
>          29 - verify_xfm_short (Failed)
>          30 - verify_xfm_byte (Failed)
>          31 - verify_xfm_2 (Failed)
>
>
> Anyone have experience with NetCDF 4.4?
>
> Thanks,
> -Steve
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150823/66e29468/attachment.html>

From vladimir.fonov at gmail.com  Sun Aug 23 09:43:27 2015
From: vladimir.fonov at gmail.com (Vladimir S. FONOV)
Date: Sun, 23 Aug 2015 09:43:27 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
References: <2204850.mNumVZCYYR@riemann>
	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
Message-ID: <AB5988E7-DFC6-4801-8DB3-2F60EAE87547@gmail.com>

Hello,

minc-toolkit is actually using NetCDF 4.3.x for  a while , and all the tests pass. See https://github.com/BIC-MNI/minc-toolkit/blob/master/cmake-modules/BuildNETCDF.cmake 
the libminc is used from develop branch of https://github.com/BIC-MNI/libminc though. 


On Aug 23, 2015, at 8:21 AM, Robert D. Vincent <robert.d.vincent at mcgill.ca> wrote:

> Hi Steve,
> 
> Yes, we are mostly building directly off the github tree these days. I have not tried building with NetCDF > 4.X. As you probably know, it's surprising to see these particular test failures, as the general transform code should not contain many code paths that rely on NetCDF.
> 
> How did you add 4.4 to the build - i.e. what is the URL of the NetCDF tarball you are using?
> 
> Which project are you building - libminc, minc-tools, or minc-toolkit (the all-encompassing meta-project)?
> 
>     -bert
> 
> On Sat, Aug 22, 2015 at 11:20 PM, Steve M. Robbins <steve at sumost.ca> wrote:
> Hi,
> 
> Debian just upgraded NetCDF to version 4.4.0, which broke the MINC build.
> Debian was using MINC 2.2.00, so I spent some time today updating MINC.
> 
> I can't find any release tarballs more recent than 2.2.00 (from 2012!).  So I
> presume everyone is working from github tree?  I updated to the tip of the
> 'master' branch and discovered that if compiled with MINC1 support, some tests
> fail.  They fail also in develop:
> 
> The following tests FAILED:
>          25 - test_xfm_3 (Failed)
>          26 - copy_xfm (Failed)
>          27 - verify_xfm_1 (Failed)
>          28 - verify_xfm_float (Failed)
>          29 - verify_xfm_short (Failed)
>          30 - verify_xfm_byte (Failed)
>          31 - verify_xfm_2 (Failed)
> 
> 
> Anyone have experience with NetCDF 4.4?
> 
> Thanks,
> -Steve
> 
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
> 
> 
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development

Best regards, Vladimir S. FONOV ~ v.s.fonov <at> ilmarin.info




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150823/4bd13483/attachment.html>

From steve at sumost.ca  Sun Aug 23 12:43:03 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Sun, 23 Aug 2015 11:43:03 -0500
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
References: <2204850.mNumVZCYYR@riemann>
	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
Message-ID: <2118458.HzWj7pAcqV@riemann>

Hi,

On August 23, 2015 08:21:02 AM Robert D. Vincent wrote:

> Yes, we are mostly building directly off the github tree these days. I have
> not tried building with NetCDF > 4.X. As you probably know, it's surprising
> to see these particular test failures, as the general transform code should
> not contain many code paths that rely on NetCDF.

Yes, but note that tests only fail if I enable MINC1 support (which is off by 
default).  Without MINC1 there are no failures.

 
> How did you add 4.4 to the build - i.e. what is the URL of the NetCDF
> tarball you are using?

I am using Debian "unstable" which contains 4.4 by default as of a few days 
ago.  Actually, I see the version tag is "4.4.0~rc2", so probably this:

http://www.unidata.ucar.edu/mailing_lists/archives/netcdfgroup/2015/msg00189.html

 
> Which project are you building - libminc, minc-tools, or minc-toolkit (the
> all-encompassing meta-project)?

I am presently building libminc.  Both master and develop branches show test 
failures.

Previously, I was building the old minc 2.2.00 (which I believe is the same as 
the current libminc + minc-tools ?) using netcdf 4.3.  I configured that with 
both MINC1 and MINC2 support and there were no test failures.  When netcdf 4.4 
showed up, there were test failures with this source base -- which is what 
prompted me to look at updating the minc sources.  

