Re: [ITP] Macaulay2 1.1

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

Dear Cygwin,

I submitted an ITP for Macaulay 2 on Nov 11 and there have been no positive
votes since then.  I would have thought that by now, if there were any interest
in having it in cygwin, there would have been a positive vote.  I understand
that 5 are required.

As I said before, Macaulay 2 is a software system for algebraic geometry
research.  Based on Groebner bases, it provides algorithms for computing
homological invariants of rings and modules.  Home page:
http://www.math.uiuc.edu/Macaulay2/".  In the past it has been included in
Fedora 8, 9, 10, see
https://admin.fedoraproject.org/pkgdb/packages/name/Macaulay2/.

Volker Zell tried compiling our older version and it didn't compile for him,
but he had installed a nonstandard fortran compiler on his system, so it may
not be relevant.

There was some discussion about whether to include the source code for
un-cygwin-packaged 3rd party libraries we use in the proposed source code
package of Macaulay 2.  I'm not sure the work to do so is worth it, but if that
is the only objection, I would have hoped to see some conditional "yes" votes
with such packaging as the condition.

Let me know *soon*, as we hope to release 1.2 by the end of the month, and if
it's not approved for cygwin, I'll just provide it, as I did with 1.1, on my
own "setup" compatible repository.

--
              Daniel R. Grayson, Professor (retired, Emeritus)
              Department of Mathematics
              University of Illinois at Urbana-Champaign
        www:  http://www.math.uiuc.edu/~dan/
      email:  dan@math.uiuc.edu
    us mail:  2409 S. Vine St., Urbana, IL 61801, USA
      phone:  +1-217-367-6384 home (88.20224W, 40.08541N)
 cell phone:  +1-217-377-0458

Re: [ITP] Macaulay2 1.1

[Christopher Faylor]

On Wed, Jan 07, 2009 at 01:04:39PM -0600, Daniel R. Grayson wrote:
>I submitted an ITP for Macaulay 2 on Nov 11 and there have been no
>positive votes since then.  I would have thought that by now, if there
>were any interest in having it in cygwin, there would have been a
>positive vote.  I understand that 5 are required.

I don't really grok this paragraph.  The point of voting is to see if
there is interest.  If no one voted then, yes, that would indicate
that there was no interest.

>As I said before, Macaulay 2 is a software system for algebraic
>geometry research.  Based on Groebner bases, it provides algorithms for
>computing homological invariants of rings and modules.  Home page:
>http://www.math.uiuc.edu/Macaulay2/".  In the past it has been included
>in Fedora 8, 9, 10, see
>https://admin.fedoraproject.org/pkgdb/packages/name/Macaulay2/.

I don't see it in Fedora 10 but it looks like it was included in the
"extras" package for older Fedoras.

>Let me know *soon*, as we hope to release 1.2 by the end of the month,
>and if it's not approved for cygwin, I'll just provide it, as I did
>with 1.1, on my own "setup" compatible repository.

It's a pretty simple process.  You need the votes, then, if you get the
votes, you need a "GTG" on the packaging.  It sounds like you don't have
either but the votes are the most crucial.

cgf

Re: [ITP] Macaulay2 1.1

[Christopher Faylor]

On Wed, Jan 07, 2009 at 02:14:26PM -0500, Christopher Faylor wrote:
>I don't see it in Fedora 10 but it looks like it was included in the
>"extras" package for older Fedoras.

...But this does not automatically qualify Macaulay2 for inclusion in
the Cygwin distribution.

cgf

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

> Date: Wed, 7 Jan 2009 14:14:26 -0500
> From: Christopher Faylor <cgf-use-the-mailinglist-please@cygwin.com>
> To: cygwin-apps@cygwin.com
> Subject: Re: [ITP] Macaulay2 1.1
> Reply-To: cygwin-apps@cygwin.com
> 
> On Wed, Jan 07, 2009 at 01:04:39PM -0600, Daniel R. Grayson wrote:
> >I submitted an ITP for Macaulay 2 on Nov 11 and there have been no
> >positive votes since then.  I would have thought that by now, if there
> >were any interest in having it in cygwin, there would have been a
> >positive vote.  I understand that 5 are required.
> 
> I don't really grok this paragraph.  The point of voting is to see if
> there is interest.  If no one voted then, yes, that would indicate
> that there was no interest.

Sorry, perhaps I should have mentioned that the number of replies to my initial
message seemed to indicate some interest.

By the way, Macaulay 2 is comparable to Singular, which is included in Cygwin.

