วันอาทิตย์ที่ 31 มกราคม พ.ศ. 2553

[sage-devel] Buildbot

Does anyone here have experience configuring and managing a Buildbot system?

http://buildbot.net/trac
http://djmitche.github.com/buildbot/docs/latest/

To enable continuous, automated builds on several platforms, we could

* Set up a meta-repository of the Sage Mercurial repositories and
spkgs. This could be an instantaneous alpha to which release managers
push patches with positive reviews.

* Modify sage-upgrade so it can pull updates directly from a set of
Sage Mercurial repositories, in addition to downloading and installing
new spkgs.

When the meta-repo changes, an autonomous buildmaster would trigger a
pull / build / test cycle on the available [remote] builders and collect
the results. Another set of builders might build Sage from scratch,
test optional spkgs, etc.

Or is it better to use sage-merge? Is it possible to pull a queue?

Anyway, since Buildbot is written in Python and depends only on Twisted,
we could make a Buildbot spkg and try to make it straightforward for
people who download Sage to submit build/test reports.


This is already

http://trac.sagemath.org/sage_trac/ticket/3524

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 2010-Jan-31 22:02:19 -0800, Nick Alexander <ncalexander@gmail.com> wrote:
>Not at all. But take away mathematics, and we don't have a
>*product*. Take away release management, fixing bugs, documentation,
>or maintaining the web site and we have an inferior project, but we
>still have a project. Take away support for Solaris or HP-UX, and we
>don't serve what appears, to me, to be a small market. (Certainly
>small compared to the potential pool of Microsoft Windows users.)

IMHO, that's a particularly unhelpful attitude. Take away the web
site and none of your potential users will know about your product or
be able to find/download it. Take away documentation and no-one will
know how to use it. Take away bug fixes and no-one can trust the
results returned by your product. Take away release management and
no-one can be sure whether the version they are running has specific
bugs fixed or not and what environment the product should run in.

Maybe you'd still have a project, but it wouldn't be one that was of
much use to anyone: If by some chance you managed to find an
executable and worked out how to drive it, it might be able to produce
an output very quickly but by the time you'd double-checked the
calculation, you might as well have done it by hand to start with.

--
Peter Jeremy

Re: [sage-devel] Re: MPIR 1.3.0 released (at last)

On 2010-Jan-30 11:51:08 +0000, "Dr. David Kirkby" <david.kirkby@onetel.net> wrote:
>Bill Hart wrote:
>> Cause of what David? MPIR 1.3.0 works absolutely fine on t2 if you set
>> the library paths correctly.
>
>I'm not convinced it should be necessary to set LD_LIBRARY_PATH like
>that. It is not with other 64-bit applications. But I may be mistaken

I'll go further: If it's necessary to set LD_LIBRARY_PATH for an
executable to find a compiler-specific library then the linker
configuration is broken. If gcc/g++ needs libgcc then it's up to it
to ensure that libgcc is linked into the executable with an appropriate
rpath (if necessary) to enable it to be found at runtime. If libtool
is overriding gcc's default linker, then it's up to libtool to ensure
that it correctly handles compiler-specific libraries.

In this case, I'm not sure whether it's gcc or libtool that is broken
but one of them is.

>But from experience, compiling code on multiple platforms often shows
>up errors not seen on other platforms, but lie in wait, ready to give
>the wrong answers at some point in the future.

Agreed.

--
Peter Jeremy

Re: [sage-devel] Re: MPIR supported Architectures/Compilers/OSes

On 31 January 2010 17:21, Bill Hart <goodwillhart@googlemail.com> wrote:
> Hmm, the link could help:
>
> http://mpir.org/supported.html
>
> It's not on the main MPIR webpage yet, as this is a proposal, which
> will need approval of the MPIR devels, after a period of discussion.
>
> Bill.

Under:

"Architectures we might like to support (but currently no assembly available)"

I would have thought 'Cuda' should go. Wolfram Research are certainly
adding Cuda support to Mathematica.

http://www.nvidia.com/object/io_1227010734073.html
http://www.youtube.com/watch?v=KfwXr_ErVlg

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 1 February 2010 05:51, Tim Lahey <tim.lahey@gmail.com> wrote:

> Solaris isn't exactly an "unusual" architecture. That's what he's done the most at
> supporting. He certainly has done "a LOT" at supporting it. I think what he's asking
> that Bill not purposely break FLINT since it does currently work.

If was MPIR I was requesting Bill did not purposely break, as that
works. I am not sure of the situation with Flint.

> I hate to think that the only people that are valid contributors to Sage are
> mathematicians. So, doing the release management, fixing bugs, documentation, or
> maintaining the web site aren't important?
>
> Cheers,
>
> Tim.

Thank you Tim.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 1 February 2010 06:02, Nick Alexander <ncalexander@gmail.com> wrote:

>> I hate to think that the only people that are valid contributors to Sage
>> are
>> mathematicians. So, doing the release management, fixing bugs,
>> documentation, or
>> maintaining the web site aren't important?
>
> Not at all.  But take away mathematics, and we don't have a *product*.  Take
> away release management, fixing bugs, documentation, or maintaining the web
> site and we have an inferior project, but we still have a project.

Without a web site, it would not be a product available to anyone easily.

> Take
> away support for Solaris or HP-UX, and we don't serve what appears, to me,
> to be a small market.  (Certainly small compared to the potential pool of
> Microsoft Windows users.)

Sun have donated a T5240, which William accepted. They have also
given him significant hardware discounts on other items. The main file
server, 'disk' is a Sun running Open Solaris. I believe sage.math is a
Sun running Linux, though I am not sure of that fact.

Only the other week, Willaim sent me an email telling me how important
it was to get the Solaris port completed.

I'm not denying the Windows port would get more users. than Solaris.
People are working on the Cygwin port.

I have no intension of running Sage on HP-UX. But I do believe that by
making code more portable, one does uncover bugs which do not show up
on one platform, but wait to hit you at a later date. I once spent
ages trying to trac down a bug on AIX, which was not reproducible, but
did occasionally occur. When I eventually found the bug, I realised it
would have affected any platform (including Linux), but had just not
showed up before. An ex-colleage who tested his code on a quad
processor SPARC for the first time, found a bug which could have
affected him on Linux, but he had not noticed it.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

>> By and large, we are a community of mathematicians. Correct me if
>> I'm wrong, but you are not contributing to the mathematical aspects
>> of Sage. Until that changes, your goals and my goals are only
>> occasionally aligned.
>
> I hate to think that the only people that are valid contributors to
> Sage are
> mathematicians. So, doing the release management, fixing bugs,
> documentation, or
> maintaining the web site aren't important?

Not at all. But take away mathematics, and we don't have a
*product*. Take away release management, fixing bugs, documentation,
or maintaining the web site and we have an inferior project, but we
still have a project. Take away support for Solaris or HP-UX, and we
don't serve what appears, to me, to be a small market. (Certainly
small compared to the potential pool of Microsoft Windows users.)

Nick

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 02-01-2010, at 12:45 AM, Nick Alexander wrote:
>
> I take issue with the claim that you have done "a LOT" for Sage. Let me be clear: I appreciate the effort you put into porting Sage to other architectures. But I question how many people are interested in actually using Sage on those architectures. You can read the obvious frustration Bill Hart feels around supporting flint and other software on these unusual architectures. I want to make it clear that he is not the only one frustrated with the recent emphasis on porting issues that are unlikely to affect me.
>

Solaris isn't exactly an "unusual" architecture. That's what he's done the most at
supporting. He certainly has done "a LOT" at supporting it. I think what he's asking
that Bill not purposely break FLINT since it does currently work.

> By and large, we are a community of mathematicians. Correct me if I'm wrong, but you are not contributing to the mathematical aspects of Sage. Until that changes, your goals and my goals are only occasionally aligned.