GIven all that, it really smells like netcdf 4.4 has changed something that 
minc1 fundamentally relies on.  I haven't had a chance to actually look at 
code, though.

Thanks,
-Steve

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150823/ed3f8f17/attachment.pgp>

From steve at sumost.ca  Sun Aug 23 12:52:11 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Sun, 23 Aug 2015 11:52:11 -0500
Subject: [MINC-development] Is MINC1 still used?
In-Reply-To: <2118458.HzWj7pAcqV@riemann>
References: <2204850.mNumVZCYYR@riemann>
	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
	<2118458.HzWj7pAcqV@riemann>
Message-ID: <15494131.WOCnsFKvfZ@riemann>

On August 23, 2015 11:43:03 AM Steve M. Robbins wrote:

> [...] note that tests only fail if I enable MINC1 support (which is off
> by default).  Without MINC1 there are no failures.

So this prompts the question: should I care?  Would anyone be inconvenienced 
if Debian's MINC only supported MINC2?

I don't personally use MINC at all these days (though if I did go back to look 
at my old datasets they will all be MINC1, so I would miss MINC1 support if it 
were lacking).

For the rest of the world, though: would they miss it?  Are the datasets -- 
like the standard BIC templates, or atlases -- distributed in MINC1 format?

Thanks,
-Steve
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150823/aeae64f4/attachment.pgp>

From steve at sumost.ca  Sun Aug 23 13:04:08 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Sun, 23 Aug 2015 12:04:08 -0500
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <2118458.HzWj7pAcqV@riemann>
References: <2204850.mNumVZCYYR@riemann>
	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
	<2118458.HzWj7pAcqV@riemann>
Message-ID: <2639690.CNgykcgFWQ@riemann>

Hi,

Last email for now -- I promise :-)

A correction on what I previously wrote:

On August 23, 2015 11:43:03 AM Steve M. Robbins wrote:

> Previously, I was building the old minc 2.2.00 (which I believe is the same
> as the current libminc + minc-tools ?) using netcdf 4.3. 

I was actually building with netcdf 4.1.3 (not 4.3).

-Steve


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150823/815d4b2f/attachment.pgp>

From vladimir.fonov at gmail.com  Sun Aug 23 19:23:19 2015
From: vladimir.fonov at gmail.com (Vladimir S. Fonov)
Date: Sun, 23 Aug 2015 19:23:19 -0400
Subject: [MINC-development] Is MINC1 still used?
In-Reply-To: <15494131.WOCnsFKvfZ@riemann>
References: <2204850.mNumVZCYYR@riemann>	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>	<2118458.HzWj7pAcqV@riemann>
	<15494131.WOCnsFKvfZ@riemann>
Message-ID: <55DA55E7.7090706@gmail.com>

Hello,

On 15-08-23 12:52 PM, Steve M. Robbins wrote:
> On August 23, 2015 11:43:03 AM Steve M. Robbins wrote:
>
>> [...] note that tests only fail if I enable MINC1 support (which is off
>> by default).  Without MINC1 there are no failures.
>
> So this prompts the question: should I care?  Would anyone be inconvenienced
> if Debian's MINC only supported MINC2?

Unfortunately you will not be able to compile most of minc tools without 
MINC1 support, since they use MINC1 api which is strongly NETCDF dependent.




> I don't personally use MINC at all these days (though if I did go back to look
> at my old datasets they will all be MINC1, so I would miss MINC1 support if it
> were lacking).
>
> For the rest of the world, though: would they miss it?  Are the datasets --
> like the standard BIC templates, or atlases -- distributed in MINC1 format?
>
> Thanks,
> -Steve
>
>
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>


From robert.d.vincent at mcgill.ca  Mon Aug 24 09:49:06 2015
From: robert.d.vincent at mcgill.ca (Robert D. Vincent)
Date: Mon, 24 Aug 2015 09:49:06 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <2639690.CNgykcgFWQ@riemann>
References: <2204850.mNumVZCYYR@riemann>
	<CAAd5DeT36ZR6Wr7q6jDhzYOvjnHDgarO=wRomFzpv6b5kjujyQ@mail.gmail.com>
	<2118458.HzWj7pAcqV@riemann> <2639690.CNgykcgFWQ@riemann>
Message-ID: <CAAd5DeRTSAR4b_GsG1jBBJadw-7_8Dr6tKO3EinXUdn7=Emqsw@mail.gmail.com>

Hi all,

The issue with netCDF 4.4 seems to be that netCDF will now open HDF5 files
successfully, but the rest of the file access will fail. Some of our code
assumes that netCDF calls will fail to open HDF5 files, this allowed us to
easily detect MINC 1 vs. MINC 2 files and use the correct library for each.
I will create a fix that should work around this problem.

