[sage-devel] Re: sage-4.1.2 and sage-4.2

[Stan Schymanski]

Dear William,

I tried to compile from source on my Macbook Pro with OSX 10.4.11 but 
got the following failure:

make[5]: *** [gnutls_buffers.lo] Killed
make[4]: *** [all-recursive] Error 1
make[3]: *** [all-recursive] Error 1
make[2]: *** [all] Error 2
failed to build GNUTLS

real1m37.444s
user0m29.013s
sys 0m51.722s
sage: An error occurred while installing gnutls-2.2.1.p4

I didn't find any binaries for 10.4.11, either. Has anyone run it on 
10.4.11?

Cheers,
Stan



William Stein wrote:
> Hi,
>
> I. SAGE-4.1.2:   Sage-4.1.2 has been released.
>
>1. You can download the source code here:
> http://sage.math.washington.edu/sage/src/sage-4.1.2.tar
> 2. I've posted many binaries here:
> http://sage.math.washington.edu/home/wstein/binaries/
> (not, a few are still being uploaded)
> Can somebody build an OS X 10.5 Intel binary?
>3. Experts: you can upgrade with "sage -upgrade".
>
> II. SAGE-4.2:  The official release manager is Mike Hansen 
> (mhan...@gmail.com).
> See http://trac.sagemath.org/sage_trac/milestone/sage-4.2 for the roadmap.
>
>  -- William
>
>   

-- 


Stan Schymanski
Scientist
Max Planck Institute for Biogeochemistry
Postfach 10 01 64
D-07701 Jena

Phone: +49.3641.576264
Fax: +49.3641.577274
WWW: http://www.bgc-jena.mpg.de/~sschym

Biospheric Theory and Modelling Group
http://www.bgc-jena.mpg.de/bgc-theory/
_


--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 10:52 PM, Bill Page  wrote:
>
> On Tue, Oct 20, 2009 at 12:29 AM, William Stein  wrote:
>>
>> 3. One could probably replace the "expand to 2GB" vmdk by one that
>> expands larger.
>
> Yes, exactly what I meant.
>
>>
>>> Using a Windows file system shared folder does not work because it
>>> does not support linux symbolic links. :-(
>>
>> It depends what you want to use it for, I guess.   But yes, Windows
>> doesn't have linux symlinks, which is definitely a problem.
>>
>
> I was thinking of using it to install a "writeable" version of Sage by
> copying from /sage.  This does not work because the Sage distribution
> requires symlinks.  Otherwise it works fine to access files and save
> them in a manner accessible to windows.
>
>>>
>>> In the meantime I created a new .vdi disk as workspace and I am
>>> playing with that. After copying /sage over to this disk I tried
>>> './sage -upgrade but something here is still read-only it seems...
>>>
>>> # ./sage -upgrade
>>
>> Look at SAGE_ROOT in the file "sage" -- you're still running the old
>> version from the read only file system.   Just edit SAGE_ROOT to your
>> new location.
>>
>
> Ah yes I forgot that.  Now it works ... up to this point:
>
> ./install: line 356: make: command not found

Yes, not having any compilers will cause trouble!

>
> real    0m0.008s
> user    0m0.000s
> sys     0m0.004s
> Error building Sage.
> Error installing Sage!
> #
>
> So now I must look for the development tools but these are apparently
> packaged differently as sfs files.  How much more would these add to
> the size of the Sage image?

Exactly 90MB.  To install them yourself get this file:


 
http://distro.ibiblio.org/pub/linux/distributions/puppylinux/puppy-4.3.1/devx_431.sfs

and stick it in  /mnt/home/, then restart.  Suddenly GCC, make, etc.,
should magically appear.  I think that's all you have to do.

By the way, sfs = "squash file system", which in puppy linux is a
compressed filesystem that gets layered on top of another filesystem
using unionfs.   Very cool.

> BTW, how do I kill that annoying "sage-ensure_notify" pop-up message?
> I deleted the crontab entry but its still live.

Look in /etc/rc.d/rc.local and you'll see -- just delete a line.   I
didn't actually end up using the crontab entry, since cron isn't even
running.


>> That is correct.  I hope somebody creates a completely separate
>> virtual machine aimed at what you want.
>
> I think an upgradable and package installable version of Sage for
> Windows+VirtualBox would be desirable.  Among other things this
> requires that the Sage shell include gcc and other tools required for
> installing spkgs.

Yep.  I think it can be done for 90MB more for compilers plus a
unionfs mounted partition, hopefully.

Note that upgrading right now might be possible by simply swapping out
the virtual disk that contains the data that is mounted as /sage/

>
>>  Maybe you just want to run a full Linux under Windows + VirtualBox
>> though, for which there are already oodles and oodles of existing options?
>>
>
> No, that's not very interesting unless for some reason you have no
> alternative. If you need windows then I think it is better to run
> Windows on VirtualBox under Linux. I do that whenever I can these days
> but unenlightened corporate IT managers sometimes do not allow it. :-(

Thanks a lot for taking interest in this VirtualBox Sage appliance
project!  It's really been great getting your feedback.  And I would
obviously be happy to help with making a version of all this that
includes fricas...

 -- 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Categories restart: call for reviewers

[Robert Bradshaw]

On Oct 13, 2009, at 12:34 PM, Nicolas M. Thiery wrote:

>
>   Hi Robert, Craig,
>
> Any chances for you to review shortly:
>
> http://combinat.sagemath.org/patches/file/tip/categories-fixsagelib-nt.patch

Looks fine to me. Seems to be mostly renaming stuff and plumbing  
category definitions around.

- Robert


--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[Bill Page]

On Tue, Oct 20, 2009 at 12:29 AM, William Stein  wrote:
>
> 3. One could probably replace the "expand to 2GB" vmdk by one that
> expands larger.

Yes, exactly what I meant.

>
>> Using a Windows file system shared folder does not work because it
>> does not support linux symbolic links. :-(
>
> It depends what you want to use it for, I guess.   But yes, Windows
> doesn't have linux symlinks, which is definitely a problem.
>

I was thinking of using it to install a "writeable" version of Sage by
copying from /sage.  This does not work because the Sage distribution
requires symlinks.  Otherwise it works fine to access files and save
them in a manner accessible to windows.

>>
>> In the meantime I created a new .vdi disk as workspace and I am
>> playing with that. After copying /sage over to this disk I tried
>> './sage -upgrade but something here is still read-only it seems...
>>
>> # ./sage -upgrade
>
> Look at SAGE_ROOT in the file "sage" -- you're still running the old
> version from the read only file system.   Just edit SAGE_ROOT to your
> new location.
>

Ah yes I forgot that.  Now it works ... up to this point:

./install: line 356: make: command not found

real0m0.008s
user0m0.000s
sys 0m0.004s
Error building Sage.
Error installing Sage!
#

So now I must look for the development tools but these are apparently
packaged differently as sfs files.  How much more would these add to
the size of the Sage image?

BTW, how do I kill that annoying "sage-ensure_notify" pop-up message?
I deleted the crontab entry but its still live.

>
> That is correct.  I hope somebody creates a completely separate
> virtual machine aimed at what you want.

I think an upgradable and package installable version of Sage for
Windows+VirtualBox would be desirable.  Among other things this
requires that the Sage shell include gcc and other tools required for
installing spkgs.

>  Maybe you just want to run a full Linux under Windows + VirtualBox
> though, for which there are already oodles and oodles of existing options?
>

No, that's not very interesting unless for some reason you have no
alternative. If you need windows then I think it is better to run
Windows on VirtualBox under Linux. I do that whenever I can these days
but unenlightened corporate IT managers sometimes do not allow it. :-(

Regards,
Bill Page.

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 4.2.alpha0 released

[John H Palmieri]

On Oct 19, 8:46 am, John H Palmieri  wrote:
> On Oct 19, 12:10 am, Mike Hansen  wrote:
>
> > Hello all,
>
> > Sage 4.2.alpha0 is out.  Source and binary areavailable at
>
> When installing on Mac OS X 10.5, either 32 or 64 bit, I see this at
> the very beginning:
>
>   ./install: line 131: PIL: command not found
>
> There are two problems with line 131: it says
>
>   PIL =`$newest pyprocessing`
>
> There shouldn't be a space before the = sign, and I assume that
> "pyprocessing" should actually say "pil".

Patch up (and a positive review) at #7252.

  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-unsubscr...@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 4.2.alpha0 released

[Dan Drake]

I wrote:
> Looking in the 4.2.alpha0 tarball, I don't see that this actually got
> merged. The makefile is the same as 4.1.2.

That's totally wrong. I was looking in the wrong directory. The ticket
was correctly merged.

Move along, nothing to see here...

Dan

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


signature.asc
Description: Digital signature

[sage-devel] Re: Sage 4.2.alpha0 released

[Dan Drake]

All long doctests pass on 64-bit Ubuntu 9.04.

On Mon, 19 Oct 2009 at 02:10PM +0700, Mike Hansen wrote:
> #7089: Dan Drake: refactor SAGE_ROOT/makefile [Reviewed by Tim Dumol]

Looking in the 4.2.alpha0 tarball, I don't see that this actually got
merged. The makefile is the same as 4.1.2.

Dan

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


signature.asc
Description: Digital signature

[sage-devel] Re: Problem executing cell contents in version 4.1.2

[William Stein]

On Mon, Oct 19, 2009 at 11:12 AM, lutusp  wrote:
>
> On Oct 19, 10:19 am, William Stein  wrote:
>> On Mon, Oct 19, 2009 at 9:30 AM, lutusp  wrote:
>>
>> > Hi all. Just downloaded 4.1.2 for Fedora 11. When I start the notebook
>> > server, everything is fine -- the worksheets get converted and placed
>> > in the new directory sage_notebook.sagenb, then I can load a
>> > worksheet. But when I try to execute a cell, I get this error message:
>>
>> > NameError: name 'sagenb' is not defined
>>
>> That error indicates that Sage is running worksheet processes
>>  using a version of Sage prior to 4.1.2.
>>
>> What precisely is the command you use to start your notebook server?
>
> 1. Installed Fedora 11 binary in my sage directory -- this creates a
> subdirectory named "sage-4.1.2-linux-Fedora_release_11_Leonidas-i686-
> Linux"
>
> 2. Renamed two prior Sage directories by prefixing "xxx_" to their
> names (to keep them from being found but to avoid having to reinstall
> them).
>
> 3. Entered the new Fedora 11 Sage 4.1.2  directory.
>
> 4. Entered "./sage -notebook"
>
> 5. Sage offers to convert the prior notebook formats, completes, runs
> server, launches Firefox.
>
> 6. Log onto Sage server -- note  "Version 4.1.2" at upper left -- open
> worksheet.
>
> 7. Try to execute a cell's contents. Message:
>
>         "Traceback (click to the left for traceback)
>         ...
>         NameError: name 'sagenb' is not defined"
>
>         (no traceback contents)
>
> 8. Create a new notebook.
>
>  (Now it gets complicated)
>
> 9. Copy example old worksheet cell contents into new worksheet. Up to
> a certain level of complexity, copied cell contents will work in a
> newly created worksheet but continue to fail if the old worksheet is
> loaded and executed. I emphasize when I say "old worksheet" I am
> talking about worksheets converted to the new format and placed in the
> new directory "sage_notebook.sagenb", and I just took the precaution
> of removing the old directory "sage_notebook" entirely to avoid any
> possible confusion.
>
> I emphasize that the old worksheet fails to execute cell contents, but
> the same cell copied to a new worksheet succeeds ... but only up to a
> certain level of complexity, after which the new and old scenarios
> (e.g. execution of old worksheet cells versus old cell content copied
> into new worksheet) become the same.
>
>> Do you have any old versions of Sage installed anywhere?
>
> Yes, two. But I have renamed the directories with a prefixed "xxx_" to
> prevent their being located. One note -- if any Sage software is
> locating resources using inodes instead of paths, this would prevent
> the above remedy from having the expected effect.
>
> If you want an example of a complex legacy worksheet, there is a
> resource .sws file attached tho the latest page of my Sage tutorial
> located at:
>
> http://arachnoid.com/sage/polynomial.html
>
> The .sws URL:
>
> http://arachnoid.com/sage/sage/polynomial_regression.sws
>
> 10. I tried compiling and installing the source (which was
> successful), but with the same outcome.
>
> Please ask any questions I have failed to proactively answer.
>

Despite your extensive feedback, I have no ideas.   Could you tar up
and post your sage_notebook.sagenb directory and just let me try it
out, assuming it doesn't have anything too personal, and just let me
try it out?

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-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 9:29 PM, William Stein  wrote:
> On Mon, Oct 19, 2009 at 7:40 PM, Bill Page  wrote:
>>
>> William,
>>
>> On Mon, Oct 19, 2009 at 7:54 PM, you wrote:
>>> Bill Page wrote:
 I think it is also useful to increase video memory to the minimum 7
 Mb. required for seamless mode. I tried this since I usually prefer to
 run in this mode where the linux system seems to share the same
 desktop as windows. But I couldn't quite get it to work. Maybe it is a
 limitation of the window manager?
>>>
>>> I don't know if it will work ever.   Regarding video memory, no matter
>>> what i set when I export the machine, it always resets the video
>>> memory to 4MB on import.
>>>
>>
>> Well, after setting the display memory to 7 Mb. it almost works. The
>> only problem is that I can't seem to find the Menu bar.  Usually it
>> would sit just above the Windows start bar (or at the top of the
>> screen if the linux default is to show the Menu at the top).  Any
>> windows that are open however remain on the windows desktop.  So I
>> supposed that this might be a limitation of jwm.
>>
 I found 'Menu/Utilities/Resize Personal File' very useful because I
 really wanted to try copying /sage into a writable area so I could
 install optional packages like fricas-aldor.