I hate to think that the only people that are valid contributors to Sage are
mathematicians. So, doing the release management, fixing bugs, documentation, or
maintaining the web site aren't important?

Cheers,

Tim.

---
Tim Lahey
PhD Candidate, Systems Design Engineering
University of Waterloo
http://www.linkedin.com/in/timlahey

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

> I think I've done a LOT for Sage - I would request you do not
> purposely break the PA-RISC support in MPIR, when it clearly passes
> all your self tests on HP-UX. I do not believe thiat is an
> unreasonable request.

I take issue with the claim that you have done "a LOT" for Sage. Let
me be clear: I appreciate the effort you put into porting Sage to
other architectures. But I question how many people are interested in
actually using Sage on those architectures. You can read the obvious
frustration Bill Hart feels around supporting flint and other software
on these unusual architectures. I want to make it clear that he is
not the only one frustrated with the recent emphasis on porting issues
that are unlikely to affect me.

By and large, we are a community of mathematicians. Correct me if I'm
wrong, but you are not contributing to the mathematical aspects of
Sage. Until that changes, your goals and my goals are only
occasionally aligned.

Nick Alexander

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: c++ question related to porting sage to Open Solaris

On 1 February 2010 04:53, Minh Nguyen <nguyenminh2@gmail.com> wrote:
> Hi David,
>
> On Mon, Feb 1, 2010 at 3:43 PM, David Kirkby <david.kirkby@onetel.net> wrote:
>
> <SNIP>
>
>> You have several options.
>>
>> 1) William has a virtual machine I believe.
>
> I vaguely recall a machine called "disk.math" or something that runs
> OpenSolaris.

Yes, that is true. But given disk's role as a central server, I do not
believe William wants to make that publicly available for development.
But you would have to check that with William. I could create a zone
on 'disk' which would provide security, and let people only use the
zone. Again that would be up to William. He would need to get me
another static IP address.

>
>> 2) You can install it in VirtualBox.
>
> That is possible. My local machine has 1 GB of RAM, which is
> sufficient for running one virtual machine with 250 MB of RAM. But my
> monthly Internet quota doesn't allow me to download Sage tarballs
> several times a month. Upgrading is an option. But for Sage
> development, I often require source tarballs.

Yes, 1 GB is not going to give you room to run many virtual machines.
In fact, I'm not sure OpenSolaris would be able to installed in 250 MB
of RAM.

My Ultra 27 has 12 GB RAM and 1.3 TB of disk space free. on the users
file system (there are a pair of 2 TB mirrored disks, plus another
pair of 500 GB mirrored disks for the system). So you wont run out of
resources too quickly.

>> 3) I would personally give you an account on my Ultra 27, which is
>> pretty quick, with a quad core 3.33 GHz Xeon.
>
> That would be kind of you.

No problem. I will do it in a few hours time, but can't do it just
now. If you still have the email I sent you with the HP-UX account
details, I will use the same password. If you have deleted it, let me
know and I'll send you the password in an email. It has SSH server,
but the port is non-standard. I can't recall what port you will need
to use, but it will not be 22. I'll sort out the details in a few
hours time.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 31 January 2010 14:27, Bill Hart <goodwillhart@googlemail.com> wrote:
> I don't see any point listing HP-UX. That platform died in 2004. I saw
> its grave.
>
> Here is one of the many obituaries:
>
> http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460

The latest release of HP-UX was September 2009 - 4 months ago.

http://h20338.www2.hp.com/enterprise/w1/en/os/hpux11i-v3-update5-overview.html

Typing HP-UX into Google I get 5,800,000 hits.

> some time between now and
> March this year, we will be removing the assembly support for the PA-
> RISC processors. Recently we found an actual PA-RISC machine (though
> not running HP-UX) and the assembly code was totally broken. No one
> has stepped forward to work on fixing it. No one we now has the
> expertise.

I would request you do not, given MPIR works and passes all tests on
HP-UX with at least the PA-RISC processor - I do not know about
Itanium.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Trac problem

On 01/31/2010 04:05 AM, Simon King wrote:
> When working on some trac tickets, I found that the "preview" button
> does not work.
>
> Trac claimed that it was an invalid operation. Afterwards, when

This *may* have happened during a server-side change, e.g., while
configuring a plug-in.

> returning to the trac ticket, "assign to SimonKing" was ticked, while
> it should have been "leaving needs_review".

I've seen this myself, but I don't know why it happens. The new
features should not have altered the existing ticket workflow, which
puts the "leave" transition at the top of the action list and selects it
by default. In particular, a diff of the "before" and "after" trac.ini
files shows the workflow configuration, at least, is the same.

> Are others experiencing similar problems? Now, as a wysywig editor can
> be used, "preview" might be unnecessary anyway. But once, it even
> happened that the "submit" button did not work when I hit it in
> wysywig mode.

The wysywig plug-in may well have some "quirks." I don't see this
particular problem mentioned at

http://trac-hacks.org/query?status=!closed&component=TracWysiwygPlugin&order=priority

but we could be seeing a new one. Is it reproducible? A possible
workaround, if it happens again, is to switch to textarea mode and try
again.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: c++ question related to porting sage to Open Solaris

Hi David,

On Mon, Feb 1, 2010 at 3:43 PM, David Kirkby <david.kirkby@onetel.net> wrote:

<SNIP>

> You have several options.
>
> 1) William has a virtual machine I believe.

I vaguely recall a machine called "disk.math" or something that runs
OpenSolaris.


> 2) You can install it in VirtualBox.

That is possible. My local machine has 1 GB of RAM, which is
sufficient for running one virtual machine with 250 MB of RAM. But my
monthly Internet quota doesn't allow me to download Sage tarballs
several times a month. Upgrading is an option. But for Sage
development, I often require source tarballs.


> 3) I would personally give you an account on my Ultra 27, which is
> pretty quick, with a quad core 3.33 GHz Xeon.

That would be kind of you.

--
Regards
Minh Van Nguyen

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: c++ question related to porting sage to Open Solaris

On 31 January 2010 21:20, Jaap Spies <j.spies@hccnet.nl> wrote:
> Jaap Spies wrote:
>>
>> Hi c++ experts,
>>
>> My C++ is a little bit rusty, so I'll ask here.

>> Any thoughts?
>>
>> Jaap
>>
>>
>>
>
> Ok, I seem to be on the ban list of everybody? Please help.
>
> jaap

Certainly not on my ban list - in fact nobody is.

But I do not know C++, so I can't help you either.

There are more than 7000 warnings issued when builing Sage on Solaris
10 SPARC. It would never surprise me if one of those is not considered
an error on another plattform. The Sun compiler considers some Sage
code as errors, and will not build it.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: c++ question related to porting sage to Open Solaris

On 31 January 2010 21:28, Minh Nguyen <nguyenminh2@gmail.com> wrote:
> Hi Jaap,
>
> On Mon, Feb 1, 2010 at 8:20 AM, Jaap Spies <j.spies@hccnet.nl> wrote:
>
> <SNIP>
>
>> Ok, I seem to be on the ban list of everybody?
>
> Certainly not. Not on mine.
>
>
>> Please help.
>
> I'm very new with OpenSolaris. But is there an OpenSolaris machine
> somewhere I could use to help?
>
> --
> Regards
> Minh Van Nguyen

You have several options.

1) William has a virtual machine I believe.

2) You can install it in VirtualBox.

3) I would personally give you an account on my Ultra 27, which is
pretty quick, with a quad core 3.33 GHz Xeon.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On 31 January 2010 14:27, Bill Hart <goodwillhart@googlemail.com> wrote:
> I don't see any point listing HP-UX. That platform died in 2004. I saw
> its grave.
>
> Here is one of the many obituaries:
>
> http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460
>
> I see you suggested Sage switch to GMP for an HP-UX port.