    -bert


On Sun, Aug 23, 2015 at 1:04 PM, Steve M. Robbins <steve at sumost.ca> wrote:

> Hi,
>
> Last email for now -- I promise :-)
>
> A correction on what I previously wrote:
>
> On August 23, 2015 11:43:03 AM Steve M. Robbins wrote:
>
> > Previously, I was building the old minc 2.2.00 (which I believe is the
> same
> > as the current libminc + minc-tools ?) using netcdf 4.3.
>
> I was actually building with netcdf 4.1.3 (not 4.3).
>
> -Steve
>
>
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/1bb9d80a/attachment.html>

From sean at rogue-research.com  Mon Aug 24 12:50:06 2015
From: sean at rogue-research.com (Sean McBride)
Date: Mon, 24 Aug 2015 12:50:06 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <2204850.mNumVZCYYR@riemann>
References: <2204850.mNumVZCYYR@riemann>
Message-ID: <20150824165006.894877609@mail.rogue-research.com>

On Sat, 22 Aug 2015 22:20:00 -0500, Steve M. Robbins said:

>I can't find any release tarballs more recent than 2.2.00 (from 2012!).

Forgive this little rant, but this really reflects badly on the MINC project.  A new release is long overdue.

2.2 doesn't even *compile* on current versions of OS X with clang (the default compiler).

>So I 
>presume everyone is working from github tree?  I updated to the tip of the 
>'master' branch and discovered that if compiled with MINC1 support, some
>tests 
>fail.  They fail also in develop:

I didn't try NetCDF 4.4, but I have several nightly builds & tests that use 4.3.3.1.

I setup a dashboard here:

<http://my.cdash.org/index.php?project=libminc>

I encourage anyone that can to also setup a nightly build so that compilation and tests can be run nightly.

Cheers,

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montr?al, Qu?bec, Canada



From vladimir.fonov at gmail.com  Mon Aug 24 13:02:29 2015
From: vladimir.fonov at gmail.com (Vladimir S. FONOV)
Date: Mon, 24 Aug 2015 13:02:29 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <20150824165006.894877609@mail.rogue-research.com>
References: <2204850.mNumVZCYYR@riemann>
	<20150824165006.894877609@mail.rogue-research.com>
Message-ID: <CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>

Well,

if everybody agrees we can merge develop branch into master and make
official 2.3.00 release (and put it on packages).  There is a problem with
backward compatibility with autoconf/automake though.

On Mon, Aug 24, 2015 at 12:50 PM, Sean McBride <sean at rogue-research.com>
wrote:

> On Sat, 22 Aug 2015 22:20:00 -0500, Steve M. Robbins said:
>
> >I can't find any release tarballs more recent than 2.2.00 (from 2012!).
>
> Forgive this little rant, but this really reflects badly on the MINC
> project.  A new release is long overdue.
>
> 2.2 doesn't even *compile* on current versions of OS X with clang (the
> default compiler).
>
> >So I
> >presume everyone is working from github tree?  I updated to the tip of the
> >'master' branch and discovered that if compiled with MINC1 support, some
> >tests
> >fail.  They fail also in develop:
>
> I didn't try NetCDF 4.4, but I have several nightly builds & tests that
> use 4.3.3.1.
>
> I setup a dashboard here:
>
> <http://my.cdash.org/index.php?project=libminc>
>
> I encourage anyone that can to also setup a nightly build so that
> compilation and tests can be run nightly.
>
> Cheers,
>
> --
> ____________________________________________________________
> Sean McBride, B. Eng                 sean at rogue-research.com
> Rogue Research                        www.rogue-research.com
> Mac Software Developer              Montr?al, Qu?bec, Canada
>
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>



-- 
Best regards,

 Vladimir S. Fonov ~ vladimir <dot> fonov <at> gmail <dot> com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/e22d6cb1/attachment.html>

From robert.d.vincent at mcgill.ca  Mon Aug 24 19:18:09 2015
From: robert.d.vincent at mcgill.ca (Robert D. Vincent)
Date: Mon, 24 Aug 2015 19:18:09 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>
References: <2204850.mNumVZCYYR@riemann>
	<20150824165006.894877609@mail.rogue-research.com>
	<CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>
Message-ID: <CAAd5DeQrqeXFE-320GYtfe1a1DZSqbtJtfG9NjYzTYTB2OKVCA@mail.gmail.com>

Hi all,

I've fixed the issue with netCDF 4.4 and checked the fix into the develop
branch.