By the way, in theory a better way to make /sage "changeable" without using
*any* extra diskspace is to create a new vdi disk, format it, then use
something called "unionfs" to combine the /sage mount with that new
formatted partition.  This will make it so /sage still only takes
380MB, but the changes get stored uncompressed and take precedence
over the read-only content of /sage.   I haven't actually done this
yet -- just read about it.   But it is one of the key basic ideas on
which the whole design of live cd's, puppy linux, etc., is based

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-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 7:40 PM, Bill Page  wrote:
>
> William,
>
> On Mon, Oct 19, 2009 at 7:54 PM, you wrote:
>> Bill Page wrote:
>>> I think it is also useful to increase video memory to the minimum 7
>>> Mb. required for seamless mode. I tried this since I usually prefer to
>>> run in this mode where the linux system seems to share the same
>>> desktop as windows. But I couldn't quite get it to work. Maybe it is a
>>> limitation of the window manager?
>>
>> I don't know if it will work ever.   Regarding video memory, no matter
>> what i set when I export the machine, it always resets the video
>> memory to 4MB on import.
>>
>
> Well, after setting the display memory to 7 Mb. it almost works. The
> only problem is that I can't seem to find the Menu bar.  Usually it
> would sit just above the Windows start bar (or at the top of the
> screen if the linux default is to show the Menu at the top).  Any
> windows that are open however remain on the windows desktop.  So I
> supposed that this might be a limitation of jwm.
>
>>> I found 'Menu/Utilities/Resize Personal File' very useful because I
>>> really wanted to try copying /sage into a writable area so I could
>>> install optional packages like fricas-aldor.
>>
>> Wow, I didn't even know about that!  Did this work.
>>
>
> Unfortunately no because the vmdk disk has a maximum fixed size of
> only 2 Gb.  It might be better and probably not take much more space
> in the compressed file if this disk were at least twice as large.
> Also you might consider using expandable disks like .vdi instead of
> .vmdk. But I am not sure of the pros and cons of this choice.

1.  vmdk disks are expanding disks.Otherwise the vmdk file would have
been 2GB to start with instead of about 300MB.

2.  when you use VirtualBox to export a "VirtualBox Appliance" then
all the disks are always exported vmdk.  The user has no choice in the
matter, as far as I can tell.  This is I think part of the "open
virtual machine format".   This might be because Vmdk is a much more
"industry standard" format than vdi.

3. One could probably replace the "expand to 2GB" vmdk by one that
expands larger.  In the meantime it's good that you simply added
another virtual disk that is whatever size you want.

> Using a Windows file system shared folder does not work because it
> does not support linux symbolic links. :-(

It depends what you want to use it for, I guess.   But yes, Windows
doesn't have linux symlinks, which is definitely a problem.

>
> In the meantime I created a new .vdi disk as workspace and I am
> playing with that. After copying /sage over to this disk I tried
> './sage -upgrade but something here is still read-only it seems...
>
> # ./sage -upgrade

Look at SAGE_ROOT in the file "sage" -- you're still running the old
version from the read only file system.   Just edit SAGE_ROOT to your
new location.

> Testing mirrors...
> [2 ] modular.fas.harvard.edu          145 [ms]
> [0 ] mirror.switch.ch                 358 [ms]
> [1 ] mirror.yandex.ru                 690 [ms]
> [3 ] sage.scipy.org                   680 [ms]
> automatically selected server modular.fas.harvard.edu
> (http://modular.fas.harvard.edu/sage/)
> Downloading packages from http://modular.fas.harvard.edu/sage//spkg
> http://modular.fas.harvard.edu/sage//spkg/standard
> Reading package lists...
> Traceback (most recent call last):
>  File "/sage/local/bin/sage-update", line 315, in 
>    do_update()
>  File "/sage/local/bin/sage-update", line 263, in do_update
>    packages = spkg_list('standard')
>  File "/sage/local/bin/sage-update", line 229, in spkg_list
>    urllib.urlretrieve(web_url, file, reporthook)
>  File "/sage/local/lib/python/urllib.py", line 94, in urlretrieve
>    return _urlopener.retrieve(url, filename, reporthook, data)
>  File "/sage/local/lib/python/urllib.py", line 239, in retrieve
>    tfp = open(filename, 'wb')
> IOError: [Errno 30] Read-only file system: 'list.tmp'
> Error getting new packages!
>
> ---
>
> This is just playing and I didn't really expect it to work. And of
> course none of this is the kind of thing for which this virtual
> machine was really intended.

That is correct.  I hope somebody creates a completely separate
virtual machine aimed at what you want.Maybe you just want to run
a full Linux under Windows + VirtualBox though, for which there are
already oodles and oodles of existing options?

> Like you said earlier perhaps there is a
> need for somewhat larger version of this sort of virtual machine that
> includes enough to actually do some development. But just how big is
> too big?  Is it really worth the effort?
>
> Puppy seems pretty solid but kind of in a "world of it's own" when it
> comes to installing packages etc. I guess the main advantage of this
> puppy is it's extremely small size for a linux distribution.

That is indeed its advantage.  It's installed size is literally *less
than* 10% the size of a minimal Ubuntu install!It is also less
resource intensiv

[sage-devel] Re: Sage in VirtualBox

[Bill Page]

William,

On Mon, Oct 19, 2009 at 7:54 PM, you wrote:
> Bill Page wrote:
>> I think it is also useful to increase video memory to the minimum 7
>> Mb. required for seamless mode. I tried this since I usually prefer to
>> run in this mode where the linux system seems to share the same
>> desktop as windows. But I couldn't quite get it to work. Maybe it is a
>> limitation of the window manager?
>
> I don't know if it will work ever.   Regarding video memory, no matter
> what i set when I export the machine, it always resets the video
> memory to 4MB on import.
>

Well, after setting the display memory to 7 Mb. it almost works. The
only problem is that I can't seem to find the Menu bar.  Usually it
would sit just above the Windows start bar (or at the top of the
screen if the linux default is to show the Menu at the top).  Any
windows that are open however remain on the windows desktop.  So I
supposed that this might be a limitation of jwm.

>> I found 'Menu/Utilities/Resize Personal File' very useful because I
>> really wanted to try copying /sage into a writable area so I could
>> install optional packages like fricas-aldor.
>
> Wow, I didn't even know about that!  Did this work.
>

Unfortunately no because the vmdk disk has a maximum fixed size of
only 2 Gb.  It might be better and probably not take much more space
in the compressed file if this disk were at least twice as large.
Also you might consider using expandable disks like .vdi instead of
.vmdk. But I am not sure of the pros and cons of this choice.

Using a Windows file system shared folder does not work because it
does not support linux symbolic links. :-(

In the meantime I created a new .vdi disk as workspace and I am
playing with that. After copying /sage over to this disk I tried
'./sage -upgrade but something here is still read-only it seems...

# ./sage -upgrade
Testing mirrors...
[2 ] modular.fas.harvard.edu  145 [ms]
[0 ] mirror.switch.ch 358 [ms]
[1 ] mirror.yandex.ru 690 [ms]
[3 ] sage.scipy.org   680 [ms]
automatically selected server modular.fas.harvard.edu
(http://modular.fas.harvard.edu/sage/)
Downloading packages from http://modular.fas.harvard.edu/sage//spkg
http://modular.fas.harvard.edu/sage//spkg/standard
Reading package lists...
Traceback (most recent call last):
  File "/sage/local/bin/sage-update", line 315, in 
do_update()
  File "/sage/local/bin/sage-update", line 263, in do_update
packages = spkg_list('standard')
  File "/sage/local/bin/sage-update", line 229, in spkg_list
urllib.urlretrieve(web_url, file, reporthook)
  File "/sage/local/lib/python/urllib.py", line 94, in urlretrieve
return _urlopener.retrieve(url, filename, reporthook, data)
  File "/sage/local/lib/python/urllib.py", line 239, in retrieve
tfp = open(filename, 'wb')
IOError: [Errno 30] Read-only file system: 'list.tmp'
Error getting new packages!

---

This is just playing and I didn't really expect it to work. And of
course none of this is the kind of thing for which this virtual
machine was really intended. Like you said earlier perhaps there is a
need for somewhat larger version of this sort of virtual machine that
includes enough to actually do some development. But just how big is
too big?  Is it really worth the effort?

Puppy seems pretty solid but kind of in a "world of it's own" when it
comes to installing packages etc. I guess the main advantage of this
puppy is it's extremely small size for a linux distribution.

Regards,
Bill Page.

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 env -- RM="rm"?

[William Stein]

On Mon, Oct 19, 2009 at 2:30 PM, Tim Dumol  wrote:
>
> The same problem happened with GLPK's spkg, to note.

I'm personally OK with the change you suggest.  Post a patch (to the
repo in local/bin/) and I'll referee it.

William

>
> On Oct 20, 5:26 am, Tim Dumol  wrote:
>> Hey,
>>
>> I am currently trying to build a `jpeg-7` spkg on Arch Linux x86_64,
>> but I've had problems with $RM
>>
>> {{{
>> ...
>> checking if gcc -std=gnu99 supports -c -o file.o... rm: cannot remove
>> `conftest*': No such file or directory
>> ...
>>
>> }}}
>>
>> etc.
>>
>> Full log here:http://sage.pastebin.com/m94e6767
>>
>> Doing:
>>
>> {{{
>> RM="rm -f" && export RM
>>
>> }}}
>>
>> fixed the problem for me.
>>
>> $SAGE_LOCAL$/bin/sage-env sets $RM to "rm" (L245), but jpeg-7's
>> configure sets $RM to "rm -f" (L7075), if $RM is not yet set. It also
>> sets $CP and $MV to "cp -f" and "mv -f", respectively.
>>
>> I wonder if `sage-env `should also set $RM to "rm -f". Thoughts?
> >
>



-- 
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-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 6:27 PM, kcrisman  wrote:
>
>
> Works fine on OSX 10.5 Intel.  Only thing is that there are two
> messages that pop up telling me where to open the browser, instead of
> just one, when I first start the VM, and in any case they seem
> redundant if there is already an alias on the desktop to Sea Monkey.
>

The idea is that you can open Safari or Firefox in OS X to that address.

 --- Wililam

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[kcrisman]

Works fine on OSX 10.5 Intel.  Only thing is that there are two
messages that pop up telling me where to open the browser, instead of
just one, when I first start the VM, and in any case they seem
redundant if there is already an alias on the desktop to Sea Monkey.

- kcrisman
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 4.2.alpha0 released

[Kiran Kedlaya]

All long doctests pass under 64-bit Fedora 10.

Kiran

On Oct 19, 3:10 am, Mike Hansen  wrote:
> Hello all,
>
> Sage 4.2.alpha0 is out.  Source and binary areavailable at
>
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-...http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0sage-4...
>
> The upgrade path is
>
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-...
>
> The tickets merged in 4.2.alpha0 are:
>
> #5556: Karl-Dieter Crisman: symbolic gamma function is incoherent
> [Reviewed by Mike Hansen]
> #5985: Craig Citro, Nicolas M. Thiéry: cPickle: adds support for class
> pickling customization [Reviewed by Craig Citro, Nicolas M. Thiéry]
> #6482: Martin Albrecht: multivariate polynomial substitution has a
> design flaw [Reviewed by Mike Hansen]
> #1509: Matthew Trumbell: Allow text rotation in plotting [Reviewed by
> Karl-Dieter Crisman]
> #2341: Karl-Dieter Crisman, Ricardo Massaro: vector subs over symbolic
> does not work [Reviewed by Jason Grout]
> #2900: Karl-Dieter Crisman: matplotlib bug in imshow (probably fixed
> in new version) [Reviewed by Jason Grout]
> #3718: Karl-Dieter Crisman: calculus -- sage treats limits involving
> floor and ceil completely wrong [Reviewed by Tim Dumol]
> #4807: Michael Yurko: bug in exponential integral [Reviewed by Mike Hansen]
> #4941: Yann Laigle-Chapuy: pari list slicing is extremely slow -- make
> it much faster [Reviewed by Mike Hansen]
> #5755: Karl-Dieter Crisman: error converting symbolic expression to
> polynomial [Reviewed by Mike Hansen]
> #5899: Tim Abbott: Update Debian build support for Sage spkg [Reviewed
> by Mike Hansen]
> #5986: Craig Citro, Nicolas M. Thiéry: Workaround mishandled nested
> classes in Python and cPickle [Reviewed by Robert Bradshaw]
> #5991: Nicolas M. Thiéry: Add a standard constructor for dynamic
> classes [Reviewed by David Roe]
> #6118: Robert Bradshaw, Craig Citro: integer shifting slow [Reviewed
> by Mike Hansen, Craig Citro]
> #6187: John Palmieri, Mitesh Patel: Fix post-cloning docbuild problems
> and add new command-line options [Reviewed by John Cremona, Mike
> Hansen]
> #6189: Karl-Dieter Crisman: 'integrate' produces incorrect answer
> [Reviewed by Mike Hansen]
> #6228: Karl-Dieter Crisman: Wrong multiplicities when solving a
> univariate polynomial equation [Reviewed by Mike Hansen]
> #6304: David Harvey: intermittent crash in bernmm (4.0.2.rc0)
> [Reviewed by Mike Hansen]
> #6420: Karl-Dieter Crisman: Maxima integration error with 1/x^3
> [Reviewed by Mike Hansen]
> #6572: John Palmieri: tutorial: put double colon on its own line
> [Reviewed by Mike Hansen]
> #6586: John Palmieri: update Sphinx to version 0.6.2 or 0.6.3
> [Reviewed by John Cremona, Tim Dumol, Mitesh Patel]
> #6605: John Palmieri: sage -docbuild DOC FORMAT should do better error
> checking on DOC [Reviewed by Mitesh Patel]
> #6642: Karl-Dieter Crisman: problem with solve (from the tutorial)
> [Reviewed by John Palmieri]
> #6823: Nathann Cohen: Kneser Graph in graph_generators [Reviewed by Rob 
> Beezer]
> #6892: John Palmieri: change dollar signs to backticks [Reviewed by Mike 
> Hansen]
> #6899: Mike Hansen, Burcin Erocal: weird bug taking float of real part
> of a symbolic [Reviewed by Burcin Erocal, Karl-Dieter Crisman]
> #6906: John Palmieri: update Mercurial to version 1.3.1 [Reviewed by
> Mike Hansen]
> #6909: Martin Albrecht: move some commutative algebra to libSingular
> functions interface [Reviewed by Michael Brickenstein, Burcin Erocal]
> #6958: Robert Miller: prove_BSD function for elliptic curves over QQ
> [Reviewed by William Stein]
> #6977: Nicolas M. Thiéry: Implement __len__ to add support for MuPAD
> objects as Python containers [Reviewed by Mike Hansen]
> #6986: Jason Grout: Make PIL standard [Reviewed by Mike Hansen]
> #7007: Mike Hansen: variables() method for polynomial rings returns
> the constant as well as the variable. [Reviewed by Jason Grout]
> #7010: Minh Van Nguyen: sanity check key value of the shift
> cryptosystem [Reviewed by Rob Beezer]
> #7015: William Stein, Minh Van Nguyen: cygwin port -- ratpoints --
> don't build executable since we don't need it (and fails on cygwin)
> [Reviewed by Mike Hansen]
> #7045: David Kirkby: update libgcrypt to the latest version 1.4.4
> [Reviewed by Adam Webb]
> #7081: Tim Dumol: sage -c "..." runs from the local/bin/ directory.
> ugh [Reviewed by William Stein]
> #7084: Karl-Dieter Crisman: Make assumption comparison work for
> GenericDeclarations [Reviewed by Tim Dumol]
> #7117: John Palmieri: latex fix for RIF [Reviewed by Mike Hansen]
> #3970: Mike Hansen: MaximaElements should not coerce into SR.
> [Reviewed by Adam Webb]
> #5243: Florent Hivert: non decreasing parking functions (depend now on
> 5255) [Reviewed by Dan Drake, Mike Hansen]
> #5812: Wilfried Huss: option zorder has no effect for point() and
> polygon() [Reviewed by Karl-Dieter Crisman, Volker Braun]
> #6828: Nathann Cohen: Random Bipartite Graph [Reviewed by Rob