Yes, but now I find MPIR does build and pass all tests on HP-UX, I
would retract that - like all web pages, that one is a bit out of
date.

> Well, not
> only will MPIR not be supporting HP-UX, but some time between now and
> March this year, we will be removing the assembly support for the PA-
> RISC processors. Recently we found an actual PA-RISC machine (though
> not running HP-UX) and the assembly code was totally broken. No one
> has stepped forward to work on fixing it. No one we now has the
> expertise.

MPIR does pass all tests on the PA-RISC. Give it is working, is it a
good idea to purposely break it? The fact it failed on your PA-RISC
machine which was not running the normal operating system for such a
platform, which is HP-UX,

> We'll also be removing the assembly support for a plethora of other
> platforms that have long died, even their manufacturers disowning
> them. The rule will be: if people aren't using it, and we don't have
> access to one and it is more than a certain number of years old,
> support will be discontinued. Supporting dead architectures is a
> massive waste of developer effort.

I'm not asking you to support them. Just leave them running if they
work. PA-RISC on HP-UX does work.

> We urgently need ports to Solaris 64 bit

which I am working on, though concentrating on Open Solaris. One
reason for doing that is that I have much faster hardware on
OpenSolaris. I expect once the 64-bit issues are resolved on Open
Solaris, and assuming we can get the 32-bit working again (broken in
4.3.1), I think there is every chance a 64-bit port will occur on
SPARC. But perhaps not just now.

I think I've done a LOT for Sage - I would request you do not
purposely break the PA-RISC support in MPIR, when it clearly passes
all your self tests on HP-UX. I do not believe thiat is an
unreasonable request.

If someone tells you they use something, and it works for them, why go
out of the way to break it for them?

Dave.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] sage keeps me warm

On 31 January 2010 22:57, chris wuthrich <christian.wuthrich@gmail.com> wrote:
> Sorry for the spam; but I thought I'd share it with you.
>
> We had a big problem with our heating in our house a few weeks ago,
> exactly when it is was so freezing cold in Nottingham that they were
> talking about "the coldest winter". I imagined it is going to be a bit
> difficult to sleep when I could see my own breath inside my home. So I
> put my laptop in my bed and did a testall to warm up my bed. Sage has
> so many useful applications !
>
> Chris.

I with one or two computers on in a room attached to my garage to keep
the room sufficiently warm to stops pipes freezing. (One is a Blade
2000 runnng Solaris, the other a HP C3600 running HP-UX)., Some Sage
developement has been done on both those machines, so I beat you to
that one!!

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Our problems solved - OpenSSL is compatible with GPL !!

On 1 February 2010 03:26, David Roe <roed@math.harvard.edu> wrote:
> Note that the OpenSSL license is in the section of that page outlining
> licenses incompatible with the GPL.  Here's the summary of OpenSSL on that
> page:
>
> The license of OpenSSL is a conjunction of two licenses, one of them being
> the license of SSLeay. You must follow both. The combination results in a
> copyleft free software license that is incompatible with the GNU GPL. It
> also has an advertising clause like the original BSD license and the Apache
> license.
>
> We recommend using GNUTLS instead of OpenSSL in software you write. However,
> there is no reason not to use OpenSSL and applications that work with
> OpenSSL.
>
> So, sadly, this doesn't solve our problems.
> David

Oops, I missed that. How stupid of me.

In which case the conclusion is clear - the OpenSSL license is
incompatible with the GPL.

That could mean the end of a legal Solaris port, as the OpenSSL
libraries do not come with Solaris 10.

I could file a request for enhancement to Oracle (who bought Sun for 7
billion USD), for them to make OpenSSL part of the Solaris
distribution. That would save our bacon. Given Sun are effectively
sponsoring a Solaris port, providing free hardware for the purpose,
they should be pretty receptive to that idea. Perhaps William could
raise that with his contact at Sun.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Sage 4.3.0.1.alpha1 released for SPARC Solaris 10 (t2.math with GCC)

On 1 February 2010 03:33, David Kirkby <david.kirkby@onetel.net> wrote:
> On 31 January 2010 21:53, Minh Nguyen <nguyenminh2@gmail.com> wrote:
>> Hi folks,
>>
>> I'm happy to announce that Sage 4.3.0.1.alpha1 [1] successfully builds
>> on t2.math. Due to an unfortunate typo, the version number should be
>> Sage 4.3.0.2.alpha1 since it's based on Sage 4.3.0.1 [2]. This alpha
>> release is based on Sage 4.3.0.1.alpha0 and merged the following
>> tickets:
>>
>> #7770, #7692, #7749, #7768, #7728
>>
>> Unfortunately, I wasn't able to produce a binary distribution using
>> the command "sage -bdist". So I tar-gzip'd the alpha1 build and put it
>> under this directory:

< SNIP>

>> Regards
>> Minh Van Nguyen
>
> I have managed to produce a 4.3.0.1 binary on my old machine running
> the first release of Solaris 10.
>
> with sage -bdist, though I added some files. I've copied it to my home
> directory on t2. The trick is to include the GNU verson of 'cp' in the
> path before the Sun one.

I should have added, I bulit the GNU coreutils on 't2' and installed
them on 't2'. They are installed under

/usr/local/coreutils-8.4

I would *not* add the directory to your path on a permanent basis, as
doing so will mask issues that are specific to Solaris. But I would be
tempted to do

$ cp /usr/local/coreutils-8.4/bin/cp /some/tmp/directory
$ export PATH=/some/tmp/directory:$PATH

before running sage-bist.


Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Sage 4.3.0.1.alpha1 released for SPARC Solaris 10 (t2.math with GCC)

On 31 January 2010 21:53, Minh Nguyen <nguyenminh2@gmail.com> wrote:
> Hi folks,
>
> I'm happy to announce that Sage 4.3.0.1.alpha1 [1] successfully builds
> on t2.math. Due to an unfortunate typo, the version number should be
> Sage 4.3.0.2.alpha1 since it's based on Sage 4.3.0.1 [2]. This alpha
> release is based on Sage 4.3.0.1.alpha0 and merged the following
> tickets:
>
> #7770, #7692, #7749, #7768, #7728
>
> Unfortunately, I wasn't able to produce a binary distribution using
> the command "sage -bdist". So I tar-gzip'd the alpha1 build and put it
> under this directory:
>
> http://sage.math.washington.edu/home/mvngu/t2.math-bin/
>
> You can find the source tarball of alpha1 at
>
> http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/
>
> I'm using GCC 4.4.1 with the Sun linker/assembler, with the following
> conditional in my /home/mvngu/.profile:
>
> if [ `uname -s` = "SunOS" ]; then
>   # Sun linker/assembler
>   . /usr/local/gcc-4.4.1-sun-linker/gcc441sun
> fi
>
> Now on to building Sage 4.3.0.1.alpha2 on t2.math.
>
>
> [1] http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/sage-4.3.0.1.alpha1.tar
>
> [2] http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/sage-4.3.0.1.tar
>
> --
> Regards
> Minh Van Nguyen

I have managed to produce a 4.3.0.1 binary on my old machine running
the first release of Solaris 10.

with sage -bdist, though I added some files. I've copied it to my home
directory on t2. The trick is to include the GNU verson of 'cp' in the
path before the Sun one. I also felt it was desirable to include the
GCC libraries, so I added them in too.

The binary can be found here

http://boxen.math.washington.edu/home/kirkby/binaries/

This is compressed in lzma format, which is part of Solaris (both in
recent versions of Solaris, and a patch available from SunSolve for
older versions of Solaris). I need to check if that patch is public,
but I think it is.

However, I HAVE NOT TESTED THIS YET. But I intend testing it and
installing on 't2'. There is of course and advantage in including your
later release on 't2', but this one has the advantage it was compiled
on the first version of Solaris 10, so is better suited for wide
distribution.