Re: [ITP] Macaulay2 1.1

[Yaakov (Cygwin/X)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Daniel R. Grayson wrote:
> I submitted an ITP for Macaulay 2 on Nov 11 and there have been no positive
> votes since then.  I would have thought that by now, if there were any interest
> in having it in cygwin, there would have been a positive vote.  I understand
> that 5 are required.

On 14 November, Corinna explained[1] that you (or someone else) will
need to ITP the yet-unpackaged dependencies as well before this can be
seriously considered.  AFAICS that hasn't happened yet, so I think the
question of interest is still moot.

[1] http://cygwin.com/ml/cygwin-apps/2008-11/msg00100.html


Yaakov
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEAREIAAYFAkllMzEACgkQpiWmPGlmQSMHjQCeO7PiT+hAq+tyrpbklIzvd1Pw
4CgAnjO+lio1ZvoUytywwap+liRydHhc
=oKLX
-----END PGP SIGNATURE-----

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

You must be reading her message differently from the way I did.

In any case, if it were to turn out that Macaulay 2 were interesting to cygwin,
but one of the libraries Macaulay 2 depends on was not interesting to cygwin,
then I wouldn't have to package that library, would I?  I could just include
the source for it in the Macaulay 2 package.  Otherwise it's a catch-22.

So I think the question of whether ITP's for the dependent packages would
succeed on their own merits is moot.

> Date: Wed, 07 Jan 2009 16:56:49 -0600
> From: "Yaakov (Cygwin/X)" <yselkowitz@users.sourceforge.net>
> To: cygwin-apps@cygwin.com
> Subject: Re: [ITP] Macaulay2 1.1
  ...
> 
> On 14 November, Corinna explained[1] that you (or someone else) will
> need to ITP the yet-unpackaged dependencies as well before this can be
> seriously considered.  AFAICS that hasn't happened yet, so I think the
> question of interest is still moot.
> 
> [1] http://cygwin.com/ml/cygwin-apps/2008-11/msg00100.html

Re: [ITP] Macaulay2 1.1

[Corinna Vinschen]

On Jan  7 19:42, Daniel R. Grayson wrote:
> You must be reading her message differently from the way I did.

Yaakov got the message basically right.  I used the gc package as an
unfortunate example but this doesn't invalidate the fact that it would
be much more feasible to have the extra libs your package depends upon
as distinct Cygwin packages.  I explained why, and you would certainly
not be the first one who would pack extra packages which your main
package depends upon.

> In any case, if it were to turn out that Macaulay 2 were interesting to cygwin,
> but one of the libraries Macaulay 2 depends on was not interesting to cygwin,
> then I wouldn't have to package that library, would I?  I could just include
> the source for it in the Macaulay 2 package.  Otherwise it's a catch-22.
> 
> So I think the question of whether ITP's for the dependent packages would
> succeed on their own merits is moot.

Not really.  Here's the list of packages you talked about:

  gc               - already available.
  ntl              - Packed in Fedora, Cygwin package only needs GTG.
  singular-factory - Didn't find that one, but...
  singular-libfac  - Is packed in Fedora, only GTG needed.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

Dear Corinna,

Yaakov said I should submit ITP's for those dependent libraries, and that the
question of interest in Macaulay2 is moot until then, but you say I don't have
to submit ITP's for them, because they are either already provided or part of
Fedora, hence only a GTG is required.

Would you be willing to agree that Macaulay2's ITP can be considered now on its
own merits, without waiting for ITP's for those libraries to be approved?

The separate question of whether the cygwin developers want those libraries to
be packaged separately can be dealt with handily by conditional "yes" votes
that say "yes, Macaulay2 should be a package, but all 3rd party libraries must
be packaged separately rather than being included as source in the Macaulay2
package", or by unconditional votes that say "yes, Macaulay2 should be a
cygwin package".