[sage-devel] Re: Sage in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 4:44 PM, Bill Page  wrote:
>
> William,
>
> Here are a few first impressions:
>
> Maybe a significant point to make here is that it is not necessary to
> be running Windows in order to import this virtual machine into
> VirtualBox. A linux or OSX machine will do just as well.  There are
> times that one might want such a small, independent and lightweight
> version of Sage anywhere.

Indeed.  In fact, I just installed it on x86 Open Solaris 11 and it
worked first try.  Total time < 10 minutes.  And it's an Open Solaris
box that nobody has run Sage on before, I think. ;-)

> Although this is presumably intended to be usable right out of the
> box, it might be good to have a small tutorial pointing out some basic

Should that be in the virtual machine, in the README.txt or in both?

> functions of VirtualBox including shared folders accessible from the
> Devices menu and mounted as follows:
>
> # mkdir /mnt/shar
> # mount -t vboxsf "Folder Name" /mnt/share
>
> (It would be nicer if the file manager or media mounter supported this
> automatically.)

Good idea.  I bet we can write our own little Daemon to figure this
out, assuming that "Folder Name" can be somehow discovered via the
command line from within virtualbox.

>
> The manual mount works fine and allows read/write access to selected
> files and directories on windows.

Yes, very nice.

>
> It might also be worthwhile to install samba-client to permit mounting
> windows shares if you think space permits in the disk image. I tried
> this but kept getting connection refused. Perhaps a local windows
> firewall issue?

Probably.   That's a great idea.

> I think it is also useful to increase video memory to the minimum 7
> Mb. required for seamless mode. I tried this since I usually prefer to
> run in this mode where the linux system seems to share the same
> desktop as windows. But I couldn't quite get it to work. Maybe it is a
> limitation of the window manager?

I don't know if it will work ever.   Regarding video memory, no matter
what i set when I export the machine, it always resets the video
memory to 4MB on import.

> I found 'Menu/Utilities/Resize Personal File' very useful because I
> really wanted to try copying /sage into a writable area so I could
> install optional packages like fricas-aldor.

Wow, I didn't even know about that!  Did this work.

>
> BTW, why are there two ethernet interfaces defined?

One to connect to the outside world, and one so your desktop can
connect to the virtual machine.  If you define only one interface then
the virtual machine can connect to the outside world, but you can't
connect to the virtual machine, which is really annoying.  (VMware is
simpler in this way.)

> Finally: Well done! Thank you for your work on this.

You're welcome.We should also make a version that includes a bit
more, without getting too heavy, e.g., Fricas, GCC, etc.

Thanks for all your suggestions and testing!

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-unsubscr...@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 in VirtualBox

[Bill Page]

William,

Here are a few first impressions:

Maybe a significant point to make here is that it is not necessary to
be running Windows in order to import this virtual machine into
VirtualBox. A linux or OSX machine will do just as well.  There are
times that one might want such a small, independent and lightweight
version of Sage anywhere.

Although this is presumably intended to be usable right out of the
box, it might be good to have a small tutorial pointing out some basic
functions of VirtualBox including shared folders accessible from the
Devices menu and mounted as follows:

# mkdir /mnt/shar
# mount -t vboxsf "Folder Name" /mnt/share

(It would be nicer if the file manager or media mounter supported this
automatically.)

The manual mount works fine and allows read/write access to selected
files and directories on windows.

It might also be worthwhile to install samba-client to permit mounting
windows shares if you think space permits in the disk image. I tried
this but kept getting connection refused. Perhaps a local windows
firewall issue?

I think it is also useful to increase video memory to the minimum 7
Mb. required for seamless mode. I tried this since I usually prefer to
run in this mode where the linux system seems to share the same
desktop as windows. But I couldn't quite get it to work. Maybe it is a
limitation of the window manager?

I found 'Menu/Utilities/Resize Personal File' very useful because I
really wanted to try copying /sage into a writable area so I could
install optional packages like fricas-aldor.

BTW, why are there two ethernet interfaces defined?

Finally: Well done! Thank you for your work on this.

Regards,
Bill Page.

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

[sage-devel] Re: [codenode-devel] Re: New Features! Updated live notebook. Support for Sage!

[William Stein]

On Mon, Oct 19, 2009 at 3:45 PM, Dorian Raymer  wrote:
> Hi Stan,
> I looked in to the solve problem and I get the same error.
>
> Solve does work when I log into the server and use Sage from the command
> line, and it works from the codenode notebook on my local linux install, so
> I am slightly baffled.
>
> Does William or another sage person have any idea what would cause the
> 'Unable to start maxima' error? I'm thinking it might be some kind of
> environment problem that a sage expert would have better intuition on.

Precisely how do you start Sage?  What command line is run?

William

>
> For plotting, this has been the number one first thing people try, and I'm
> currently working on getting plots to show up in the notebook. Most of the
> code supporting this already exists, it has just been a matter of
> updating/adapting it to work with the new backend/frontend, so it won't be
> long.
>
> -Dorian
>
>
> On Mon, Oct 19, 2009 at 4:39 AM, Stan Schymanski  wrote:
>>
>> Hi Dorian,
>>
>> I got all excited but the live public notebook in sage mode does not
>> seem to work for me.
>>
>> plot(x^3 - 3*x^2 + 2*x -3, (x,-3,3))
>>
>> produces an empty field for me, and
>>
>> solve(0 == x^3 - a*x^2 + b*x -3, x)
>>
>> results in a traceback:
>>
>> Traceback (most recent call last):
>> [snip]
>> TypeError: Unable to start maxima
>>
>> Is there a problem with the sage installation?
>>
>> Cheers,
>> Stan
>>
>> Dorian Raymer wrote:
>> > Hello everyone,
>> > We have merged a big development effort into the master repository
>> > !
>> >
>> > This represents a milestone in the ongoing effort to bring the whole
>> > design to a new stable point. The Frontend was re-implemented using
>> > Django earlier this year, and we now have completely re-done the
>> > Backend!
>> >
>> > High level features:
>> > - Engine plug-ins
>> >   An engine is the computation process behind the notebook (Before
>> > today, Codenode only supported the regular Python engine).
>> >   Engine types can be added without modifying library code!
>> > - Sage Engine
>> >   Template engine type for using Sage in codenode!
>> > - Admin management of Backends
>> >   A frontend can be associated with any number of backend servers,
>> > running anywhere on the network. Each of those backend servers can
>> > support any number of engine types.
>> > - Development version of servers
>> >   Improved work flow; development of running code. Raw support for
>> > reloading of modules with out restarting the process.
>> >
>> > Updated documentation . These still need
>> > work, so please make as many comments and ask as many questions as you
>> > have. This will aid improving the quality of the documentation greatly.
>> >
>> > Updated live public notebook , with two
>> > engine types: Python and Sage!!! This backend server is also much more
>> > powerful than what we had running on app engine. This makes the live
>> > site an actually useful service, so please use it for day to day
>> > python/sage things as much as you like.
>> >
>> >
>> > Making it easy to develop codenode is an important design feature we
>> > are striving to fulfill. The major pieces of the system are becoming
>> > sufficiently modularized such that working on a specific component
>> > does not require deep knowledge of the entire system. We hope this
>> > will allow people with specific specializations to have a fun time
>> > developing features either independently, or with some counterpart
>> > collaborator who might have a complementary expertise.
>> >
>> > As a project, codenode is finding a balance between providing a solid
>> > base design for long term viability/extendability, and writing useful
>> > features that people care about and need NOW! With this milestone, the
>> > focus of our efforts will noticeably shift back to fulling feature
>> > requests, fixing bugs, and sharing/proposing our own feature ideas.
>> >
>> > A few quick mentions of features:
>> > General plotting capability for plain Python that is compatible with
>> > Sage.
>> > Sage version of notebook style and key bindings.
>> > Collaborative running notebooks, portable notebook formats, etc.
>> > Uploading code and data for use in a notebook.
>> > Engines for other languages.
>> >
>> > Looking forward to feedback!
>> >
>> > Thanks,
>> > Dorian and Alex
>> >
>> >
>> > >
>>
>>
>>
>
>
> >
>



-- 
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-unsubscr...@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 releases

[Nicolas M. Thiery]

On Wed, Oct 14, 2009 at 10:49:18AM +0200, Burcin Erocal wrote:
> > I wonder if the "lieutenant" model used by Linux kernel development
> > might be helpful here?  If there was one or two people (lieutenants)
> > responsible for each broad area of Sage, and trusted to merge patches,

Somehow, that's precisely the point of the Sage-Combinat community for
combinatorics. By going through our patch server, our patches are in
effect continuously prereviewed:

 - Integration with the other patches being developed on the same topic
 - Usage and feedback by a variety of people
 - Usage on different platforms

My impression so far is that this has worked quite well, and improved
the quality of the patches we submitted to trac (reducing the risks of
them actually causing trouble). On the other hand:

 - There is some technical overhead (hopefully that will become easier
   at some point, maybe by switching to pbranches rather than queues)

 - With the category stuff, our patch queue got close to become non
   manageable. But I don't foresee further developments at that scale
   (both in sheer code size, life cycle, amount of reviewers, number
   of tickets, depth and numbers of areas where Sage was changed, ...)

 - We definitely should have an automatic testbot running every day
   all Sage tests with our patches applied on a bunch of platforms.

Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" 
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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: [codenode-devel] Re: New Features! Updated live notebook. Support for Sage!

[Alex Clemesha]

On Mon, Oct 19, 2009 at 3:45 PM, Dorian Raymer  wrote:
> Hi Stan,
> I looked in to the solve problem and I get the same error.
>
> Solve does work when I log into the server and use Sage from the command
> line, and it works from the codenode notebook on my local linux install, so
> I am slightly baffled.
>
> Does William or another sage person have any idea what would cause the
> 'Unable to start maxima' error? I'm thinking it might be some kind of
> environment problem that a sage expert would have better intuition on.
>
> For plotting, this has been the number one first thing people try, and I'm
> currently working on getting plots to show up in the notebook. Most of the
> code supporting this already exists, it has just been a matter of
> updating/adapting it to work with the new backend/frontend, so it won't be
> long.
Yeah, plotting and data management are incredibly important to actual
real-person-usability.  I'm very excited about the fact that our architecture is
now feeling very solid and we can (finally) shift our main focus to usability.

We already have most of the plotting code in place, and I have a branch
going with some good data management functionality, so these features should
come along pretty quickly.