However, given my comments about the OpenSSL license being compatible
with the GPL, and the fact the OpenSSL libraries do not come with
Solaris, I think it would be wise if I uploaded those libraries, and
created a new .binary with those libraries included. I will not
however do that until there is agreement of my logic about the OpenSSL
issue.

For what it is worth, it used 2 hours of CPU time on my 500 MHz
machine to create the lzma format binary, but it allowed me to upload
it in a few hours. With hindsight, it would have been better to copy a
tar file to a faster machine, and create the lzma format files there.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Our problems solved - OpenSSL is compatible with GPL !!

Note that the OpenSSL license is in the section of that page outlining licenses incompatible with the GPL.  Here's the summary of OpenSSL on that page:

The license of OpenSSL is a conjunction of two licenses, one of them being the license of SSLeay. You must follow both. The combination results in a copyleft free software license that is incompatible with the GNU GPL. It also has an advertising clause like the original BSD license and the Apache license.

We recommend using GNUTLS instead of OpenSSL in software you write. However, there is no reason not to use OpenSSL and applications that work with OpenSSL.

So, sadly, this doesn't solve our problems.
David

On Sun, Jan 31, 2010 at 10:05 PM, David Kirkby <david.kirkby@onetel.net> wrote:
I believe our problems are solved reguards the OpenSSL / GPL issue.
Follow this logic.

1) The GPL is copyright the Free Software Federation. It says that at
the top of the GPL license.

http://www.gnu.org/licenses/gpl-2.0.txt

2) The Free Sofware Federation contain a list of licenses compatible
with the GPL

http://www.fsf.org/licensing/licenses/#GPLCompatibleLicenses

In that list is the OpenSSL license

Dave.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Our problems solved - OpenSSL is compatible with GPL !!

I believe our problems are solved reguards the OpenSSL / GPL issue.
Follow this logic.

1) The GPL is copyright the Free Software Federation. It says that at
the top of the GPL license.

http://www.gnu.org/licenses/gpl-2.0.txt

2) The Free Sofware Federation contain a list of licenses compatible
with the GPL

http://www.fsf.org/licensing/licenses/#GPLCompatibleLicenses

In that list is the OpenSSL license

Dave.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] arm compile sage

Just curious how this is going if anyone has given up or its still happening.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] SageIDE/MathPiperIDE: another way to make a desktop app for Sage

Since the topic of desktop-based apps for Sage has come up recently on
this list, I thought it might be a good time to point out that a very
good way to create a desktop-based GUI for Sage is to use one of the
open source Java IDEs as a foundation. This is what I did when I
created SageIDE a couple of years ago and I have been very pleased
with the results.

Of course, when I left the Sage project I renamed SageIDE to
MathPiperIDE and converted it to use another CAS as its computation
engine, but what I would like to focus on is the benefits of using
this approach for creating a Sage desktop app. Here are some of the
benefits:

- Programmer-quality text editing capabilities.
- Programmer-quality search and replace tool.
- Edit modes for over 150 file types.
- Easy-to-use declarative syntax highlighting extension mechanism.
- Error highlighting.
- Plugin-based extension mechanism with versioning and dependencies.
- Dockable windows for the plugins.
- The whole application is extensible using numerous scripting languages.
- Solid 2D and 3D graphics capabilities.
- Cross platform (runs on Windows, Mac, Linux, and Solaris).
- Many other benefits.

Anyway, if anyone is interested in playing with SageIDE/MathPiperIDE
to get a feel for what a desktop GUI app for Sage might be like, it
can be downloaded from here:

http://mathpiper.org

Ted

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: sage keeps me warm

Given the climate here in Duluth, I would be happy to host a build
farm in my house...

-Marshall

On Jan 31, 5:23 pm, Martin Albrecht <m...@informatik.uni-bremen.de>
wrote:
> On Sunday 31 January 2010, chris wuthrich wrote:
>
> > Sorry for the spam; but I thought I'd share it with you.
>
> > We had a big problem with our heating in our house a few weeks ago,
> > exactly when it is was so freezing cold in Nottingham that they were
> > talking about "the coldest winter". I imagined it is going to be a bit
> > difficult to sleep when I could see my own breath inside my home. So I
> > put my laptop in my bed and did a testall to warm up my bed. Sage has
> > so many useful applications !
>
> I thought about the same thing yesterday when our heating broke down :)
>
> Cheers,
> Martin
>
> --
> name: Martin Albrecht
> _pgp:http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
> _otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
> _www:http://www.informatik.uni-bremen.de/~malb
> _jab: martinralbre...@jabber.ccc.de

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Unknown control sequence '\texttt'

On Jan 31, 1:37 pm, lutusp <lut...@gmail.com> wrote:
> > Then why do you bother posting things that look like requests for
> > help?
>
> In the hope that someone will fix the source, rather than offer new
> patches?

This is contradictory: the way the source gets fixed is that people
post patches, and then other people test them and review them.

--
John

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] sage keeps me warm

On Sunday 31 January 2010, chris wuthrich wrote:
> Sorry for the spam; but I thought I'd share it with you.
>
> We had a big problem with our heating in our house a few weeks ago,
> exactly when it is was so freezing cold in Nottingham that they were
> talking about "the coldest winter". I imagined it is going to be a bit
> difficult to sleep when I could see my own breath inside my home. So I
> put my laptop in my bed and did a testall to warm up my bed. Sage has
> so many useful applications !

I thought about the same thing yesterday when our heating broke down :)

Cheers,
Martin

--
name: Martin Albrecht
_pgp: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8EF0DC99
_otr: 47F43D1A 5D68C36F 468BAEBA 640E8856 D7951CCF
_www: http://www.informatik.uni-bremen.de/~malb
_jab: martinralbrecht@jabber.ccc.de

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] sage keeps me warm

Hi Chris,

On Mon, Feb 1, 2010 at 9:57 AM, chris wuthrich
<christian.wuthrich@gmail.com> wrote:

<SNIP>

> Sage has
> so many useful applications !

No one knows more about this than David Kirkby [1] :-)

Sage: creating a viable free open source alternative to Magma, Maple,
Mathematica and Matlab, preventing frozen water pipes and keeping
people warm at night. :-)

[1] http://groups.google.com/group/sage-devel/browse_thread/thread/73c2516a54c57c1

--
Regards
Minh Van Nguyen

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] sage keeps me warm

Sorry for the spam; but I thought I'd share it with you.

We had a big problem with our heating in our house a few weeks ago,
exactly when it is was so freezing cold in Nottingham that they were
talking about "the coldest winter". I imagined it is going to be a bit
difficult to sleep when I could see my own breath inside my home. So I
put my laptop in my bed and did a testall to warm up my bed. Sage has
so many useful applications !

Chris.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Sage 4.3.0.1.alpha1 released for SPARC Solaris 10 (t2.math with GCC)

Hi folks,

I'm happy to announce that Sage 4.3.0.1.alpha1 [1] successfully builds
on t2.math. Due to an unfortunate typo, the version number should be
Sage 4.3.0.2.alpha1 since it's based on Sage 4.3.0.1 [2]. This alpha
release is based on Sage 4.3.0.1.alpha0 and merged the following
tickets:

#7770, #7692, #7749, #7768, #7728

Unfortunately, I wasn't able to produce a binary distribution using
the command "sage -bdist". So I tar-gzip'd the alpha1 build and put it
under this directory:

http://sage.math.washington.edu/home/mvngu/t2.math-bin/

You can find the source tarball of alpha1 at

http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/

I'm using GCC 4.4.1 with the Sun linker/assembler, with the following
conditional in my /home/mvngu/.profile:

if [ `uname -s` = "SunOS" ]; then
# Sun linker/assembler
. /usr/local/gcc-4.4.1-sun-linker/gcc441sun
fi

Now on to building Sage 4.3.0.1.alpha2 on t2.math.