I agree that we should do an official release of the minc libraries and
tools. I would like to either fix the autoconf/automake build, or remove
that capability entirely. Votes?

    -bert

On Mon, Aug 24, 2015 at 1:02 PM, Vladimir S. FONOV <vladimir.fonov at gmail.com
> wrote:

> Well,
>
> if everybody agrees we can merge develop branch into master and make
> official 2.3.00 release (and put it on packages).  There is a problem with
> backward compatibility with autoconf/automake though.
>
> On Mon, Aug 24, 2015 at 12:50 PM, Sean McBride <sean at rogue-research.com>
> wrote:
>
>> On Sat, 22 Aug 2015 22:20:00 -0500, Steve M. Robbins said:
>>
>> >I can't find any release tarballs more recent than 2.2.00 (from 2012!).
>>
>> Forgive this little rant, but this really reflects badly on the MINC
>> project.  A new release is long overdue.
>>
>> 2.2 doesn't even *compile* on current versions of OS X with clang (the
>> default compiler).
>>
>> >So I
>> >presume everyone is working from github tree?  I updated to the tip of
>> the
>> >'master' branch and discovered that if compiled with MINC1 support, some
>> >tests
>> >fail.  They fail also in develop:
>>
>> I didn't try NetCDF 4.4, but I have several nightly builds & tests that
>> use 4.3.3.1.
>>
>> I setup a dashboard here:
>>
>> <http://my.cdash.org/index.php?project=libminc>
>>
>> I encourage anyone that can to also setup a nightly build so that
>> compilation and tests can be run nightly.
>>
>> Cheers,
>>
>> --
>> ____________________________________________________________
>> Sean McBride, B. Eng                 sean at rogue-research.com
>> Rogue Research                        www.rogue-research.com
>> Mac Software Developer              Montr?al, Qu?bec, Canada
>>
>>
>> _______________________________________________
>> MINC-development mailing list
>> MINC-development at bic.mni.mcgill.ca
>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>>
>
>
>
> --
> Best regards,
>
>  Vladimir S. Fonov ~ vladimir <dot> fonov <at> gmail <dot> com
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/368f41b1/attachment.html>

From steve at sumost.ca  Mon Aug 24 19:35:17 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Mon, 24 Aug 2015 18:35:17 -0500
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <CAAd5DeQrqeXFE-320GYtfe1a1DZSqbtJtfG9NjYzTYTB2OKVCA@mail.gmail.com>
References: <2204850.mNumVZCYYR@riemann>
	<CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>
	<CAAd5DeQrqeXFE-320GYtfe1a1DZSqbtJtfG9NjYzTYTB2OKVCA@mail.gmail.com>
Message-ID: <12867754.8JI4MIoxtx@riemann>

On August 24, 2015 07:18:09 PM Robert D. Vincent wrote:
> Hi all,
> 
> I've fixed the issue with netCDF 4.4 and checked the fix into the develop
> branch.

Super!  Thanks so much Bert!!


> I agree that we should do an official release of the minc libraries and
> tools. I would like to either fix the autoconf/automake build, or remove
> that capability entirely. 

The ChangeLog -- even on master -- led me to believe it was already removed 
(2013-11-28).

> Votes?

My vote: make life easier for yourselves; support just one build system.