-Alex




>
> -Dorian
>
>
> On Mon, Oct 19, 2009 at 4:39 AM, Stan Schymanski  wrote:
>>
>> Hi Dorian,
>>
>> I got all excited but the live public notebook in sage mode does not
>> seem to work for me.
>>
>> plot(x^3 - 3*x^2 + 2*x -3, (x,-3,3))
>>
>> produces an empty field for me, and
>>
>> solve(0 == x^3 - a*x^2 + b*x -3, x)
>>
>> results in a traceback:
>>
>> Traceback (most recent call last):
>> [snip]
>> TypeError: Unable to start maxima
>>
>> Is there a problem with the sage installation?
>>
>> Cheers,
>> Stan
>>
>> Dorian Raymer wrote:
>> > Hello everyone,
>> > We have merged a big development effort into the master repository
>> > !
>> >
>> > This represents a milestone in the ongoing effort to bring the whole
>> > design to a new stable point. The Frontend was re-implemented using
>> > Django earlier this year, and we now have completely re-done the
>> > Backend!
>> >
>> > High level features:
>> > - Engine plug-ins
>> >   An engine is the computation process behind the notebook (Before
>> > today, Codenode only supported the regular Python engine).
>> >   Engine types can be added without modifying library code!
>> > - Sage Engine
>> >   Template engine type for using Sage in codenode!
>> > - Admin management of Backends
>> >   A frontend can be associated with any number of backend servers,
>> > running anywhere on the network. Each of those backend servers can
>> > support any number of engine types.
>> > - Development version of servers
>> >   Improved work flow; development of running code. Raw support for
>> > reloading of modules with out restarting the process.
>> >
>> > Updated documentation . These still need
>> > work, so please make as many comments and ask as many questions as you
>> > have. This will aid improving the quality of the documentation greatly.
>> >
>> > Updated live public notebook , with two
>> > engine types: Python and Sage!!! This backend server is also much more
>> > powerful than what we had running on app engine. This makes the live
>> > site an actually useful service, so please use it for day to day
>> > python/sage things as much as you like.
>> >
>> >
>> > Making it easy to develop codenode is an important design feature we
>> > are striving to fulfill. The major pieces of the system are becoming
>> > sufficiently modularized such that working on a specific component
>> > does not require deep knowledge of the entire system. We hope this
>> > will allow people with specific specializations to have a fun time
>> > developing features either independently, or with some counterpart
>> > collaborator who might have a complementary expertise.
>> >
>> > As a project, codenode is finding a balance between providing a solid
>> > base design for long term viability/extendability, and writing useful
>> > features that people care about and need NOW! With this milestone, the
>> > focus of our efforts will noticeably shift back to fulling feature
>> > requests, fixing bugs, and sharing/proposing our own feature ideas.
>> >
>> > A few quick mentions of features:
>> > General plotting capability for plain Python that is compatible with
>> > Sage.
>> > Sage version of notebook style and key bindings.
>> > Collaborative running notebooks, portable notebook formats, etc.
>> > Uploading code and data for use in a notebook.
>> > Engines for other languages.
>> >
>> > Looking forward to feedback!
>> >
>> > Thanks,
>> > Dorian and Alex
>> >
>> >
>> > >
>>
>>
>>
>
>
> >
>



-- 
Alex Clemesha
clemesha.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] Re: [codenode-devel] Re: New Features! Updated live notebook. Support for Sage!

[Dorian Raymer]

Hi Stan,
I looked in to the solve problem and I get the same error.

Solve does work when I log into the server and use Sage from the command
line, and it works from the codenode notebook on my local linux install, so
I am slightly baffled.

Does William or another sage person have any idea what would cause the
'Unable to start maxima' error? I'm thinking it might be some kind of
environment problem that a sage expert would have better intuition on.

For plotting, this has been the number one first thing people try, and I'm
currently working on getting plots to show up in the notebook. Most of the
code supporting this already exists, it has just been a matter of
updating/adapting it to work with the new backend/frontend, so it won't be
long.

-Dorian


On Mon, Oct 19, 2009 at 4:39 AM, Stan Schymanski  wrote:

>
> Hi Dorian,
>
> I got all excited but the live public notebook in sage mode does not
> seem to work for me.
>
> plot(x^3 - 3*x^2 + 2*x -3, (x,-3,3))
>
> produces an empty field for me, and
>
> solve(0 == x^3 - a*x^2 + b*x -3, x)
>
> results in a traceback:
>
> Traceback (most recent call last):
> [snip]
> TypeError: Unable to start maxima
>
> Is there a problem with the sage installation?
>
> Cheers,
> Stan
>
> Dorian Raymer wrote:
> > Hello everyone,
> > We have merged a big development effort into the master repository
> > !
> >
> > This represents a milestone in the ongoing effort to bring the whole
> > design to a new stable point. The Frontend was re-implemented using
> > Django earlier this year, and we now have completely re-done the Backend!
> >
> > High level features:
> > - Engine plug-ins
> >   An engine is the computation process behind the notebook (Before
> > today, Codenode only supported the regular Python engine).
> >   Engine types can be added without modifying library code!
> > - Sage Engine
> >   Template engine type for using Sage in codenode!
> > - Admin management of Backends
> >   A frontend can be associated with any number of backend servers,
> > running anywhere on the network. Each of those backend servers can
> > support any number of engine types.
> > - Development version of servers
> >   Improved work flow; development of running code. Raw support for
> > reloading of modules with out restarting the process.
> >
> > Updated documentation . These still need
> > work, so please make as many comments and ask as many questions as you
> > have. This will aid improving the quality of the documentation greatly.
> >
> > Updated live public notebook , with two
> > engine types: Python and Sage!!! This backend server is also much more
> > powerful than what we had running on app engine. This makes the live
> > site an actually useful service, so please use it for day to day
> > python/sage things as much as you like.
> >
> >
> > Making it easy to develop codenode is an important design feature we
> > are striving to fulfill. The major pieces of the system are becoming
> > sufficiently modularized such that working on a specific component
> > does not require deep knowledge of the entire system. We hope this
> > will allow people with specific specializations to have a fun time
> > developing features either independently, or with some counterpart
> > collaborator who might have a complementary expertise.
> >
> > As a project, codenode is finding a balance between providing a solid
> > base design for long term viability/extendability, and writing useful
> > features that people care about and need NOW! With this milestone, the
> > focus of our efforts will noticeably shift back to fulling feature
> > requests, fixing bugs, and sharing/proposing our own feature ideas.
> >
> > A few quick mentions of features:
> > General plotting capability for plain Python that is compatible with
> > Sage.
> > Sage version of notebook style and key bindings.
> > Collaborative running notebooks, portable notebook formats, etc.
> > Uploading code and data for use in a notebook.
> > Engines for other languages.
> >
> > Looking forward to feedback!
> >
> > Thanks,
> > Dorian and Alex
> >
> >
> > >
>
>
> >
>

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 2:51 PM, mhampton  wrote:
>
> I think I figured it out using ifconfig.  The virtual sage session
> incorrectly detected the ip address.  I have "eth0" and a "eth1"
> addresses, and it gave me the eth1 number.  Using the eth0 number
> works.

Interesting.  That's really weird/disturbing.  What are the addresses
for each.  We do have to figure this out.   I  could make the window
report both addresses, with a preference for the eth1 address...

William

>
> -Marshall
>
> On Oct 19, 3:46 pm, mhampton  wrote:
>> Seems to work well, at least in the virtual machine.  I can't seem to
>> access it in my windows firefox though, nor can I ssh into it as the
>> popup directs.
>>
>> Speed seems pretty good, I'm impressed.  I had trouble with the vmware
>> image on the machine I am testing it on, this seems better and
>> easier.
>>
>> -Marshall
>>
>> On Oct 19, 12:32 pm, William Stein  wrote:
>>
>> > Hello,
>>
>> > Despite sage_vmware-*.zip being bar far the most popular Sage
>> > download, we will not be distributing Sage as a vmware image anymore.
>> >  Over the weekend I created a Sage distribution aimed at Windows users
>> > based on "puppy Linux" using Virtual Box:
>>
>> >      
>> > http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>>
>> > It's a 659MB download.  However, once you download and extract the
>> > zip, the resulting files still only use 659MB.  If you then import the
>> > virtual machine into VirtualBox (follow the included readme), you'll
>> > use a total of just over 700MB disk space.  This is *much* better than
>> > the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
>> > it's also much better than the 1.5GB used by an extracted Sage install
>> > on Linux or OS X.
>>
>> > What I need very much is for some people to test sage_vbox-4.1.2.zip.
>> > Please test and report back.
>>
>> >  -- William
>>
>> > --
>> > William Stein
>> > Associate Professor of Mathematics
>> > University of Washingtonhttp://wstein.org
> >
>



-- 
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-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 3:13 PM, Harald Schilly
 wrote:
>
> On Oct 19, 11:29 pm, William Stein  wrote:
>> Harald, can you post the sage_vbox here to the list of binaries for
>> windows for 4.1.2 onhttp://sagemath.org?
>
> Yes, I'll add it to the mirrors. Tell me if you change it, (or is the
> readme now included?) so that i can update it.
>

Yes.

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 4.2.alpha0 released

[Jaap Spies]

Mike Hansen wrote:
> Hello all,
> 
> Sage 4.2.alpha0 is out.  Source and binary areavailable at
> 
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0.tar
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0sage-4.2.alpha0-sage.math.washington.edu-x86_64-Linux.tar.gz
> 
> The upgrade path is
> 
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0
> 

With Fedora 11, 32 bit upgrading from 4.1.2.rc2 I get one failure:

sage -t  "devel/sage/doc/en/tutorial/tour_algebra.rst"
**
File 
"/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/en/tutorial/tour_algebra.rst",
 line 87:
 sage: find_root(cos(phi)==sin(phi),0,pi/2)
Expected:
 0.78539816339744839
Got:
 0.78539816339744828
**
1 items had failures:
1 of   4 in __main__.example_5
***Test Failed*** 1 failures.
For whitespace errors, see the file 
/home/jaap/.sage//tmp/.doctest_tour_algebra.py
 [21.1 s]


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-unsubscr...@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 in VirtualBox

[Harald Schilly]

On Oct 19, 11:29 pm, William Stein  wrote:
> Harald, can you post the sage_vbox here to the list of binaries for
> windows for 4.1.2 onhttp://sagemath.org?

Yes, I'll add it to the mirrors. Tell me if you change it, (or is the
readme now included?) so that i can update it.

H
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[mhampton]

I think I figured it out using ifconfig.  The virtual sage session
incorrectly detected the ip address.  I have "eth0" and a "eth1"
addresses, and it gave me the eth1 number.  Using the eth0 number
works.

-Marshall

On Oct 19, 3:46 pm, mhampton  wrote:
> Seems to work well, at least in the virtual machine.  I can't seem to
> access it in my windows firefox though, nor can I ssh into it as the
> popup directs.
>
> Speed seems pretty good, I'm impressed.  I had trouble with the vmware
> image on the machine I am testing it on, this seems better and
> easier.
>
> -Marshall
>
> On Oct 19, 12:32 pm, William Stein  wrote:
>
> > Hello,
>
> > Despite sage_vmware-*.zip being bar far the most popular Sage
> > download, we will not be distributing Sage as a vmware image anymore.
> >  Over the weekend I created a Sage distribution aimed at Windows users
> > based on "puppy Linux" using Virtual Box:
>
> >  
> > http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>
> > It's a 659MB download.  However, once you download and extract the
> > zip, the resulting files still only use 659MB.  If you then import the
> > virtual machine into VirtualBox (follow the included readme), you'll
> > use a total of just over 700MB disk space.  This is *much* better than
> > the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
> > it's also much better than the 1.5GB used by an extracted Sage install
> > on Linux or OS X.
>
> > What I need very much is for some people to test sage_vbox-4.1.2.zip.
> > Please test and report back.
>
> >  -- William
>
> > --
> > William Stein
> > Associate Professor of Mathematics
> > University of Washingtonhttp://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-unsubscr...@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 env -- RM="rm"?

[Tim Dumol]

The same problem happened with GLPK's spkg, to note.

On Oct 20, 5:26 am, Tim Dumol  wrote:
> Hey,
>
> I am currently trying to build a `jpeg-7` spkg on Arch Linux x86_64,
> but I've had problems with $RM
>
> {{{
> ...
> checking if gcc -std=gnu99 supports -c -o file.o... rm: cannot remove
> `conftest*': No such file or directory
> ...
>
> }}}
>
> etc.
>
> Full log here:http://sage.pastebin.com/m94e6767
>
> Doing:
>
> {{{
> RM="rm -f" && export RM
>
> }}}
>
> fixed the problem for me.
>
> $SAGE_LOCAL$/bin/sage-env sets $RM to "rm" (L245), but jpeg-7's
> configure sets $RM to "rm -f" (L7075), if $RM is not yet set. It also
> sets $CP and $MV to "cp -f" and "mv -f", respectively.
>
> I wonder if `sage-env `should also set $RM to "rm -f". Thoughts?
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[mhampton]

Seems to work well, at least in the virtual machine.  I can't seem to
access it in my windows firefox though, nor can I ssh into it as the
popup directs.

Speed seems pretty good, I'm impressed.  I had trouble with the vmware
image on the machine I am testing it on, this seems better and
easier.

-Marshall

On Oct 19, 12:32 pm, William Stein  wrote:
> Hello,
>
> Despite sage_vmware-*.zip being bar far the most popular Sage
> download, we will not be distributing Sage as a vmware image anymore.
>  Over the weekend I created a Sage distribution aimed at Windows users
> based on "puppy Linux" using Virtual Box:
>
>  http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>
> It's a 659MB download.  However, once you download and extract the
> zip, the resulting files still only use 659MB.  If you then import the
> virtual machine into VirtualBox (follow the included readme), you'll
> use a total of just over 700MB disk space.  This is *much* better than
> the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
> it's also much better than the 1.5GB used by an extracted Sage install
> on Linux or OS X.
>
> What I need very much is for some people to test sage_vbox-4.1.2.zip.
> Please test and report back.
>
>  -- William
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://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-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 12:50 PM, Harald Schilly
 wrote:
>
> On Oct 19, 9:37 pm, William Stein  wrote:
>> The files in sage_vbox-4.1.2 are already gzip compressed squashfs
>> images
>
> ok, that's good. compressing compressed data is useless and yes, speed
> is more important than a few MB + or -. More important would be a
> scerencast, virtually holding the hand of new users through the
> installation of vbox and mounting this image ;)

Harald, can you post the sage_vbox here to the list of binaries for
windows for 4.1.2 on http://sagemath.org?

http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip

 -- 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Sage env -- RM="rm"?