[1] http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/sage-4.3.0.1.alpha1.tar

[2] http://sage.math.washington.edu/home/mvngu/sage-src/sage-4.3.0.1/sage-4.3.0.1.tar

--
Regards
Minh Van Nguyen

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Prototype Sage Desktop App

On 01-31-2010, at 4:29 PM, Andy Somogyi wrote:

> Hello All
>
> Here is a link to a prototype sage desktop app for the Mac.
>
> http://numerator.sourceforge.net/SageApp.dmg
>
> Its 100% native, Cocoa model / view application. On startup, it creates a background process with the sage notebook server, and users can open as many windows as they want to it. It also has a console where it dumps the output of the sage process. As it is a native app, the download is about 100K in size. Its universal binary, I've tested it on Snow Leopard - core 2 duo, and Leopard - G4. Will not work with Tiger.
>
> Each window is just hosts a WebKit widget.
>
> The first time it is run, it will ask the user for the location of the sage runtime, this will typically be:
> /Application/sage/sage,
>
> this needs to be the actual sage executable script, you can set this with a file chooser dialog. If you want to change it in the future, it is stored in the standard defaults file of USERNAME/Library/Preferences/org.sagemath.Sage.App.plist
>
> The sage process takes about 5 - 10 seconds to start, when it starts, the first window will be told to open the localhost:8000 url.
>
> This is a first prototype, it has little functionality currently, will probably crash, almost certainly has numerous memory leaks, and is FAR, FAR, FAR from complete, but it is a START.
>
> Note, I am very happy to maintain and update this for the long term.
>
> Eventually, I want to have something that behaves similarly to the Mathematica desktop, and I think this is certainly possible with a Javascript/HTML5 based UI hosted in a set of WebKit views, and wrapped in a nice Cocoa app. The backed end / server is 100% cross platform, the client fits in nicely with the native desktop. It would be fairly simple for Windows person to come up with a Windows type app like this, and same for Linux / GTK person.
>

I certainly like this idea. However, the notebook isn't really the best thing for longer
functions. It's great for trying ideas out.

> So, try it out, let me know what features you would like to see.
>

Something along the lines of the MATLAB editor to edit functions that you can then
send to a cell in the WebKit view.

> Where would be a good place to store the source code to this?


Probably one of Google Code, BitBucket, or GitHub. This came up in regards to another Sage
related piece of code. Google Code supports Mercurial (like Sage) and offers 1GB of space. So, it's great for larger projects. BitBucket is solely Mercurial, but only offers 150MB. However, it supports a different workflow for sharing code and bringing in updates. Lastly, GitHub offers 300MB and supports a similar workflow to BitBucket (but better implemented in my opinion). The only problem is that it's designed for Git. It supports Mercurial through an hg-to-git extension that's in development.

Cheers,

Tim.

---
Tim Lahey
PhD Candidate, Systems Design Engineering
University of Waterloo
http://www.linkedin.com/in/timlahey

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Implementation of conjugacy classes

Dear Javier,

On Sat, Jan 09, 2010 at 02:01:07PM -0800, javier wrote:
> The implementation of the conjugacy classes is now ticket #7886:
> http://sagetrac.org/sage_trac/ticket/7886

I went through your patch, and am definitely +1 on the overall design
and user interface! Here are some suggestions for categorifying the
internals:

Split the ConjugacyClass into two classes (in the same module as before):

class ConjucacyClass(Parent):
# A parent, in the FiniteEnumeratedSets() category
# (see FiniteEnumeratedSets().example())
# Generic implementation with naive algorithm(s)

class ConjugacyClassGap(ConjugacyClass):
# Specializes the relevant methods of the above by calling GAP

In sage.categories.finite_groups.FiniteGroups, put in ParentMethods
the conjugacy class method for groups:

def conjugacy_class(self, g)
return ConjugacyClass(self, g)

and in ElementMethods the conjugacy class method for group elements:

def conjugacy_class(self):
return self.parent().conjugacy_class(self)

And finally, in those Sage groups which are implemented as GAP groups
(that is matrix and permutation groups), and only for those, override
the default method by:

def conjugacy_class(self, g)
return ConjugacyClassGAP(self, g)

Being in FiniteEnumeratedSets will give you for free some default
methods (like ``cardinality`` implemented by default in terms of
``list``), as well as sanity checks between
list/iter/cardinality/... when calling ``TestSuite(c).run()``.

It will be easy to later extend this to infinite groups when there
will be a concrete need/implementation.

By the way, a related feature which could afford a similar design, and
that would be useful in the short run for my PhD student would be:

sage: G = PermutationGroup(...)
sage: for C in G.conjugacy_classes():
... ... C.cardinality() ... C.representative() ...

I'd be happy to review it (and the above), should you feel like
working on it :-)

Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthiery@users.sf.net>
http://Nicolas.Thiery.name/

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Unknown control sequence '\texttt'

> Then why do you bother posting things that look like requests for
> help?

In the hope that someone will fix the source, rather than offer new
patches? It's not as though it hasn't been reported yet, or discussed
yet, and several new releases have gone by without resolution.
Meanwhile I'll get around the problem any way I can. The reason? Sage
is terrific.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: c++ question related to porting sage to Open Solaris

Minh Nguyen wrote:
> Hi Jaap,
>
> On Mon, Feb 1, 2010 at 8:20 AM, Jaap Spies<j.spies@hccnet.nl> wrote:
>
> <SNIP>
>
>> Ok, I seem to be on the ban list of everybody?
>
> Certainly not. Not on mine.
>
>

Thanks!

>> Please help.
>
> I'm very new with OpenSolaris. But is there an OpenSolaris machine
> somewhere I could use to help?
>


I use Open Solaris in VirtualBox, so be welcome.

My question is c++ related. I did my search, but found no solution yet.

Jaap

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Prototype Sage Desktop App

Hello All

Here is a link to a prototype sage desktop app for the Mac.

http://numerator.sourceforge.net/SageApp.dmg

Its 100% native, Cocoa model / view application. On startup, it creates a background process with the sage notebook server, and users can open as many windows as they want to it. It also has a console where it dumps the output of the sage process. As it is a native app, the download is about 100K in size. Its universal binary, I've tested it on Snow Leopard - core 2 duo, and Leopard - G4. Will not work with Tiger.

Each window is just hosts a WebKit widget.

The first time it is run, it will ask the user for the location of the sage runtime, this will typically be:
/Application/sage/sage,

this needs to be the actual sage executable script, you can set this with a file chooser dialog. If you want to change it in the future, it is stored in the standard defaults file of USERNAME/Library/Preferences/org.sagemath.Sage.App.plist

The sage process takes about 5 - 10 seconds to start, when it starts, the first window will be told to open the localhost:8000 url.

This is a first prototype, it has little functionality currently, will probably crash, almost certainly has numerous memory leaks, and is FAR, FAR, FAR from complete, but it is a START.

Note, I am very happy to maintain and update this for the long term.

Eventually, I want to have something that behaves similarly to the Mathematica desktop, and I think this is certainly possible with a Javascript/HTML5 based UI hosted in a set of WebKit views, and wrapped in a nice Cocoa app. The backed end / server is 100% cross platform, the client fits in nicely with the native desktop. It would be fairly simple for Windows person to come up with a Windows type app like this, and same for Linux / GTK person.

So, try it out, let me know what features you would like to see.

Where would be a good place to store the source code to this?

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: c++ question related to porting sage to Open Solaris

Hi Jaap,

On Mon, Feb 1, 2010 at 8:20 AM, Jaap Spies <j.spies@hccnet.nl> wrote:

<SNIP>

> Ok, I seem to be on the ban list of everybody?

Certainly not. Not on mine.


> Please help.