-Steve
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/d2d306ba/attachment.pgp>

From a.janke at gmail.com  Mon Aug 24 22:03:05 2015
From: a.janke at gmail.com (Andrew Janke)
Date: Tue, 25 Aug 2015 12:03:05 +1000
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <12867754.8JI4MIoxtx@riemann>
References: <2204850.mNumVZCYYR@riemann>
	<CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>
	<CAAd5DeQrqeXFE-320GYtfe1a1DZSqbtJtfG9NjYzTYTB2OKVCA@mail.gmail.com>
	<12867754.8JI4MIoxtx@riemann>
Message-ID: <CAG0CB20+1YQFY78=0sRfBA4JLLjgYsXkWc96=MenkNcw6u7vzA@mail.gmail.com>

On 25 August 2015 at 09:35, Steve M. Robbins <steve at sumost.ca> wrote:
>
> My vote: make life easier for yourselves; support just one build system.

+1.  I thought we were long done with automake/conf for the core of
minc. Not so with some of the extra tools in minc-toolkit but
certainly for libminc + minc-tools


a

From steve at sumost.ca  Tue Aug 25 00:42:57 2015
From: steve at sumost.ca (Steve M. Robbins)
Date: Mon, 24 Aug 2015 23:42:57 -0500
Subject: [MINC-development] fix SOVERSION generation
Message-ID: <1864792.8IK23feRCq@riemann>

Hi,

Attached patch to 'develop' fixes SOVERSION on the two generated libraries. 

THanks,
-Steve

-------------- next part --------------
A non-text attachment was scrubbed...
Name: diff
Type: text/x-patch
Size: 815 bytes
Desc: not available
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/847773c4/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: This is a digitally signed message part.
URL: <http://www.bic.mni.mcgill.ca/pipermail/minc-development/attachments/20150824/847773c4/attachment.pgp>

From a.janke at gmail.com  Tue Aug 25 00:51:37 2015
From: a.janke at gmail.com (Andrew Janke)
Date: Tue, 25 Aug 2015 14:51:37 +1000
Subject: [MINC-development] fix SOVERSION generation
In-Reply-To: <1864792.8IK23feRCq@riemann>
References: <1864792.8IK23feRCq@riemann>
Message-ID: <CAG0CB23NjUmrXNDy0J3o-su6jnusd3aeu_qaEBdO+DCO0Cokkw@mail.gmail.com>

On this note I don't think we follow convention for the libminc
SOVERSION as per what used to happen with the automake for Major+Minor
version numbers.

   https://github.com/BIC-MNI/libminc/blob/master/README.release

Certainly the instructions are not up to date as I still see things
like "make distcheck"

I take it this is still important for a Debian/Ubuntu build. Vlad +
Bert: I'll be in Montreal on the 7th + 8th of September so am happy to
go through what we used to do for a release to make sure that it
worked for Steve.


a

On 25 August 2015 at 14:42, Steve M. Robbins <steve at sumost.ca> wrote:
> Hi,
>
> Attached patch to 'develop' fixes SOVERSION on the two generated libraries.
>
> THanks,
> -Steve
>
>
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development
>

From sean at rogue-research.com  Tue Aug 25 10:19:04 2015
From: sean at rogue-research.com (Sean McBride)
Date: Tue, 25 Aug 2015 10:19:04 -0400
Subject: [MINC-development] MINC with NetCDF 4.4?
In-Reply-To: <CAG0CB20+1YQFY78=0sRfBA4JLLjgYsXkWc96=MenkNcw6u7vzA@mail.gmail.com>
References: <2204850.mNumVZCYYR@riemann>
	<CAMCwi7NcYS9vP=GSUXKeC6YhVRU3NCqCuutM8AUbveddsdNBGg@mail.gmail.com>
	<CAAd5DeQrqeXFE-320GYtfe1a1DZSqbtJtfG9NjYzTYTB2OKVCA@mail.gmail.com>
	<12867754.8JI4MIoxtx@riemann>
	<CAG0CB20+1YQFY78=0sRfBA4JLLjgYsXkWc96=MenkNcw6u7vzA@mail.gmail.com>
Message-ID: <20150825141904.1570702632@mail.rogue-research.com>

On Tue, 25 Aug 2015 12:03:05 +1000, Andrew Janke said:

>On 25 August 2015 at 09:35, Steve M. Robbins <steve at sumost.ca> wrote:
>>
>> My vote: make life easier for yourselves; support just one build system.
>
>+1.  I thought we were long done with automake/conf for the core of

+1

Cheers,

-- 
____________________________________________________________
Sean McBride, B. Eng                 sean at rogue-research.com
Rogue Research                        www.rogue-research.com 
Mac Software Developer              Montr?al, Qu?bec, Canada