[Tim Dumol]

Hey,

I am currently trying to build a `jpeg-7` spkg on Arch Linux x86_64,
but I've had problems with $RM

{{{
...
checking if gcc -std=gnu99 supports -c -o file.o... rm: cannot remove
`conftest*': No such file or directory
...
}}}

etc.

Full log here: http://sage.pastebin.com/m94e6767

Doing:

{{{
RM="rm -f" && export RM
}}}

fixed the problem for me.

$SAGE_LOCAL$/bin/sage-env sets $RM to "rm" (L245), but jpeg-7's
configure sets $RM to "rm -f" (L7075), if $RM is not yet set. It also
sets $CP and $MV to "cp -f" and "mv -f", respectively.

I wonder if `sage-env `should also set $RM to "rm -f". Thoughts?
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[Harald Schilly]

On Oct 19, 9:37 pm, William Stein  wrote:
> The files in sage_vbox-4.1.2 are already gzip compressed squashfs
> images

ok, that's good. compressing compressed data is useless and yes, speed
is more important than a few MB + or -. More important would be a
scerencast, virtually holding the hand of new users through the
installation of vbox and mounting this image ;)

H
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 12:02 PM, Harald Schilly
 wrote:
>
> On Oct 19, 8:32 pm, William Stein  wrote:
>> It's a 659MB download.  However, once you download and extract the
>> zip, the resulting files still only use 659MB.
>
> Just curious, is there any compression at all?

No.  I zipped it with

   zip -0 -r sage_vbox-4.1.2.zip sage_vbox-4.1.2

>  It might be worthwhile
> testing 7z (which you can get in windows very easy via 7-zip.org) - if
> it saves >10% of the download it might be interesting to consider as
> an option.

The files in sage_vbox-4.1.2 are already gzip compressed squashfs
images, so compressing them again should be interesting...

I just tried compression level 9 (the highest I think) with 7zip, and
creating the compressed archive takes a while. In the end I get the
following size:

657MB

compared to 659MB for zip.   I don't think it is worth using 7zip for
only 2MB savings, given that users will have to wait 10 minutes (say)
to extract Sage instead of waiting only a few seconds with an
uncompressed zip.In fact, that extraction and install only takes a
few seconds is a big plus of this approach.

If we were to use 7zip, we must make it a self-extracting archive for
windows, or people will get too frustrated/confused (again).   We
could provide two versions of the same files: a standard tarball for
OSX/Linux/Solaris, and a 7zip self-extracting archive for Windows.

 -- 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-unsubscr...@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 in VirtualBox

[Bill Page]

William,

In spite of missing the readme, installation is very easy. If you
already have VirtualBox installed just use File/Import Applicance et
voila! Sage-in-five-minutes.  Puppy linux looks nice but maybe is just
a little daunting to the anyone who hasn't previously used linux on a
virtual machine under Windows.  Of course you can ignore all that and
just access Sage notebook by a local ip address from your windows
desktop.

Nice.

I'll play some more, but right now I like it.  It's reasonably fast
even on a Windows XP machine with limited ram  ( 1 Gbyte).

Thanks

Bill Page.

On Mon, Oct 19, 2009 at 2:32 PM, William Stein  wrote:
>
> Hello,
>
> Despite sage_vmware-*.zip being bar far the most popular Sage
> download, we will not be distributing Sage as a vmware image anymore.
>  Over the weekend I created a Sage distribution aimed at Windows users
> based on "puppy Linux" using Virtual Box:
>
>     http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>
> It's a 659MB download.  However, once you download and extract the
> zip, the resulting files still only use 659MB.  If you then import the
> virtual machine into VirtualBox (follow the included readme), you'll
> use a total of just over 700MB disk space.  This is *much* better than
> the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
> it's also much better than the 1.5GB used by an extracted Sage install
> on Linux or OS X.
>
> What I need very much is for some people to test sage_vbox-4.1.2.zip.
> Please test and report back.
>

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[William Stein]

On Mon, Oct 19, 2009 at 11:56 AM, Bill Page  wrote:
>
> William,
>
> Did you include the readme ? I don't see it after extraction.
>
> Regards,
> Bill Page.

I've attached it to this email, and uploaded a new version that contains it.

 -- William

>
> On Mon, Oct 19, 2009 at 2:32 PM, William Stein  wrote:
>>
>> Hello,
>>
>> Despite sage_vmware-*.zip being bar far the most popular Sage
>> download, we will not be distributing Sage as a vmware image anymore.
>>  Over the weekend I created a Sage distribution aimed at Windows users
>> based on "puppy Linux" using Virtual Box:
>>
>>     http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>>
>> It's a 659MB download.  However, once you download and extract the
>> zip, the resulting files still only use 659MB.  If you then import the
>> virtual machine into VirtualBox (follow the included readme), you'll
>> use a total of just over 700MB disk space.  This is *much* better than
>> the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
>> it's also much better than the 1.5GB used by an extracted Sage install
>> on Linux or OS X.
>>
>> What I need very much is for some people to test sage_vbox-4.1.2.zip.
>> Please test and report back.
>>
>>  -- William
>>
>>
>> --
>> William Stein
>> Associate Professor of Mathematics
>> University of Washington
>> http://wstein.org
>>
>> >
>>
>
> >
>



-- 
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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---


This is Sage (see http://www.sagemath.org) distributed as a compressed
VirtualBox appliance.  It requires 1GB of free disk space and 1GB RAM
to install and run.

INSTALLATION:
   1. Install the free open source program VirtualBox (70MB) for your
  operating system (Microsoft Windows, OS X, Linux, or x86
  Solaris):  http://www.virtualbox.org/
   2. In VirtualBox, click File --> Import Appliance, and select the
  file "sage.ovf", which is in the same directory as README.txt.
   3. Click "Next", then "Import", and wait about two minutes.
   4. From within VirtualBox, start the new virtual machine "Sage
  VirtualBox Appliance", then open your web browser in Microsoft
  Windows (or OS X, etc.) to the address that is displayed.

YOUR WORKSHEETS: 
   All of your worksheets are stored in the virtual machine in the
   file sage_root.vmdk.  To download them all to a file
   download_worksheets.zip, click the link "Download All Active" from
   the list of Sage worksheets.  The file download_worksheets.zip
   contains all of your worksheets.  You can upload that zip file into
   another copy of Sage to open *all* of your worksheets.

DEVELOPMENT:
   You cannot currently modify the copy of Sage included in this
   virtual machine, since it is stored in a read-only compressed
   filesystem, which saves a huge amount of space.

SUPPORT: 
   Join http://groups.google.com/group/sage-support or get free
   live help at http://www.sagemath.org/help-irc.html

[sage-devel] Re: Sage in VirtualBox

[Harald Schilly]

On Oct 19, 8:32 pm, William Stein  wrote:
> It's a 659MB download.  However, once you download and extract the
> zip, the resulting files still only use 659MB.

Just curious, is there any compression at all? It might be worthwhile
testing 7z (which you can get in windows very easy via 7-zip.org) - if
it saves >10% of the download it might be interesting to consider as
an option.

H
--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 in VirtualBox

[Bill Page]

William,

Did you include the readme ? I don't see it after extraction.

Regards,
Bill Page.

On Mon, Oct 19, 2009 at 2:32 PM, William Stein  wrote:
>
> Hello,
>
> Despite sage_vmware-*.zip being bar far the most popular Sage
> download, we will not be distributing Sage as a vmware image anymore.
>  Over the weekend I created a Sage distribution aimed at Windows users
> based on "puppy Linux" using Virtual Box:
>
>     http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip
>
> It's a 659MB download.  However, once you download and extract the
> zip, the resulting files still only use 659MB.  If you then import the
> virtual machine into VirtualBox (follow the included readme), you'll
> use a total of just over 700MB disk space.  This is *much* better than
> the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
> it's also much better than the 1.5GB used by an extracted Sage install
> on Linux or OS X.
>
> What I need very much is for some people to test sage_vbox-4.1.2.zip.
> Please test and report back.
>
>  -- William
>
>
> --
> 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Sage in VirtualBox

[William Stein]

Hello,

Despite sage_vmware-*.zip being bar far the most popular Sage
download, we will not be distributing Sage as a vmware image anymore.
 Over the weekend I created a Sage distribution aimed at Windows users
based on "puppy Linux" using Virtual Box:

 http://sage.math.washington.edu/home/wstein/binaries/sage_vbox-4.1.2.zip

It's a 659MB download.  However, once you download and extract the
zip, the resulting files still only use 659MB.  If you then import the
virtual machine into VirtualBox (follow the included readme), you'll
use a total of just over 700MB disk space.  This is *much* better than
the 3.5GB used by the extracted sage-vmware-*.zip!  Interestingly,
it's also much better than the 1.5GB used by an extracted Sage install
on Linux or OS X.

What I need very much is for some people to test sage_vbox-4.1.2.zip.
Please test and report back.

 -- William


-- 
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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Problem executing cell contents in version 4.1.2

[lutusp]

On Oct 19, 10:19 am, William Stein  wrote:
> On Mon, Oct 19, 2009 at 9:30 AM, lutusp  wrote:
>
> > Hi all. Just downloaded 4.1.2 for Fedora 11. When I start the notebook
> > server, everything is fine -- the worksheets get converted and placed
> > in the new directory sage_notebook.sagenb, then I can load a
> > worksheet. But when I try to execute a cell, I get this error message:
>
> > NameError: name 'sagenb' is not defined
>
> That error indicates that Sage is running worksheet processes
>  using a version of Sage prior to 4.1.2.
>
> What precisely is the command you use to start your notebook server?

1. Installed Fedora 11 binary in my sage directory -- this creates a
subdirectory named "sage-4.1.2-linux-Fedora_release_11_Leonidas-i686-
Linux"

2. Renamed two prior Sage directories by prefixing "xxx_" to their
names (to keep them from being found but to avoid having to reinstall
them).

3. Entered the new Fedora 11 Sage 4.1.2  directory.

4. Entered "./sage -notebook"

5. Sage offers to convert the prior notebook formats, completes, runs
server, launches Firefox.

6. Log onto Sage server -- note  "Version 4.1.2" at upper left -- open
worksheet.

7. Try to execute a cell's contents. Message:

 "Traceback (click to the left for traceback)
 ...
 NameError: name 'sagenb' is not defined"

 (no traceback contents)

8. Create a new notebook.

  (Now it gets complicated)

9. Copy example old worksheet cell contents into new worksheet. Up to
a certain level of complexity, copied cell contents will work in a
newly created worksheet but continue to fail if the old worksheet is
loaded and executed. I emphasize when I say "old worksheet" I am
talking about worksheets converted to the new format and placed in the
new directory "sage_notebook.sagenb", and I just took the precaution
of removing the old directory "sage_notebook" entirely to avoid any
possible confusion.

I emphasize that the old worksheet fails to execute cell contents, but
the same cell copied to a new worksheet succeeds ... but only up to a
certain level of complexity, after which the new and old scenarios
(e.g. execution of old worksheet cells versus old cell content copied
into new worksheet) become the same.

> Do you have any old versions of Sage installed anywhere?

Yes, two. But I have renamed the directories with a prefixed "xxx_" to
prevent their being located. One note -- if any Sage software is
locating resources using inodes instead of paths, this would prevent
the above remedy from having the expected effect.

If you want an example of a complex legacy worksheet, there is a
resource .sws file attached tho the latest page of my Sage tutorial
located at:

http://arachnoid.com/sage/polynomial.html

The .sws URL:

http://arachnoid.com/sage/sage/polynomial_regression.sws

10. I tried compiling and installing the source (which was
successful), but with the same outcome.

Please ask any questions I have failed to proactively answer.

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

[sage-devel] Re: SCons in Sage

[William Stein]

On Mon, Oct 19, 2009 at 9:59 AM, Dr. David Kirkby
 wrote:
>
> brandon.bar...@gmail.com wrote:
>> The SCons wiki even seems to be trying to say cmake is better:
>> http://www.scons.org/wiki/SconsVsOtherBuildTools ;-)  I have no
>> opinion myself, other than the problems you have already outlined;
>> another dev I know has also expressed his dislike of it, but he cannot
>> be trusted in this matter since he is anti-python.
>
> I would have thought it been sensible to make SCons compatible with GNU make,
> and not require its cmake to work optimally. I would not be surprised if
> switching to 'cmake' instead of the GNU make does not cause problems 
> elsewhere.

"cmake" is *not* just an alternative implementation of GNU make.
Instead cmake is "...the cross-platform, open-source build system.
CMake is a family of tools designed to build, test and package
software. CMake is used to control the software compilation process
using simple platform and compiler independent configuration files.
CMake generates native makefiles and workspaces that can be used in
the compiler environment of your choice."  It's more an alternative to
autoconf than "make".