I'm very new with OpenSolaris. But is there an OpenSolaris machine
somewhere I could use to help?

--
Regards
Minh Van Nguyen

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: c++ question related to porting sage to Open Solaris

Jaap Spies wrote:
> Hi c++ experts,
>
> My C++ is a little bit rusty, so I'll ask here.
>
> Building matplotlib, pynac, scipy and scipysandbox fail in the end with
>
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:
>> In static member function 'static int
>> std::char_traits<wchar_t>::compare(const wchar_t*, const wchar_t*,
>> size_t)':
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:330:
>> error: 'wmemcmp' was not declared in this scope
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:
>> In static member function 'static const wchar_t*
>> std::char_traits<wchar_t>::find(const wchar_t*, size_t, const wchar_t&)':
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:338:
>> error: invalid conversion from 'const wchar_t*' to 'wchar_t*'
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:338:
>> error: initializing argument 1 of 'wchar_t* std::wmemchr(wchar_t*,
>> wchar_t, size_t)'
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:
>> In static member function 'static wchar_t*
>> std::char_traits<wchar_t>::move(wchar_t*, const wchar_t*, size_t)':
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:342:
>> error: 'wmemmove' was not declared in this scope
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:
>> In static member function 'static wchar_t*
>> std::char_traits<wchar_t>::copy(wchar_t*, const wchar_t*, size_t)':
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:346:
>> error: 'wmemcpy' was not declared in this scope
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:
>> In static member function 'static wchar_t*
>> std::char_traits<wchar_t>::assign(wchar_t*, size_t, wchar_t)':
>> /usr/local/gcc-4.4.2/lib/gcc/i386-pc-solaris2.11/4.4.2/../../../../include/c++/4.4.2/bits/char_traits.h:350:
>> error: 'wmemset' was not declared in this scope
>
>
> Any thoughts?
>
> Jaap
>
>
>

Ok, I seem to be on the ban list of everybody? Please help.

jaap

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] sage-mode and indentation

On Sat, Jan 23, 2010 at 11:09:24AM -0800, Nick Alexander wrote:
> Why don't you update the sage-mode wiki ...

Upon googling for the wiki, the first hit was:

http://naruto.wikia.com/wiki/Sage_Mode

"Sage Mode is the result of using natural energy along with a
ninja's normal chakra in perfect balance to drastically
increase a ninja's abilities."

Wow! Sage + Emacs really seems the way go :-)

Nicolas
--
Nicolas M. Thiéry "Isil" <nthiery@users.sf.net>
http://Nicolas.Thiery.name/

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] sage-mode and indentation

On Sat, Jan 23, 2010 at 08:58:28PM +0000, John Cremona wrote:
> Some of those might have come from me since I use emacs and I edited
> some combinat docstrings recently (they were giveing errors when
> building the docs).

Those we just fixed were from someone else :-)

> I would like to have emacs automatically replace tabs by spaces, and
> if the 3 lines you supply will do that (after being inserted in my
> .emacs) then I'll use them right away.

That should do the job for new / reindented lines.

> I already have to use emacs's untabify command after editing .pyx
> files since they don't compile at all if they have tabs. (So the
> .emacs will need to know to apply this to .pyx files and not just
> .py files).

I have this in my .emacs to put .pyx files in sage-mode instead of
pyrex-mode.

(setq auto-mode-alist (cons '("\\.pyx\\'" . sage-mode) auto-mode-alist))

Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthiery@users.sf.net>
http://Nicolas.Thiery.name/

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Supported Platforms web page is *very* out of date

Ah!

http://www.linux.com/news/enterprise/biz-enterprise/266916-red-hat-pulls-plug-on-itanium-with-rhel-6

That leaves debian, which still supports it officially, unofficial
support on Ubuntu and support for ia32 on SUSE.

But that leads me to question the future of ia64 itself. I don't
personally believe HP will keep it alive for long.

Bill.

On Jan 31, 6:28 pm, Bill Hart <goodwillh...@googlemail.com> wrote:
> OK, I did some reading and I now see the point of the question.
>
> At this point I don't see any problem with Linux on Itanium 2. For
> example the gcc build farm contains an Itanium 2 (though no longer an
> Itanium), and gcc itself support Itanium 2, as does the assembler
> (obviously).
>
> Are there any articles which you could point me to which might change
> my mind about Linux on Itanium?
>
> Certainly if Linux didn't have a future on Itanium, then that would
> indeed be relevant.
>
> Bill.
>
> On Jan 31, 3:28 pm, William Stein <wst...@gmail.com> wrote:
>
>
>
> > On Sun, Jan 31, 2010 at 6:27 AM, Bill Hart <goodwillh...@googlemail.com> wrote:
> > > I don't see any point listing HP-UX. That platform died in 2004. I saw
> > > its grave.
>
> > > Here is one of the many obituaries:
>
> > >http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460
>
> > > I see you suggested Sage switch to GMP for an HP-UX port. Well, not
> > > only will MPIR not be supporting HP-UX, but some time between now and
> > > March this year, we will be removing the assembly support for the PA-
> > > RISC processors. Recently we found an actual PA-RISC machine (though
> > > not running HP-UX) and the assembly code was totally broken. No one
> > > has stepped forward to work on fixing it. No one we now has the
> > > expertise.
>
> > > We'll also be removing the assembly support for a plethora of other
> > > platforms that have long died, even their manufacturers disowning
> > > them. The rule will be: if people aren't using it, and we don't have
> > > access to one and it is more than a certain number of years old,
> > > support will be discontinued. Supporting dead architectures is a
> > > massive waste of developer effort.
>
> > > We urgently need ports to Solaris 64 bit and to Windows Vista and  and
> > > Windows 7, 32 and 64 bit. We should be focusing all our efforts on
> > > these important ports, instead of dead platforms/architectures like
> > > AIX, HP-UX and TRU64.
>
> > > Bill.
>
> > In your mind, is Linux on Itanium2 dead or alive?
>
> > William

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] possible singular bug?

The following example

sage: p = 17
sage: F = GF(p)
sage: P2.<X,Y,Z> = ProjectiveSpace(F,2)
sage: C = Curve(X^2+Y^2-Z^2)
sage: len(C.rational_points())
18
sage: C.rational_points(algorithm='bn')
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
...
RuntimeError: Singular error:
Computing affine singular points ...
Computing all points at infinity ...
Computing affine singular places ...
Computing singular places at infinity ...
Computing non-singular places at infinity ...
The given polynomial is a unit in the power series ring!
// ** right side is not a datum, assignment ignored
? `_` is undefined
? error occurred in brnoeth.lib::place line 1155: ` return(HND); `
? leaving brnoeth.lib::place
skipping text from `;` error at token `)`
? leaving brnoeth.lib::Adj_div

** Unable to use the Brill-Noether Singular package to compute all
points (see above).

suggests to me that there is a bug in Singular's code to enumerate the
rational points on a (very simple) curve. If it's not a bug then we
need to find out what the restrictions are on the singular function
being called, and catch this.

Has anyone seen this before?

John

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Supported Platforms web page is *very* out of date

OK, I did some reading and I now see the point of the question.

At this point I don't see any problem with Linux on Itanium 2. For
example the gcc build farm contains an Itanium 2 (though no longer an
Itanium), and gcc itself support Itanium 2, as does the assembler
(obviously).

Are there any articles which you could point me to which might change
my mind about Linux on Itanium?

Certainly if Linux didn't have a future on Itanium, then that would
indeed be relevant.

Bill.