> Date: Thu, 8 Jan 2009 11:30:11 +0100
> From: Corinna Vinschen <corinna-cygwin@cygwin.com>
> To: cygwin-apps@cygwin.com
> Subject: Re: [ITP] Macaulay2 1.1
> 
> On Jan  7 19:42, Daniel R. Grayson wrote:
> > You must be reading her message differently from the way I did.
> 
> Yaakov got the message basically right.  I used the gc package as an
> unfortunate example but this doesn't invalidate the fact that it would
> be much more feasible to have the extra libs your package depends upon
> as distinct Cygwin packages.  I explained why, and you would certainly
> not be the first one who would pack extra packages which your main
> package depends upon.
> 
> > In any case, if it were to turn out that Macaulay 2 were interesting to cygwin,
> > but one of the libraries Macaulay 2 depends on was not interesting to cygwin,
> > then I wouldn't have to package that library, would I?  I could just include
> > the source for it in the Macaulay 2 package.  Otherwise it's a catch-22.
> > 
> > So I think the question of whether ITP's for the dependent packages would
> > succeed on their own merits is moot.
> 
> Not really.  Here's the list of packages you talked about:
> 
>   gc               - already available.
>   ntl              - Packed in Fedora, Cygwin package only needs GTG.
>   singular-factory - Didn't find that one, but...
>   singular-libfac  - Is packed in Fedora, only GTG needed.

Re: [ITP] Macaulay2 1.1

[Yaakov (Cygwin/X)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Daniel R. Grayson wrote:
> Yaakov said I should submit ITP's for those dependent libraries, and that the
> question of interest in Macaulay2 is moot until then, but you say I don't have
> to submit ITP's for them, because they are either already provided or part of
> Fedora, hence only a GTG is required.

Let me clarify.  Anyone proposing to add a package to the Cygwin distro
requires two things:

1) show presence in other major distro (e.g. Fedora, Debian, or Ubuntu)
stable release OR 5 positive votes for inclusion.

2) a GTG review from a current Cygwin package maintainer.

The fact that Fedora has some of your prerequisites only helps with the
first point.  Cygwin != Linux, so we can't use Fedora's packages.  So
somebody needs to build and ITP those prereqs first before this package
can be properly reviewed.

> Would you be willing to agree that Macaulay2's ITP can be considered now on its
> own merits, without waiting for ITP's for those libraries to be approved?

No.


Yaakov
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEAREIAAYFAklmQksACgkQpiWmPGlmQSNdCwCfR92ZPTpQsey+YLUhds+Bb32K
DWEAoKRssd3qwouFa3BOdxZuJEMITKaa
=ptQf
-----END PGP SIGNATURE-----

[ITP] Macaulay2 1.1

[Daniel R. Grayson]

Dear Cygwin,

We'd like to distribute Macaulay 2 as an official cygwin package.  The package
is used by hundreds of mathematicians as part of their research, and hundreds
of published research papers, listed on our web site have cited it.  It has
been developed over the years through generous funding from the National
Science Foundation.

It is not included in a major Linux distribution.

The package files, including the setup.hint file, are visible at 

    http://www.math.uiuc.edu/Macaulay2/Repositories/cygwin/Macaulay2/

and are ready for uploading.

Here is the setup.hint file:

    # -*- sh -*-
    # this file is part of the Macaulay 2 cygwin package
    sdesc: "a software system for algebraic geometry research"
    ldesc: "Macaulay 2 is a software system for algebraic geometry research.
     Macaulay 2 is a software system for algebraic geometry research, written by
     Daniel R. Grayson and Michael E. Stillman.  Based on Groebner bases, it
     provides algorithms for computing homological invariants of rings and
     modules.
     Home page: http://www.math.uiuc.edu/Macaulay2/"
    category: Math
    # here are the requirements for version 1.1
    #   requires: cygwin lapack libgdbm4 libreadline6 libncurses8
    # the requirements for later versions are listed automatically and ought to be similar:
    requires: cygwin libgdbm4 libncurses8 libreadline6

--
              Daniel R. Grayson, Professor (retired, Emeritus)
              Department of Mathematics
              University of Illinois at Urbana-Champaign
        www:  http://www.math.uiuc.edu/~dan/
      email:  dan@math.uiuc.edu
    us mail:  2409 S. Vine St., Urbana, IL 61801, USA
      phone:  +1-217-367-6384 home (88.20224W, 40.08541N)
 cell phone:  +1-217-377-0458

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

> Date: Tue, 11 Nov 2008 19:52:16 -0600
> From: "Daniel R. Grayson" <dan@math.uiuc.edu>
> To: cygwin-apps@cygwin.com
> Subject: [ITP] Macaulay2 1.1

  ...

> It is not included in a major Linux distribution.

Oops, no, I was wrong.  It seems to be included in Fedora 8, 9, 10, thanks to
Rex Dieter, see

      https://admin.fedoraproject.org/pkgdb/packages/name/Macaulay2

Perhaps that will make the cygwin approval simpler.