William


> But to me, it seems SCons sends the wrong flags to the compiler. Even if you
> force it too use the right compiler, it mixes up the flags. I'd rather control
> the flags myself, not have some program insert them unless it can do it
> properly. Which I am not convinced SCons can with Sun Studio.
>
> In short, it seems a mess to me.
>
> Dave
>
>
> >
>



-- 
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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Problem executing cell contents in version 4.1.2

[William Stein]

On Mon, Oct 19, 2009 at 9:30 AM, lutusp  wrote:
>
> Hi all. Just downloaded 4.1.2 for Fedora 11. When I start the notebook
> server, everything is fine -- the worksheets get converted and placed
> in the new directory sage_notebook.sagenb, then I can load a
> worksheet. But when I try to execute a cell, I get this error message:
>
> NameError: name 'sagenb' is not defined

That error indicates that Sage is running worksheet processes
 using a version of Sage prior to 4.1.2.

What precisely is the command you use to start your notebook server?
Do you have any old versions of Sage installed anywhere?

William

>
> I have tried a bunch of things -- I downloaded the source and compiled
> it, with no change. I can create a new worksheet not derived from
> 4.1.1, with no change.
>
> Stumped.
>
> >
>



-- 
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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: SCons in Sage

[Dr. David Kirkby]

brandon.bar...@gmail.com wrote:
> The SCons wiki even seems to be trying to say cmake is better:
> http://www.scons.org/wiki/SconsVsOtherBuildTools ;-)  I have no
> opinion myself, other than the problems you have already outlined;
> another dev I know has also expressed his dislike of it, but he cannot
> be trusted in this matter since he is anti-python.

I would have thought it been sensible to make SCons compatible with GNU make, 
and not require its cmake to work optimally. I would not be surprised if 
switching to 'cmake' instead of the GNU make does not cause problems elsewhere.

But to me, it seems SCons sends the wrong flags to the compiler. Even if you 
force it too use the right compiler, it mixes up the flags. I'd rather control 
the flags myself, not have some program insert them unless it can do it 
properly. Which I am not convinced SCons can with Sun Studio.

In short, it seems a mess to me.

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-unsubscr...@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 on Solaris - now builds 'out of the box' with gcc 4.4.1

[Dr. David Kirkby]

brandon.bar...@gmail.com wrote:
> On the latest opensolaris running on amd64 and using gcc 4.4.1 I did
> hit the bug you mentioned, but in sqlite (running the gnu autotools
> again for sqlite did not seem to help):
> 
> 
> ar cru .libs/libsqlite3.a  sqlite3.o
> ranlib .libs/libsqlite3.a
> creating libsqlite3.la
> (cd .libs && rm -f libsqlite3.la && ln -s ../libsqlite3.la
> libsqlite3.la)
> if /usr/local/bin/gcc -DPACKAGE_NAME=\"sqlite\" -DPACKAGE_TARNAME=
> \"sqlite\" -DPACKAGE_VERSION=\"3.6.17\" -DPACKAGE_STRING=\"sqlite\
> 3.6.17\" -DPACKAGE_BUGREPORT=\"http://www.sqlite.org\"; -DPACKAGE=
> \"sqlite\
> " -DVERSION=\"3.6.17\" -D_FILE_OFFSET_BITS=64 -DSTDC_HEADERS=1 -
> DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -
> DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -
> DHAVE_INTTYPES_H=1 -DHAVE_STD
> INT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DHAVE_FDATASYNC=1 -
> DHAVE_USLEEP=1 -DHAVE_LOCALTIME_R=1 -DHAVE_GMTIME_R=1 -
> DHAVE_READLINE=1 -I. -I.   -I /export/home/brandon/sage-4.1.2/local/
> include -DSQLITE_THRE
> ADSAFE=1  -g  -O2  -MT shell.o -MD -MP -MF ".deps/shell.Tpo" -c -o
> shell.o shell.c; \
> then mv -f ".deps/shell.Tpo" ".deps/shell.Po"; else rm -f
> ".deps/shell.Tpo"; exit 1; fi
> /bin/sh ./libtool --tag=CC --mode=link /usr/local/bin/gcc -
> DSQLITE_THREADSAFE=1  -g  -O2-o sqlite3  shell.o ./libsqlite3.la -
> lreadline -lcurses
> /usr/local/bin/gcc -DSQLITE_THREADSAFE=1 -g -O2 -o .libs/sqlite3
> shell.o  ./.libs/libsqlite3.so -lreadline -lcurses  -Wl,--rpath -Wl,/
> export/home/brandon/sage-4.1.2/local/lib
> usage: ld
> [-3:6:abc:d:e:f:h:il:mo:p:rstu:z:B:CD:F:GI:L:M:N:P:Q:R:S:VW:Y:?] file
> (s)
>


Can you give me the output of

$ gcc -v
$ ld --version.
$ make --version
$ which make
$ which tar
$ which gcc
$ echo $PATH

(even if these create errors, tell me what they are)

What I think is happening is that you have a GNU version of a linker names 'ld' 
in your path before the Sun linker at /usr/ccs/bin AND gcc is configured to use 
the Sun linker.

If gcc -v shows something like

--with-ld=/usr/ccs/bin/ld

then is configured to use the Sun linker, in /usr/ccs/bin. If however you have 
another linker in your path named 'ld' (i.e. probably from GNU binutils) before 
the Sun linker, the some of the build scripts in Sage will assume you are using 
the GNU linker, and pass GNU-specific flags to the linker.


I suspect by changing the order of your the things in your path, so the Sun 
linker is picked up first, then it will solve it. However, if changing your 
path 
means that you pick up the Sun version of make before then GNU one, then you 
will get a problem. The trick will be to put the GNU version of 'make' and 
'tar' 
in your path before the Sun versions, but put the Sun versions of the other 
tools before the GNU ones. (See my previous note about creating a directory and 
copying the GNU version of make and tar to it.

Perhaps with the notes above, you can solve this yourself, but if not, let me 
have the output of the commands

Here you will see what I have.

1)
$ gcc -v

--with-ld=/usr/ccs/bin/ld --without-gnu-ld


so gcc is configured to use the Sun linker.

2)
$ ld --version
ld: illegal option -- version
usage: ld [-6:abc:d:e:f:h:il:mo:p:rstu:z:B:CD:F:GI:L:M:N:P:Q:R:S:VY:?] file(s)

So I have the Sun linker in my path before any GNU one

3)
$ make --version
GNU Make 3.80

So I have the GNU make in my path first, before the Sun one.


4)
$ which make
/export/home/drkirkby/bin/make

I've put a copy of the GNU make in a directory which is picked up early.

It basically boils down to this

* If gcc is configured to use the Sun linker, then the Sun linker must be in 
your path first, as some Sage scrips check what linker you have in your path 
first, and use that to determine what linker flags to use. (At some point I 
hope 
to do a better test, but it hard to determine unambiguously what linker gcc 
uses.


* The GNU version of make and tar must be in your path before any GNU version.

* Since the Sun version of 'make' is in the same directory as the Sun version 
of 
'ld' then you must copy GNU versions of make and tar to a directory such as 
$HOME/bin-for-sage, and ensure that is in your path first.


I hope that helps somewhat



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

[sage-devel] Problem executing cell contents in version 4.1.2

[lutusp]

Hi all. Just downloaded 4.1.2 for Fedora 11. When I start the notebook
server, everything is fine -- the worksheets get converted and placed
in the new directory sage_notebook.sagenb, then I can load a
worksheet. But when I try to execute a cell, I get this error message:

NameError: name 'sagenb' is not defined

I have tried a bunch of things -- I downloaded the source and compiled
it, with no change. I can create a new worksheet not derived from
4.1.1, with no change.

Stumped.

--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 4.2.alpha0 released

[John H Palmieri]

On Oct 19, 12:10 am, Mike Hansen  wrote:
> Hello all,
>
> Sage 4.2.alpha0 is out.  Source and binary areavailable at

When installing on Mac OS X 10.5, either 32 or 64 bit, I see this at
the very beginning:

  ./install: line 131: PIL: command not found

There are two problems with line 131: it says

  PIL =`$newest pyprocessing`

There shouldn't be a space before the = sign, and I assume that
"pyprocessing" should actually say "pil".

  John


  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-unsubscr...@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 on Solaris - now builds 'out of the box' with gcc 4.4.1

[brandon.bar...@gmail.com]

On the latest opensolaris running on amd64 and using gcc 4.4.1 I did
hit the bug you mentioned, but in sqlite (running the gnu autotools
again for sqlite did not seem to help):


ar cru .libs/libsqlite3.a  sqlite3.o
ranlib .libs/libsqlite3.a
creating libsqlite3.la
(cd .libs && rm -f libsqlite3.la && ln -s ../libsqlite3.la
libsqlite3.la)
if /usr/local/bin/gcc -DPACKAGE_NAME=\"sqlite\" -DPACKAGE_TARNAME=
\"sqlite\" -DPACKAGE_VERSION=\"3.6.17\" -DPACKAGE_STRING=\"sqlite\
3.6.17\" -DPACKAGE_BUGREPORT=\"http://www.sqlite.org\"; -DPACKAGE=
\"sqlite\
" -DVERSION=\"3.6.17\" -D_FILE_OFFSET_BITS=64 -DSTDC_HEADERS=1 -
DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -
DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -
DHAVE_INTTYPES_H=1 -DHAVE_STD
INT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DHAVE_FDATASYNC=1 -
DHAVE_USLEEP=1 -DHAVE_LOCALTIME_R=1 -DHAVE_GMTIME_R=1 -
DHAVE_READLINE=1 -I. -I.   -I /export/home/brandon/sage-4.1.2/local/
include -DSQLITE_THRE
ADSAFE=1  -g  -O2  -MT shell.o -MD -MP -MF ".deps/shell.Tpo" -c -o
shell.o shell.c; \
then mv -f ".deps/shell.Tpo" ".deps/shell.Po"; else rm -f
".deps/shell.Tpo"; exit 1; fi
/bin/sh ./libtool --tag=CC --mode=link /usr/local/bin/gcc -
DSQLITE_THREADSAFE=1  -g  -O2-o sqlite3  shell.o ./libsqlite3.la -
lreadline -lcurses
/usr/local/bin/gcc -DSQLITE_THREADSAFE=1 -g -O2 -o .libs/sqlite3
shell.o  ./.libs/libsqlite3.so -lreadline -lcurses  -Wl,--rpath -Wl,/
export/home/brandon/sage-4.1.2/local/lib
usage: ld
[-3:6:abc:d:e:f:h:il:mo:p:rstu:z:B:CD:F:GI:L:M:N:P:Q:R:S:VW:Y:?] file
(s)

On Oct 19, 4:58 am, "Dr. David Kirkby" 
wrote:
> solarg wrote:
> > hello all,
> > is there any binaries for solaris/x86 (or amd64) that i can download
> > somewhere?
>
> > thanks in advance,
>
> > gerard
>
> I do not know if there is. There were, but  when one goes to the download page
>
> http://www.sagemath.org/download.html
>
> and click 'Solaris binaries'
>
> one goes tohttp://www.sagemath.org/download.html/solaris/
>
> However, that has a 404 error - page not found.
>
> The x86 binaries are quite old.
>
> I would suggest you try to build from source, though I have not done so myself
> on x86, so I can only suggest what you will need for SPARC. It would be *very*
> useful if you could try, and let us know of any issues you get.
>
> Here are a few tips, which are needed for SPARC, and probably are Solaris x86 
> too.
>
> 1) Get yourself a gcc of >= 4.0.1 which is built to include Fortran.
>
> $ gcc -v should show something like
>
> --enable-languages=c,c++,fortran
>
> If not, then it includes all languages, and will be ok. Some are built with 
> only
> C and C++ support.
>
> 2)  You will probably need to copy the file
>
> http://sage.math.washington.edu/home/kirkby/Solaris-fixes/pari-2.3.3
>
> to spkg/standard/ directory, then 'touch' it, to make it the latest version.
>
> This needs reviewing by someone and is ticket
>
> http://sagetrac.org/sage_trac/ticket/6579
>
> 3) If you have Sun Studio compiler installed (usually in /opt/SUNWspro or
> /opt/sunstudio12.1) move the directory constraining the Sun Studio compiler to
> somewhere else temporarily if you have it installed.
>
> # mv /opt/SUNWspro /opt/xxxSUNWspro
>
> will do it. You will need root access to do that. Sorry, but SCons screws up 
> the
> build otherwise due to this bug
>
> http://sagetrac.org/sage_trac/ticket/6595
>
> which has not been resolved yet.
>
> 4) You *might* hit this bug, which has only been observed on Solaris x86 to my
> knowledge
>
> http://sagetrac.org/sage_trac/ticket/6539
>
> I'm not aware of a solution to that.
>
> Most important, also see the notes I posted earlier.
>
> 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: SCons in Sage

[brandon.bar...@gmail.com]

The SCons wiki even seems to be trying to say cmake is better:
http://www.scons.org/wiki/SconsVsOtherBuildTools ;-)  I have no
opinion myself, other than the problems you have already outlined;
another dev I know has also expressed his dislike of it, but he cannot
be trusted in this matter since he is anti-python.