On Jan 31, 3:28 pm, William Stein <wst...@gmail.com> wrote:
> On Sun, Jan 31, 2010 at 6:27 AM, Bill Hart <goodwillh...@googlemail.com> wrote:
> > I don't see any point listing HP-UX. That platform died in 2004. I saw
> > its grave.
>
> > Here is one of the many obituaries:
>
> >http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460
>
> > I see you suggested Sage switch to GMP for an HP-UX port. Well, not
> > only will MPIR not be supporting HP-UX, but some time between now and
> > March this year, we will be removing the assembly support for the PA-
> > RISC processors. Recently we found an actual PA-RISC machine (though
> > not running HP-UX) and the assembly code was totally broken. No one
> > has stepped forward to work on fixing it. No one we now has the
> > expertise.
>
> > We'll also be removing the assembly support for a plethora of other
> > platforms that have long died, even their manufacturers disowning
> > them. The rule will be: if people aren't using it, and we don't have
> > access to one and it is more than a certain number of years old,
> > support will be discontinued. Supporting dead architectures is a
> > massive waste of developer effort.
>
> > We urgently need ports to Solaris 64 bit and to Windows Vista and  and
> > Windows 7, 32 and 64 bit. We should be focusing all our efforts on
> > these important ports, instead of dead platforms/architectures like
> > AIX, HP-UX and TRU64.
>
> > Bill.
>
> In your mind, is Linux on Itanium2 dead or alive?
>
> William

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: MPIR supported Architectures/Compilers/OSes

Hmm, the link could help:

http://mpir.org/supported.html

It's not on the main MPIR webpage yet, as this is a proposal, which
will need approval of the MPIR devels, after a period of discussion.

Bill.

On Jan 31, 5:17 pm, Bill Hart <goodwillh...@googlemail.com> wrote:
> I've made a page of Architectures/Compilers/OSes that MPIR should
> recognise, along with a *proposed* categorisation according to how
> much attention the MPIR developers do/should pay to each.
>
> I'm posting this to the mpir-devel and sage-devel lists for comment,
> as this is just as relevant for the Sage folks as it is for MPIR.
>
> Have we missed anything important? Anyone think something should be
> moved up/down the list?
>
> Anyone want to volunteer to maintain support/complete a port of MPIR
> for one or more of these, or to volunteer to make a machine available
> for testing (especially if it lives in the supported but not tested
> category)?
>
> Bill.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Supported Platforms web page is *very* out of date

I've posted a list of arches/compilers/OSes that MPIR currently does/
perhaps should support, in another thread. That should answer the
question, I think.

On Jan 31, 3:28 pm, William Stein <wst...@gmail.com> wrote:
> On Sun, Jan 31, 2010 at 6:27 AM, Bill Hart <goodwillh...@googlemail.com> wrote:
> > I don't see any point listing HP-UX. That platform died in 2004. I saw
> > its grave.
>
> > Here is one of the many obituaries:
>
> >http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460
>
> > I see you suggested Sage switch to GMP for an HP-UX port. Well, not
> > only will MPIR not be supporting HP-UX, but some time between now and
> > March this year, we will be removing the assembly support for the PA-
> > RISC processors. Recently we found an actual PA-RISC machine (though
> > not running HP-UX) and the assembly code was totally broken. No one
> > has stepped forward to work on fixing it. No one we now has the
> > expertise.
>
> > We'll also be removing the assembly support for a plethora of other
> > platforms that have long died, even their manufacturers disowning
> > them. The rule will be: if people aren't using it, and we don't have
> > access to one and it is more than a certain number of years old,
> > support will be discontinued. Supporting dead architectures is a
> > massive waste of developer effort.
>
> > We urgently need ports to Solaris 64 bit and to Windows Vista and  and
> > Windows 7, 32 and 64 bit. We should be focusing all our efforts on
> > these important ports, instead of dead platforms/architectures like
> > AIX, HP-UX and TRU64.
>
> > Bill.
>
> In your mind, is Linux on Itanium2 dead or alive?
>
> William

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] MPIR supported Architectures/Compilers/OSes

I've made a page of Architectures/Compilers/OSes that MPIR should
recognise, along with a *proposed* categorisation according to how
much attention the MPIR developers do/should pay to each.

I'm posting this to the mpir-devel and sage-devel lists for comment,
as this is just as relevant for the Sage folks as it is for MPIR.

Have we missed anything important? Anyone think something should be
moved up/down the list?

Anyone want to volunteer to maintain support/complete a port of MPIR
for one or more of these, or to volunteer to make a machine available
for testing (especially if it lives in the supported but not tested
category)?

Bill.

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Re: Supported Platforms web page is *very* out of date

On Sun, Jan 31, 2010 at 6:27 AM, Bill Hart <goodwillhart@googlemail.com> wrote:
> I don't see any point listing HP-UX. That platform died in 2004. I saw
> its grave.
>
> Here is one of the many obituaries:
>
> http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460
>
> I see you suggested Sage switch to GMP for an HP-UX port. Well, not
> only will MPIR not be supporting HP-UX, but some time between now and
> March this year, we will be removing the assembly support for the PA-
> RISC processors. Recently we found an actual PA-RISC machine (though
> not running HP-UX) and the assembly code was totally broken. No one
> has stepped forward to work on fixing it. No one we now has the
> expertise.
>
> We'll also be removing the assembly support for a plethora of other
> platforms that have long died, even their manufacturers disowning
> them. The rule will be: if people aren't using it, and we don't have
> access to one and it is more than a certain number of years old,
> support will be discontinued. Supporting dead architectures is a
> massive waste of developer effort.
>
> We urgently need ports to Solaris 64 bit and to Windows Vista and  and
> Windows 7, 32 and 64 bit. We should be focusing all our efforts on
> these important ports, instead of dead platforms/architectures like
> AIX, HP-UX and TRU64.
>
> Bill.

In your mind, is Linux on Itanium2 dead or alive?

William

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Supported Platforms web page is *very* out of date

I don't see any point listing HP-UX. That platform died in 2004. I saw
its grave.

Here is one of the many obituaries:

http://www.chillingeffects.org/responses/notice.cgi?NoticeID=1460

I see you suggested Sage switch to GMP for an HP-UX port. Well, not
only will MPIR not be supporting HP-UX, but some time between now and
March this year, we will be removing the assembly support for the PA-
RISC processors. Recently we found an actual PA-RISC machine (though
not running HP-UX) and the assembly code was totally broken. No one
has stepped forward to work on fixing it. No one we now has the
expertise.

We'll also be removing the assembly support for a plethora of other
platforms that have long died, even their manufacturers disowning
them. The rule will be: if people aren't using it, and we don't have
access to one and it is more than a certain number of years old,
support will be discontinued. Supporting dead architectures is a
massive waste of developer effort.

We urgently need ports to Solaris 64 bit and to Windows Vista and and
Windows 7, 32 and 64 bit. We should be focusing all our efforts on
these important ports, instead of dead platforms/architectures like
AIX, HP-UX and TRU64.

Bill.