Re: [ITP] Macaulay2 1.1

[Corinna Vinschen]

On Nov 13 08:23, Daniel R. Grayson wrote:
> 
> > Date: Tue, 11 Nov 2008 19:52:16 -0600
> > From: "Daniel R. Grayson" <dan@math.uiuc.edu>
> > To: cygwin-apps@cygwin.com
> > Subject: [ITP] Macaulay2 1.1
> 
>   ...
> 
> > It is not included in a major Linux distribution.
> 
> Oops, no, I was wrong.  It seems to be included in Fedora 8, 9, 10, thanks to
> Rex Dieter, see
> 
>       https://admin.fedoraproject.org/pkgdb/packages/name/Macaulay2
> 
> Perhaps that will make the cygwin approval simpler.

Yes, definitely.  I'm not familiar with this stuff at all, so I would be
glad if some other maintainer (Reini?  You're mathomatica maintainer so
you might have some affinity...) would take a look.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

PS: Macaulay2 depends on various other 3rd party packages, which are not yet
set up as cygwin packages, namely: singular-factory, gc, singular-libfac, and
ntl.  

I don't suppose someone has come forward to create those packages...

Our Makefile is set up to automatically download the source code for those
packages, in case they aren't found.  Perhaps you have a policy against package
makefiles doing that.  If so, then a simple option to the configure script will
prevent it.  If not, then I don't have to make all those other packages!

Alternatively, I could provide a bulky source code package that includes the
source code of those other packages.  But you might prefer me not to do that,
either.

Let me know your preferences.

Re: [ITP] Macaulay2 1.1

[Corinna Vinschen]

On Nov 13 12:16, Daniel R. Grayson wrote:
> 
> PS: Macaulay2 depends on various other 3rd party packages, which are not yet
> set up as cygwin packages, namely: singular-factory, gc, singular-libfac, and
> ntl.  

Ouch.

> I don't suppose someone has come forward to create those packages...
> 
> Our Makefile is set up to automatically download the source code for those
> packages, in case they aren't found.  Perhaps you have a policy against package
> makefiles doing that.  If so, then a simple option to the configure script will
> prevent it.  If not, then I don't have to make all those other packages!
> 
> Alternatively, I could provide a bulky source code package that includes the
> source code of those other packages.  But you might prefer me not to do that,
> either.
> 
> Let me know your preferences.

Usually you should provide the packages which your package depend on as
well, if nobody else is doing it.  It's really not feasible to depend on
something which isn't available to the user.  I'm wondering, why don't
you just provide and maintain these additional packages instead of
creating this bulky source code package?  In the end the effort
shouldn't be very different, given that they are otherwise built
automatically anyway.  There's also the fact that some of the above
packages provide shared libraries usually, at least the gc package.  You
can't use the shared libs from your package if you don't provide them as
well.  So you're linking statically, but does that make sense?  The gc
package at least might be useful for others as well.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

Re: [ITP] Macaulay2 1.1

[Yaakov (Cygwin Ports)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Corinna Vinschen wrote:
> The gc package at least might be useful for others as well.

If by gc you mean the Boehm-Demers-Weiser garbage collector library,
it's already packaged as libgc.


Yaakov
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEAREIAAYFAkkdPiIACgkQpiWmPGlmQSPlxQCg6kmrOzvMk+0X1M1WOFYoXJys
OzcAnA0bDHi8QpBPAKeJAm3pqXFUEecq
=GFCf
-----END PGP SIGNATURE-----

Re: [ITP] Macaulay2 1.1

[Corinna Vinschen]

On Nov 14 03:00, Yaakov (Cygwin Ports) wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
> 
> Corinna Vinschen wrote:
> > The gc package at least might be useful for others as well.
> 
> If by gc you mean the Boehm-Demers-Weiser garbage collector library,
> it's already packaged as libgc.

Uh, right.  Thanks for the hint.  I was just looking for gc.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

Re: [ITP] Macaulay2 1.1

[Daniel R. Grayson]

That's the one, thanks!

> Date: Fri, 14 Nov 2008 03:00:18 -0600
> From: "Yaakov (Cygwin Ports)" <yselkowitz@users.sourceforge.net>
> To: cygwin-apps@cygwin.com
> Subject: Re: [ITP] Macaulay2 1.1
> Mail-Followup-To: cygwin-apps@cygwin.com
> 
> Corinna Vinschen wrote:
> > The gc package at least might be useful for others as well.
> 
> If by gc you mean the Boehm-Demers-Weiser garbage collector library,
> it's already packaged as libgc.