On Oct 19, 5:25 am, "Dr. David Kirkby" 
wrote:
> What is the view of others on the use of SCons in code in Sage? Personally 
> I've
> come to the opinion that it is more trouble than it is worth.
>
> It may be that few people know how to use it properly, in which case you might
> ask why do they use it. I do not use a chainsaw, as I do not know how to use 
> one
> properly! OK, the consequences for using a chainsaw are somewhat larger than
> using SCons, but the point is, if you do not know how to use something, it's
> probably best you do not.
>
> But I believe SCons has more fundamental issues. Here are a fraction of the
> issues I know about.
>
> 1) cliquer was trying to use a sun compiler 'cc' at one time, despite the fact
> there was no such compiler in the path, so it failed
>
> http://sagetrac.org/sage_trac/ticket/6852
>
> I gather from Georg Weber that SCons was removed from Cliquer?
>
> http://sagetrac.org/sage_trac/ticket/6681
>
> due to problems on OS X. That is one good bit of news to me.
>
> 2) PolyBoRi used to ignore CC and CXX. That was fixed,  but now SCons uses a
> mixture of Sun and GNU flags if trying to build PolyBoRi with Sun Studio.
>
> http://sagetrac.org/sage_trac/ticket/7034
>
> 3)  c_lib uses a mixture of Sun and GNU compilers, despite I've tried my best 
> to
> force it to honor the environment variables CC and CXX.
>
> http://sagetrac.org/sage_trac/ticket/6595
>
> What do others think about SCons? Is it the best thing since sliced bread, or
> more trouble than it is worth? If something will not work with SCons, is
> converting it to a simple makefile and/or autoconf/automake system considered 
> ok?
>
> Personally, I'd like to see the back of SCons.
>
> 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: Error rebuilding MPIR and ATLAS

[William Stein]

On Mon, Oct 19, 2009 at 7:36 AM, Pablo Angulo  wrote:
>
> Hello: After an "Illegal instruction" error, I tried to rebuild MPIR and
> ATLAS (as explained in
> http://wiki.sagemath.org/faq#Otherquestions), but make failed, with the
> following log:
>
>
> /bin/ls: cannot access bzip2-*-install: No such file or directory
> /bin/ls: cannot access dir-*-install: No such file or directory
> /bin/ls: cannot access prereq-*-install: No such file or directory
> make[1]: Entering directory
> `/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg'
> standard/deps:43: warning: overriding commands for target `installed/'
> standard/deps:40: warning: ignoring old commands for target `installed/'
> standard/deps:46: warning: overriding commands for target `installed/'
> standard/deps:43: warning: ignoring old commands for target `installed/'
> standard/deps:188: warning: overriding commands for target `installed/'
> standard/deps:46: warning: ignoring old commands for target `installed/'
> make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
> dropped.
> make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
> dropped.
> make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
> dropped.
> make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
> make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
> make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
> make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
> dropped.
> make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
> dropped.
> make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
> dropped.
> make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
> dropped.
> make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
> dropped.
> make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
> dropped.
> make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
> make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
> make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
> make[1]: Circular installed/ <- installed/ dependency dropped.
> make[1]: Circular installed/ <- installed/ dependency dropped.
> make[1]: Circular installed/ <- installed/ dependency dropped.
> sage-spkg mpir-1.2.p4 2>&1
> You must set the SAGE_ROOT environment variable or
> run this script from the SAGE_ROOT or
> SAGE_ROOT/local/bin/ directory.
> mpir-1.2.p4
> Machine:
> Linux nevera 2.6.24-24-generic #1 SMP Fri Sep 18 16:49:39 UTC 2009 i686
> GNU/Linux
> Deleting directories from past builds of previous/current versions of
> mpir-1.2.p4
> Extracting package
> /opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/standard/mpir-1.2.p4.spkg
> ...
> -rw-r--r-- 1 pablo pablo 107 2009-08-16 01:00
> /opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/standard/mpir-1.2.p4.spkg
> tar: This does not look like a tar archive
> tar: Error exit delayed from previous errors
> Finished extraction
> sage: After decompressing the directory mpir-1.2.p4 does not exist
> This means that the corresponding .spkg needs to be downloaded
> again.
> http://www.sagemath.org//packages/optional/mpir-1.2.p4.spkg -->
> mpir-1.2.p4.spkg
> [ ]
> http://www.sagemath.org//packages/standard/mpir-1.2.p4.spkg -->
> mpir-1.2.p4.spkg
> [ ]
> http://www.sagemath.org//packages/experimental/mpir-1.2.p4.spkg -->
> mpir-1.2.p4.spkg
> [ ]
> http://www.sagemath.org//packages/archive/mpir-1.2.p4.spkg -->
> mpir-1.2.p4.spkg
> [ ]
> **
> * Unable to download mpir-1.2.p4
> * Please see http://www.sagemath.org//packages for a list of valid
> * packages or check the package name.
> **
> /opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/build
> bunzip2: Can't open input file mpir-1.2.p4.spkg: No such file or directory.
> tar: This does not look like a tar archive
> tar: Error exit delayed from previous errors
> tar: mpir-1.2.p4.spkg: Cannot open: No such file or directory
> tar: Error is not recoverable: exiting now
> Second download resulted in a corrupted package.
> make[1]: *** [installed/mpir-1.2.p4] Error 1
> make[1]: Leaving directory
> `/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg'
>
> real    0m3.567s
> user    0m0.108s
> sys     0m0.036s
> Error building Sage.
> make: *** [all] Error 1
>
>
> Any ideas? I can try to completely build from source, but what I'd
> really like to know is if "rebuild MPIR and ATLAS" solve the "Illegal
> instruction" problem.

Yes, it would.

You should use sage-4.1.2, by the way, which just came out.

Also, the problem above looks like a networking issue?  You can
download spkg's directly from http://sagemath.org/packages/standard/
and do
  ./sage -f name_of_packa

[sage-devel] Error rebuilding MPIR and ATLAS

[Pablo Angulo]

Hello: After an "Illegal instruction" error, I tried to rebuild MPIR and
ATLAS (as explained in
http://wiki.sagemath.org/faq#Otherquestions), but make failed, with the
following log:


/bin/ls: cannot access bzip2-*-install: No such file or directory
/bin/ls: cannot access dir-*-install: No such file or directory
/bin/ls: cannot access prereq-*-install: No such file or directory
make[1]: Entering directory
`/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg'
standard/deps:43: warning: overriding commands for target `installed/'
standard/deps:40: warning: ignoring old commands for target `installed/'
standard/deps:46: warning: overriding commands for target `installed/'
standard/deps:43: warning: ignoring old commands for target `installed/'
standard/deps:188: warning: overriding commands for target `installed/'
standard/deps:46: warning: ignoring old commands for target `installed/'
make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
dropped.
make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
dropped.
make[1]: Circular installed/python-2.6.2.p1 <- installed/ dependency
dropped.
make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
make[1]: Circular installed/zlib-1.2.3.p4 <- installed/ dependency dropped.
make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
dropped.
make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
dropped.
make[1]: Circular installed/termcap-1.3.1.p0 <- installed/ dependency
dropped.
make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
dropped.
make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
dropped.
make[1]: Circular installed/readline-5.2.p7 <- installed/ dependency
dropped.
make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
make[1]: Circular installed/libpng-1.2.35 <- installed/ dependency dropped.
make[1]: Circular installed/ <- installed/ dependency dropped.
make[1]: Circular installed/ <- installed/ dependency dropped.
make[1]: Circular installed/ <- installed/ dependency dropped.
sage-spkg mpir-1.2.p4 2>&1
You must set the SAGE_ROOT environment variable or
run this script from the SAGE_ROOT or
SAGE_ROOT/local/bin/ directory.
mpir-1.2.p4
Machine:
Linux nevera 2.6.24-24-generic #1 SMP Fri Sep 18 16:49:39 UTC 2009 i686
GNU/Linux
Deleting directories from past builds of previous/current versions of
mpir-1.2.p4
Extracting package
/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/standard/mpir-1.2.p4.spkg
...
-rw-r--r-- 1 pablo pablo 107 2009-08-16 01:00
/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/standard/mpir-1.2.p4.spkg
tar: This does not look like a tar archive
tar: Error exit delayed from previous errors
Finished extraction
sage: After decompressing the directory mpir-1.2.p4 does not exist
This means that the corresponding .spkg needs to be downloaded
again.
http://www.sagemath.org//packages/optional/mpir-1.2.p4.spkg -->
mpir-1.2.p4.spkg
[ ]
http://www.sagemath.org//packages/standard/mpir-1.2.p4.spkg -->
mpir-1.2.p4.spkg
[ ]
http://www.sagemath.org//packages/experimental/mpir-1.2.p4.spkg -->
mpir-1.2.p4.spkg
[ ]
http://www.sagemath.org//packages/archive/mpir-1.2.p4.spkg -->
mpir-1.2.p4.spkg
[ ]
**
* Unable to download mpir-1.2.p4
* Please see http://www.sagemath.org//packages for a list of valid
* packages or check the package name.
**
/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg/build
bunzip2: Can't open input file mpir-1.2.p4.spkg: No such file or directory.
tar: This does not look like a tar archive
tar: Error exit delayed from previous errors
tar: mpir-1.2.p4.spkg: Cannot open: No such file or directory
tar: Error is not recoverable: exiting now
Second download resulted in a corrupted package.
make[1]: *** [installed/mpir-1.2.p4] Error 1
make[1]: Leaving directory
`/opt/sage-4.1.1-linux-Debian_GNU_Linux_5.0_lenny-i686-Linux/spkg'

real0m3.567s
user0m0.108s
sys 0m0.036s
Error building Sage.
make: *** [all] Error 1


Any ideas? I can try to completely build from source, but what I'd
really like to know is if "rebuild MPIR and ATLAS" solve the "Illegal
instruction" problem.

Thanks!


--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 notebook servers

[Joaquim Puig]

 1. Where?     (E.g., "Dept. of Mathematics, University of Maryland")

Faculty of Mathematics and Statistics (FME)
Polytechnical University of Catalonia (UPC)
Barcelona, Catalonia-Spain

>    2. Why?   (E.g, "for our Math 411 course on differential equations")

Presently we are using it to allow members of the university (it is
behind a firewall) to try SAGE without having to install it.

Regards,

Joaquim Puig

--
Departament de Matemàtica Aplicada I
Universitat Politècnica de Catalunya
Diagonal 647
08028 Barcelona

> Thanks!!
>
> William
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://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-unsubscr...@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 4.2.alpha0 released

[Jaap Spies]

Pat LeSmithe wrote:
> On 10/19/2009 03:35 AM, Jaap Spies wrote:
>> Upgrading from sage-4.1.2.rc2 failed on Fedora 11, 32 bit:
>>
>>  env = pickle.load(picklefile)
>> AttributeError: 'module' object has no attribute 'RedirStream'
>> [j...@paix sage-4.1.2.rc2]$
> 
> Can you try deleting SAGE_ROOT/devel/sage/doc/output and rebuilding the
> docs?  It seems Sphinx 0.6.3 does not automatically detect and upgrade
> an environment pickle made by Sphinx 0.5.1.
> 

That seems to work. Thanks,

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-unsubscr...@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 4.2.alpha0 released

[Pat LeSmithe]

On 10/19/2009 03:35 AM, Jaap Spies wrote:
> Upgrading from sage-4.1.2.rc2 failed on Fedora 11, 32 bit:
> 
>  env = pickle.load(picklefile)
> AttributeError: 'module' object has no attribute 'RedirStream'
> [j...@paix sage-4.1.2.rc2]$

Can you try deleting SAGE_ROOT/devel/sage/doc/output and rebuilding the
docs?  It seems Sphinx 0.6.3 does not automatically detect and upgrade
an environment pickle made by Sphinx 0.5.1.

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

[sage-devel] Looking for a good name...

[Florent Hivert]

  Dear all,

I'm looking for a good name for the following concept:

In several place in Sage, we need to build a set (ie a parent) which will
generated some objects whose parents are actually a different set. Some
examples:

- I want to model the enumerated set of non negative integer or prime numbers,
but I'd rather the element generated by this set to be plain sage Integers
with parent Integer Ring.
- More generally, I'm building a subset of a certain set and I want the
element to be generated as element for the larger set.
- I'm building the union of different sets...

Of course a different way to do it is to set up coercion to automatically
convert element, but I'd rather to leave this choice to the user.

In MuPAD such a parent (= domain in MuPAD's terminology) was called a "facade
domain". I'm not sure this name of facade is very meaning full for the average
sage user. So I'm looking for suggestions. By the way, maybe this concept is
already used somewhere else in Sage. If so, please give me some pointers.

Thanks for your help,

Cheers,

Florent



--~--~-~--~~~---~--~~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel-unsubscr...@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 4.2.alpha0 released

[Jaap Spies]

Mike Hansen wrote:
> Hello all,
> 
> Sage 4.2.alpha0 is out.  Source and binary areavailable at
> 
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0.tar
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0sage-4.2.alpha0-sage.math.washington.edu-x86_64-Linux.tar.gz
> 
> The upgrade path is
> 
> http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0
> 

Upgrading from sage-4.1.2.rc2 failed on Fedora 11, 32 bit:


sage-spkg dsage-1.0.1.p0 2>&1
Warning: Attempted to overwrite SAGE_ROOT environment variable
dsage-1.0.1.p0
Machine:
Linux paix 2.6.30.8-64.fc11.i686.PAE #1 SMP Fri Sep 25 04:56:58 EDT 2009 i686 
i686 i386 GNU/Linux
sage: dsage-1.0.1.p0 is already installed
cd /home/jaap/downloads/sage-4.1.2.rc2 && . local/bin/sage-env && cd local/bin/ 
&& ./sage-make_relative
Making script relocatable

real0m0.914s
user0m0.300s
sys 0m0.371s
Traceback (most recent call last):
   File "/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/common/builder.py", 
line 1009, in 
 getattr(get_builder(name), type)()
   File "/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/common/builder.py", 
line 260, in _wrapper
 getattr(get_builder(document), name)(*args, **kwds)
   File "/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/common/builder.py", 
line 354, in _wrapper
 for module_name in self.get_modified_modules():
   File "/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/common/builder.py", 
line 445, in get_modified_modules
 env = self.get_sphinx_environment()
   File "/home/jaap/downloads/sage-4.1.2.rc2/devel/sage/doc/common/builder.py", 
line 415, in get_sphinx_environment
 env = BuildEnvironment.frompickle(config, env_pickle)
   File 
"/home/jaap/downloads/sage-4.1.2.rc2/local/lib/python2.6/site-packages/Sphinx-0.6.3-py2.6.egg/sphinx/environment.py",
 line 204, in frompickle
 env = pickle.load(picklefile)
AttributeError: 'module' object has no attribute 'RedirStream'
[j...@paix sage-4.1.2.rc2]$


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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] SCons in Sage

[Dr. David Kirkby]

What is the view of others on the use of SCons in code in Sage? Personally I've 
come to the opinion that it is more trouble than it is worth.