On Jan 31, 10:44 am, "Dr. David Kirkby" <david.kir...@onetel.net>
wrote:
> http://wiki.sagemath.org/SupportedPlatforms
>
> is very out of date. This is no criticism o the web master - I know only too
> well it is next to impossible to keep a web site up to date.
>
> Here are the points.
>
> 1) Solaris 9 on Sparc 32 bit (ongoing, getting close, mabshoff is working on this)
>
> Truth: mabshoff has left.
>
> Unless someone else comes along and does some work, I can't see this really
> happening.
>
> 2) Solaris 9 on Sparc 64 bit (work will start on this once the 32 bit port works
>
> Truth: Nobody is making any effort on this front.
>
> 3) Solaris 10 on Sparc 32 bit (ongoing, getting close, mabshoff is working on this)
>
> Truth: Solaris 10 32-bit SPARC builds easily on 4.3.0.1, though some test
> failures are observed, which need to be resolved.
>
> 4) Solaris 10 on Sparc 64 bit (work will start on this once the 32 bit port works)
>
> Truth: Open Solaris 64-bit is seen as a higher priority, but one might expect
> that 64-bit Solaris 10 on SPARC will work at some point.
>
> 5) Solaris 10 on x86 32 bit, Opteron/EMT64 32 bit (ongoing, getting close,
> mabshoff is working on this)
>
> Truth: Nobody is working on a Solaris 10 32-bit x86 port and I doubt they ever
> will. We are going directly to 64-bit, and certainly will not bother then making
> a 32-bit build.
>
> 6) Solaris 10 on Opteron/EMT64 64 bit (work will start on this once the 32 bit
> port works)
>
> Truth. Active development for a 64-bit Open Solaris build, which will run on any
> modern processor from Intel or AMD.
>
> 7) BSD support on x86, x86-64: most likely FreeBSD
> Truth. There is active development for a FreeBSD port, but not for the other BSDs'
>
> 8) the AIX 64 bit port is in the early stages - it can be followed at The AIX 64
> bit port page
>
> Truth: Nobody is seriously working on an AIX port.
>
> 9) I am actually putting some effort into an HP-UX port. Whether this ever gets
> completed is another matter. Details at
>
> http://wiki.sagemath.org/HP-UX
>
> Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Should SSL support be checked in 'prereq' ?

On Sun, 31 Jan 2010 at 10:15AM +0000, Dr. David Kirkby wrote:
> 3) Change Sage so that the hashlib module of python is not essential
> for a functioning Sage. That is I suspect the easiest option. I
> don't claim to understand how Sage builds fully, but I would have
> thought crypto support was not a requirement.

People in this thread have been discussing this issue only in terms of
encryption, but hashlib also does...well, hashing, such as MD5 and SHA1.
We make a big deal out of Sage being Python-based, and Python in turn
says that the "batteries are included" -- these batteries typically
include being able to easily take hashes of strings, so I would strongly
prefer that the hashlib module stay the way it is.

(Also, SageTeX needs md5 hashing support in its current state, and I'd
be unhappy if, just after getting it merged, some of its functionality
needs to be changed!)

Dan

--
--- Dan Drake
----- http://mathsci.kaist.ac.kr/~drake
-------

Re: [sage-devel] Should SSL support be checked in 'prereq' ?

On Sun, Jan 31, 2010 at 12:46 AM, Dr. David Kirkby
<david.kirkby@onetel.net> wrote:
> William Stein wrote:
> Personally, I do not believe it is legal to ship OpenSSL and for Sage to
> remain GPL, unless you could get the python developers agree to add an
> clause that permits linking against OpenSSL.
>
> http://www.openssl.org/support/faq.html#LEGAL2
>
> says. "If you are using GPL software developed by others, you may want to
> ask the copyright holder for permission to use their software with OpenSSL."

Python is not GPL software.

>> Note that a few years ago, Sage did ship opensll and not ship the
>> gnutls stack.   Then a student in my Sage class pointed out that
>> openssl is licensed in a GPL-incompatible way, and that's when we
>> switched to GNUtls.
>
> I agree with him/her.

Do you mean that you agree with the statement: "The Openssl license is
not GPL-compatible?"


--
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

[sage-devel] Re: Should updates only be permitted with same gcc version?

Dr. David Kirkby wrote:
[snipped]
>
> Fair enough. I mis understood this. But I still think we should not let
> someone upgrade Sage with a version different to what they used to build
> it.
>

-1

I've a system wide install of sage upgraded 86 times:
[root@paix installed]# ls sage-*
sage-1.5.1.2 sage-2.1.0.1 sage-2.4.1.2 sage-2.8.12 sage-2.8.6 sage-3.0.6 sage-4.0
sage-1.5.3 sage-2.10.1 sage-2.4.2 sage-2.8.13 sage-2.8.7 sage-3.1 sage-4.0.1
sage-1.6 sage-2.10.2 sage-2.5 sage-2.8.14 sage-2.8.8 sage-3.1.1 sage-4.0.2
sage-1.6.1 sage-2.10.3 sage-2.5.0.2 sage-2.8.15 sage-2.8.8.1 sage-3.1.2 sage-4.1
sage-1.7 sage-2.10.4 sage-2.5.1 sage-2.8.2 sage-2.8.9 sage-3.1.4 sage-4.1.1
sage-1.7.1 sage-2.11 sage-2.5.2 sage-2.8.3 sage-2.9 sage-3.2 sage-4.1.2
sage-1.8 sage-2.1.3 sage-2.5.3 sage-2.8.3.3 sage-2.9.1.1 sage-3.2.1 sage-4.2
sage-1.8.1 sage-2.1.3.1 sage-2.6 sage-2.8.3.6 sage-2.9.2 sage-3.2.2 sage-4.2.1
sage-1.8.2.1 sage-2.1.4 sage-2.7 sage-2.8.4 sage-3.0 sage-3.2.3 sage-4.3
sage-1.9 sage-2.2 sage-2.7.3 sage-2.8.4.1 sage-3.0.2 sage-3.3
sage-2.0 sage-2.3 sage-2.8 sage-2.8.4.2 sage-3.0.3 sage-3.4
sage-2.1 sage-2.4 sage-2.8.10 sage-2.8.5 sage-3.0.4 sage-3.4.1
sage-2.10 sage-2.4.1 sage-2.8.11 sage-2.8.5.1 sage-3.0.5 sage-3.4.2
[root@paix installed]# ls sage-* | wc
87 87 945

Only last time things went wrong in upgrading to sage-4.3.1.

I have to look into that.

Jaap


> Dave
>


--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Re: [sage-devel] Should updates only be permitted with same gcc version?

Alex Ghitza wrote:
> On Sun, 31 Jan 2010 09:47:00 +0000, "Dr. David Kirkby" <david.kirkby@onetel.net> wrote:
>> The issue reported on sage-support makes me think we should insist that updates
>> are performed with the same version of gcc as that which was used to build gcc
>> in the first placed. Basically, two people have found
>>
>> 1) Install Sage with older gcc.
>> 2) Install the latest gcc 4.4.3
>> 3) Try to update Sage, and it fails
>> 4) Build Sage from the start, using gcc 4.4.3 and it works.
>>
>
> Again, that's not what happened (but I guess I didn't explain it
> properly). Here's the actual run of events:

Sorry. I should have read the thread more carefully - it is not your fault.

> 1) Build sage-4.3.1 with gcc-4.4.2 on Arch Linux; Sage runs fine, passes
> doctests, etc. Someone else gets the binary and can use it to run Sage
> 2) Upgrade gcc to gcc-4.4.3
> 3) Try to run sage, get errors that the original poster listed

So you are saying that simply updating gcc, causes Sage to no longer work? That
seems very naughty of the shipped gcc/libc if that is happening. I could
understand if one removed a new version of gcc, and replaced it with an old
version this might happen.

But I think when shipping binaries, we should include ALL the libraries used by
gcc with Sage. That means:

* libgcc_s
* libgfortran
* libstdc++.

I've said this before, but nobody has really agreed with me. I think there could
be issues if Sage is built with a new gcc, and someone has only an old one.

I'm just in the process of building a SPARC binary package. I decided to use
LZMA compression, which is going to use 2 hours of CPU time on my oldest SPARC
machine. But I have included those libraries.

On Solaris at least, only a very old gcc is shipped, so I suspect the libraries
will be too out of date. If I build Sage with gcc 4.4.1, and someone has only
gcc 3.4.3 (shipped with Solaris), then I think their library is likely to be too
old. So it's better if we ship one with the same version of gcc as used to build
Sage.

> 4) Rebuild sage-4.3.1 (so the same version) from scratch with gcc-4.4.3,
> and everything is fine again (for me, but not for the other person who
> couldn't use my binary because of a cpu flag issue)

Fair enough. I mis understood this. But I still think we should not let someone
upgrade Sage with a version different to what they used to build it.

Dave

--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to sage-devel+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org