Re: [ITP] Macaulay2 1.1

[Reini Urban]

2008/11/13 Corinna:
> Yes, definitely.  I'm not familiar with this stuff at all, so I would be
> glad if some other maintainer (Reini?  You're mathomatica maintainer so
> you might have some affinity...) would take a look.

Sorry, I'm terribly busy at work, and only have my laptop around.
-- 
Reini

Re: [ITP] Macaulay2 1.1

[Dr. Volker Zell]

>>>>> Daniel R Grayson writes:

    > Dear Cygwin,

    > We'd like to distribute Macaulay 2 as an official cygwin package.  The package
    > is used by hundreds of mathematicians as part of their research, and hundreds
    > of published research papers, listed on our web site have cited it.  It has
    > been developed over the years through generous funding from the National
    > Science Foundation.

    > It is not included in a major Linux distribution.

    > The package files, including the setup.hint file, are visible at 

    >     http://www.math.uiuc.edu/Macaulay2/Repositories/cygwin/Macaulay2/

    > and are ready for uploading.

    > Here is the setup.hint file:

    >     # -*- sh -*-
    >     # this file is part of the Macaulay 2 cygwin package
    >     sdesc: "a software system for algebraic geometry research"
    >     ldesc: "Macaulay 2 is a software system for algebraic geometry research.
    >      Macaulay 2 is a software system for algebraic geometry research, written by
    >      Daniel R. Grayson and Michael E. Stillman.  Based on Groebner bases, it
    >      provides algorithms for computing homological invariants of rings and
    >      modules.
    >      Home page: http://www.math.uiuc.edu/Macaulay2/"
    >     category: Math
    >     # here are the requirements for version 1.1
    >     #   requires: cygwin lapack libgdbm4 libreadline6 libncurses8
    >     # the requirements for later versions are listed automatically and ought to be similar:
    >     requires: cygwin libgdbm4 libncurses8 libreadline6


Building from source gives:

gcc -g -O2 -O3 -Wno-unused-label -Werror-implicit-function-declaration -I../../include -I../../../../Macaulay2/d/../../include -I../../../../Macaulay2/d -I. -I../util -I../e -I../../../../Macaulay2/d/../regex -I/misc/src/Macaulay2-1.1-6/BUILD/normal/libraries/final/usr/lib/Macaulay2/Core/include  -Wno-strict-aliasing -DNDEBUG -pipe -Wall              -Wshadow -Winline -Wcast-qual -Wno-parentheses -Wno-sign-compare  -c tmp_init.c -o tmp_init.o
time gcc  -L/misc/src/Macaulay2-1.1-6/BUILD/normal/libraries/final/usr/lib/Macaulay2/Core/lib   M2types.o M2mem.o scclib.o M2lib.o gmp_aux.o gdbm_interface.o gmp_memory.o factory_allocator.o main.o version.o M2-factory.o startup.o ../regex/regex.o gmp_init.o interp.oo texmacs.oo interface.oo actors5.oo actors4.oo actors3.oo actors2.oo actors.oo evaluate.oo libfac.oo objects.oo struct.oo GC.oo util.oo common.oo convertr.oo basic.oo binding.oo parser.oo lex.oo tokens.oo engine.oo gmp.oo err.oo stdiop.oo getline.oo stdio.oo varnets.oo nets.oo ctype.oo vararray.oo varstrin.oo strings.oo system.oo C.oo  tmp_init.o ../e/error.o ../e/table.o ../e/exptable.o ../e/monordering.o ../e/complex.o ../e/rand.o ../e/points.o ../e/overflow.o ../e/exceptions.o ../e/imonorder.o ../e/sagbi.o ../e/monideal-minprimes.o ../e/dmat-LU.o ../e/lapack.o ../e/LLL.o ../e/fractionfreeLU.o ../e/ntl-interface.o ../e/mat.o ../e/dmat.o ../e/smat.o ../e/coeffrings.o ../e/Eschreyer.o ../e/gbweight.o ../e/qring.o ../e/polyquotient.o ../e/poly.o ../e/debug.o ../e/comp.o ../e/comp-gb.o ../e/comp-gb-proxy.o ../e/comp-gb-declared.o ../e/comp-res.o ../e/res-a0.o ../e/res-a0-poly.o ../e/res-a1.o ../e/res-a1-poly.o ../e/res-a2.o ../e/reducedgb.o ../e/reducedgb-field.o ../e/reducedgb-marked.o ../e/reducedgb-field-local.o ../e/reducedgb-ZZ.o ../e/montable.o ../e/montableZZ.o ../e/schorder.o ../e/monorder.o ../e/GF.o ../e/ZZ.o ../e/QQ.o ../e/RRR.o ../e/CCC.o ../e/assprime.o ../e/buffer.o ../e/comb.o ../e/det.o ../e/frac.o ../e/freemod.o ../e/gb-default.o ../e/gb-homog2.o ../e/gb-sugarless.o ../e/gb-toric.o ../e/gb-walk.o ../e/gbring.o ../e/gauss.o ../e/hermite.o ../e/hilb.o ../e/int-bag.o ../e/intarray.o ../e/matrix.o ../e/matrix-con.o ../e/mem.o ../e/monideal.o ../e/monoid.o ../e/monomial.o ../e/ntuple.o ../e/polyring.o ../e/pfaff.o ../e/relem.o ../e/ring.o ../e/ringmap.o ../e/schur.o ../e/skew.o ../e/skewpoly.o ../e/solvable.o ../e/spair.o ../e/text-io.o ../e/varpower.o ../e/weylalg.o ../e/ZZp.o ../e/ntl-internal.o ../e/ntl-debugio.o ../e/matrix-kbasis.o ../e/matrix-symm.o ../e/matrix-sort.o ../e/ring-vecs.o ../e/res-a2-gb.o ../e/engine.o ../e/factor.o ../e/x-monom.o ../e/x-monoid.o ../e/x-relem.o ../e/x-factor.o ../e/x-free.o ../e/x-mat.o ../e/x-monideal.o ../e/x-gb.o ../e/x-ringmap.o ../e/x-mutablemat.o  ../e/f4/f4.o ../e/f4/f4-syz.o ../e/f4/f4-mem.o ../e/f4/f4-monlookup.o ../e/f4/f4-types.o ../e/f4/f4-computation.o ../e/f4/f4-spairs.o ../e/f4/f4-m2-interface.o ../e/f4/monhashtable.o ../e/f4/monsort.o ../e/f4/moninfo.o ../e/f4/ntuple-monomial.o ../e/f4/varpower-monomial.o ../e/f4/gausser.o  M2inits.o  -L. -lfactory_init2 -lfac -lcf -lfactory_init1 -llapack -lblas -lntl -lgc  -lntl -lgdbm -lgc -lmpfr -lgmp -lreadline -lhistory -lstdc++ -L/usr/local/gfortran/lib/gcc/i686-pc-cygwin/4.3.0 -L/usr/local/gfortran/lib/gcc/i686-pc-cygwin/4.3.0/../../.. -lgfortranbegin -lgfortran -lcygwin -luser32 -lkernel32 -ladvapi32 -lshell32 -lncurses  -lM2inits2 -lm -o "../bin/M2.exe".tmp
/usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../../i686-pc-cygwin/bin/ld: warning: auto-importing has been activated without --enable-auto-import specified on the command line.
This should work unless it involves constant data structures referencing symbols from auto-imported DLLs./usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_throw.o):(.text+0x76): undefined reference to `___w32_sharedptr_unexpected'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_throw.o):(.text+0x84): undefined reference to `___w32_sharedptr_terminate'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0xa5): undefined reference to `___w32_sharedptr_terminate'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0x122): undefined reference to `___w32_sharedptr_terminate'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0x195): undefined reference to `___w32_sharedptr_unexpected'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0x1a9): undefined reference to `___w32_sharedptr_terminate'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0x1e5): undefined reference to `___w32_sharedptr_terminate'
/usr/lib/gcc/i686-pc-cygwin/3.4.4/libstdc++.a(eh_terminate.o):(.text+0x205): undefined reference to `___w32_sharedptr_unexpected'
Info: resolving _rl_attempted_completion_over by linking to __imp__rl_attempted_completion_over (auto-import)
Info: resolving __rl_comment_begin by linking to __imp___rl_comment_begin (auto-import)
Info: resolving _rl_readline_name by linking to __imp__rl_readline_name (auto-import)
Info: resolving _rl_attempted_completion_function by linking to __imp__rl_attempted_completion_function (auto-import)
Info: resolving _rl_basic_word_break_characters by linking to __imp__rl_basic_word_break_characters (auto-import)
Info: resolving _rl_readline_version by linking to __imp__rl_readline_version (auto-import)
Info: resolving _gdbm_errno by linking to __imp__gdbm_errno (auto-import)
collect2: ld returned 1 exit status


By the way, check out Method Three: cygport from
http://cygwin.com/setup.html for packaging.

Ciao
  Volker