It may be that few people know how to use it properly, in which case you might 
ask why do they use it. I do not use a chainsaw, as I do not know how to use 
one 
properly! OK, the consequences for using a chainsaw are somewhat larger than 
using SCons, but the point is, if you do not know how to use something, it's 
probably best you do not.

But I believe SCons has more fundamental issues. Here are a fraction of the 
issues I know about.

1) cliquer was trying to use a sun compiler 'cc' at one time, despite the fact 
there was no such compiler in the path, so it failed

http://sagetrac.org/sage_trac/ticket/6852

I gather from Georg Weber that SCons was removed from Cliquer?

http://sagetrac.org/sage_trac/ticket/6681

due to problems on OS X. That is one good bit of news to me.

2) PolyBoRi used to ignore CC and CXX. That was fixed,  but now SCons uses a 
mixture of Sun and GNU flags if trying to build PolyBoRi with Sun Studio.

http://sagetrac.org/sage_trac/ticket/7034

3)  c_lib uses a mixture of Sun and GNU compilers, despite I've tried my best 
to 
force it to honor the environment variables CC and CXX.

http://sagetrac.org/sage_trac/ticket/6595

What do others think about SCons? Is it the best thing since sliced bread, or 
more trouble than it is worth? If something will not work with SCons, is 
converting it to a simple makefile and/or autoconf/automake system considered 
ok?

Personally, I'd like to see the back of SCons.


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-unsubscr...@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 on Solaris - now builds 'out of the box' with gcc 4.4.1

[Dr. David Kirkby]

solarg wrote:
> hello all,
> is there any binaries for solaris/x86 (or amd64) that i can download 
> somewhere?
> 
> thanks in advance,
> 
> gerard

I do not know if there is. There were, but  when one goes to the download page

http://www.sagemath.org/download.html

and click 'Solaris binaries'

one goes to http://www.sagemath.org/download.html/solaris/

However, that has a 404 error - page not found.

The x86 binaries are quite old.

I would suggest you try to build from source, though I have not done so myself 
on x86, so I can only suggest what you will need for SPARC. It would be *very* 
useful if you could try, and let us know of any issues you get.

Here are a few tips, which are needed for SPARC, and probably are Solaris x86 
too.

1) Get yourself a gcc of >= 4.0.1 which is built to include Fortran.

$ gcc -v should show something like

--enable-languages=c,c++,fortran

If not, then it includes all languages, and will be ok. Some are built with 
only 
C and C++ support.

2)  You will probably need to copy the file

http://sage.math.washington.edu/home/kirkby/Solaris-fixes/pari-2.3.3.p5/pari-2.3.3.p5.spkg

to spkg/standard/ directory, then 'touch' it, to make it the latest version.

This needs reviewing by someone and is ticket

http://sagetrac.org/sage_trac/ticket/6579


3) If you have Sun Studio compiler installed (usually in /opt/SUNWspro or 
/opt/sunstudio12.1) move the directory constraining the Sun Studio compiler to 
somewhere else temporarily if you have it installed.

# mv /opt/SUNWspro /opt/xxxSUNWspro

will do it. You will need root access to do that. Sorry, but SCons screws up 
the 
build otherwise due to this bug

http://sagetrac.org/sage_trac/ticket/6595

which has not been resolved yet.

4) You *might* hit this bug, which has only been observed on Solaris x86 to my 
knowledge

http://sagetrac.org/sage_trac/ticket/6539

I'm not aware of a solution to that.



Most important, also see the notes I posted earlier.


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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Re: html5 video in the notebook

[Wilfried Huss]

William Stein schrieb:
> On Sun, Oct 18, 2009 at 5:10 AM, Wilfried Huss
>  wrote:
>   
>> I have problems using the html5  tag with the notebook.
>>
>> Linking to an external video file works as expected. But if I
>> try to use an ogg theora video, which is uploaded into the DATA
>> directory, as the source of the  element, the file is not
>> recognized. I suspect that the notebook webserver does not
>> use the correct Mimetype for ogv files.
>>
>> Strangely it works on sagenb, but not on my local sage installation.
>> 
>
> What version of Sage are you running locally?
>   
I tried with 4.1.1 and 4.1.2 on mandriva, and now with
4.1.2 on debian where it works.

So it looks like this is a bug in mandriva.
>> I have a test worksheet at:
>>  http://www.sagenb.org/home/pub/755/
>>
>> Does anybody who is familiar with the notebook knows what is
>> needed to make this work on a local sage installation?
>> 
>
> I don't know why it doesn't work.  Maybe something needs to be changed
> in notebook/twist.py or something else having to do with
> Twisted.web2...
>
> One thing is that sagenb.org is proxied through apache, which might
> possibly add appropriate mime-type info?
>
> 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-unsubscr...@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 on Solaris - now builds 'out of the box' with gcc 4.4.1

[solarg]

hello all,
is there any binaries for solaris/x86 (or amd64) that i can download 
somewhere?

thanks in advance,

gerard

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

[sage-devel] Re: Grant Ideas: sage days workshops

[Rado]

+1 for workshop on error-correcting codes. One thing I can contribute
to such a workshop (if it is accepted) is adding improved Feng-Rao
(aka order bounds) algorithms for min. distance of AG codes, developed
by my advisor Iwan Duursma and me. Currently we have standalone code
in Python, which means it should be relatively painless. Also the bug
with Riemann-Roch spaces coming from Singular can be tackled (unless
it is solved already?).

Rado

On Oct 16, 4:03 pm, David Joyner  wrote:
> On Fri, Oct 16, 2009 at 9:38 AM, William Stein  wrote:
>
> > 2009/10/16 David Joyner :
>
> >> I'd love to see an error-correcting codes workshop!
>
> > What would be involved?  Can you send more details?
>
> There are a huge number of things missing.
>
> 1. a fast minimum_distance for an arbitrary non-binary code (minimum distance 
> of
> linear binary codes can be computed very quickly thanks to code by
> Robert Miller)
>
> 2. a function to compute a basis for the Riemann-Roch space of a
> reasonably general curve
> at a reasonably general divisor. This is needed to implement AG codes.)
>
> 3. implementation of the "weight function" machinery (this is an alternate
> way to compute certain AG-type codes, avoiding the Riemann-Roch space
> machinery; see the survey papers by Hoeholdt on AG codes...)
>
> 4. decoding algorithms
>
> 5. covering codes and covering radius algorithms (more generally,
> porting over Guava
> functions to Sage)
>
> 6. more generally, porting over what Magma has that Sage doesn't.
>
>
>
>
>
> >> On Thu, Oct 15, 2009 at 9:00 PM, William Stein  wrote:
>
> >>> Hi,
>
> >>> I'm applying to NSF for some Sage Days workshops (off in the future:
> >>> think 2-4 years from right now).   If anybody has any idea about what
> >>> might constitute a good topic for their dream Sage Days, please
> >>> respond to this email with the topic.  The more details you can give
> >>> the better.
>
> >>> Thanks!
>
> >>>  -- William
>
> >>> --
> >>> William Stein
> >>> Associate Professor of Mathematics
> >>> University of Washington
> >>>http://wstein.org
>
> > --
> > 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-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-devel] Sage 4.2.alpha0 released

[Mike Hansen]

Hello all,

Sage 4.2.alpha0 is out.  Source and binary areavailable at

http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0.tar
http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0sage-4.2.alpha0-sage.math.washington.edu-x86_64-Linux.tar.gz

The upgrade path is

http://sage.math.washington.edu/home/mhansen/release/4.2/alpha0/sage-4.2.alpha0

The tickets merged in 4.2.alpha0 are:

#5556: Karl-Dieter Crisman: symbolic gamma function is incoherent
[Reviewed by Mike Hansen]
#5985: Craig Citro, Nicolas M. Thiéry: cPickle: adds support for class
pickling customization [Reviewed by Craig Citro, Nicolas M. Thiéry]
#6482: Martin Albrecht: multivariate polynomial substitution has a
design flaw [Reviewed by Mike Hansen]
#1509: Matthew Trumbell: Allow text rotation in plotting [Reviewed by
Karl-Dieter Crisman]
#2341: Karl-Dieter Crisman, Ricardo Massaro: vector subs over symbolic
does not work [Reviewed by Jason Grout]
#2900: Karl-Dieter Crisman: matplotlib bug in imshow (probably fixed
in new version) [Reviewed by Jason Grout]
#3718: Karl-Dieter Crisman: calculus -- sage treats limits involving
floor and ceil completely wrong [Reviewed by Tim Dumol]
#4807: Michael Yurko: bug in exponential integral [Reviewed by Mike Hansen]
#4941: Yann Laigle-Chapuy: pari list slicing is extremely slow -- make
it much faster [Reviewed by Mike Hansen]
#5755: Karl-Dieter Crisman: error converting symbolic expression to
polynomial [Reviewed by Mike Hansen]
#5899: Tim Abbott: Update Debian build support for Sage spkg [Reviewed
by Mike Hansen]
#5986: Craig Citro, Nicolas M. Thiéry: Workaround mishandled nested
classes in Python and cPickle [Reviewed by Robert Bradshaw]
#5991: Nicolas M. Thiéry: Add a standard constructor for dynamic
classes [Reviewed by David Roe]
#6118: Robert Bradshaw, Craig Citro: integer shifting slow [Reviewed
by Mike Hansen, Craig Citro]
#6187: John Palmieri, Mitesh Patel: Fix post-cloning docbuild problems
and add new command-line options [Reviewed by John Cremona, Mike
Hansen]
#6189: Karl-Dieter Crisman: 'integrate' produces incorrect answer
[Reviewed by Mike Hansen]
#6228: Karl-Dieter Crisman: Wrong multiplicities when solving a
univariate polynomial equation [Reviewed by Mike Hansen]
#6304: David Harvey: intermittent crash in bernmm (4.0.2.rc0)
[Reviewed by Mike Hansen]
#6420: Karl-Dieter Crisman: Maxima integration error with 1/x^3
[Reviewed by Mike Hansen]
#6572: John Palmieri: tutorial: put double colon on its own line
[Reviewed by Mike Hansen]
#6586: John Palmieri: update Sphinx to version 0.6.2 or 0.6.3
[Reviewed by John Cremona, Tim Dumol, Mitesh Patel]
#6605: John Palmieri: sage -docbuild DOC FORMAT should do better error
checking on DOC [Reviewed by Mitesh Patel]
#6642: Karl-Dieter Crisman: problem with solve (from the tutorial)
[Reviewed by John Palmieri]
#6823: Nathann Cohen: Kneser Graph in graph_generators [Reviewed by Rob Beezer]
#6892: John Palmieri: change dollar signs to backticks [Reviewed by Mike Hansen]
#6899: Mike Hansen, Burcin Erocal: weird bug taking float of real part
of a symbolic [Reviewed by Burcin Erocal, Karl-Dieter Crisman]
#6906: John Palmieri: update Mercurial to version 1.3.1 [Reviewed by
Mike Hansen]
#6909: Martin Albrecht: move some commutative algebra to libSingular
functions interface [Reviewed by Michael Brickenstein, Burcin Erocal]
#6958: Robert Miller: prove_BSD function for elliptic curves over QQ
[Reviewed by William Stein]
#6977: Nicolas M. Thiéry: Implement __len__ to add support for MuPAD
objects as Python containers [Reviewed by Mike Hansen]
#6986: Jason Grout: Make PIL standard [Reviewed by Mike Hansen]
#7007: Mike Hansen: variables() method for polynomial rings returns
the constant as well as the variable. [Reviewed by Jason Grout]
#7010: Minh Van Nguyen: sanity check key value of the shift
cryptosystem [Reviewed by Rob Beezer]
#7015: William Stein, Minh Van Nguyen: cygwin port -- ratpoints --
don't build executable since we don't need it (and fails on cygwin)
[Reviewed by Mike Hansen]
#7045: David Kirkby: update libgcrypt to the latest version 1.4.4
[Reviewed by Adam Webb]
#7081: Tim Dumol: sage -c "..." runs from the local/bin/ directory.
ugh [Reviewed by William Stein]
#7084: Karl-Dieter Crisman: Make assumption comparison work for
GenericDeclarations [Reviewed by Tim Dumol]
#7117: John Palmieri: latex fix for RIF [Reviewed by Mike Hansen]
#3970: Mike Hansen: MaximaElements should not coerce into SR.
[Reviewed by Adam Webb]
#5243: Florent Hivert: non decreasing parking functions (depend now on
5255) [Reviewed by Dan Drake, Mike Hansen]
#5812: Wilfried Huss: option zorder has no effect for point() and
polygon() [Reviewed by Karl-Dieter Crisman, Volker Braun]
#6828: Nathann Cohen: Random Bipartite Graph [Reviewed by Rob Beezer]
#6891: Karl-Dieter Crisman: Bring functions.spike_functions.py to 100%
coverage [Reviewed by Mike Hansen]
#6928: Mitesh Patel: Upgrade the Trac spkg to 0.11.5 [Reviewed by Tim Dumol]
#6998: Mike Hanse