[sage-support] Re: Problems with Sage 2.11

[mabshoff]

On Apr 3, 6:23 pm, Georg Muntingh <[EMAIL PROTECTED]> wrote:
> I just installed Sage 2.11 on an old computer that I upgraded to
> Ubuntu Gutsy a couple of hours before. Any idea what's wrong?
>
> [EMAIL PROTECTED]:~/Apps/sage-2.11-ubuntu32-intel-i686-Linux$ ./sage
> --
> | SAGE Version 2.11, Release Date: 2008-03-30                        |
> | Type notebook() for the GUI, and license() for information.        |
> --
> The SAGE install tree may have moved.
> Regenerating Python.pyo and .pyc files that hardcode the install PATH
> (please wait at most a few minutes)...
> Please do not interrupt this.
>
> /home/georgm/Apps/sage-2.11-ubuntu32-intel-i686-Linux/local/bin/sage-
> sage: line 214:  7325 Illegal instruction     (core dumped) sage-
> ipython "$@" -c "$SAGE_STARTUP_COMMAND;"

That release has an ATLAS that uses instructions not available on that
CPU. You can compile from source [which I assume would take an
unreasonable amount of time] or uses the system's ATLAS build. See
"QUESTION: I downloaded a Sage binary and it crashes on startup with
Illegal instruction." from http://wiki.sagemath.org/faq

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: 3d plotting (jmol) problems

[mabshoff]

On Apr 2, 7:28 pm, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> On Wed, Apr 2, 2008 at 12:29 PM, mabshoff
>
> <[EMAIL PROTECTED]> wrote:
>
> >  FC6 ships gcj instead of the Sun Java SDK. Recent version of gcj do
> >  mostly work with jmol, but FC6's gcj is known to be broken. I would
> >  recommend to install the Sun Java SDK. Alternatively you could update
> >  the gcj runtime, but that is potentially painful and will likely break
> >  other Java related things.
>
> OK, I can use it with Sun's plugin in the notebook.  (Firefox did not
> freeze as before.)  I will ask the system administrator to set Sun's
> Java as the default.  Right now:
>
> boole[~]$ ll /usr/bin/java
> lrwxrwxrwx 1 root root 22 Apr 25  2007 /usr/bin/java -> 
> /etc/alternatives/java*
> boole[~]$ ll /etc/alternatives/java
> lrwxrwxrwx 1 root root 35 Apr 25  2007 /etc/alternatives/java ->
> /usr/lib/jvm/jre-1.4.2-gcj/bin/java*
>
> Thanks for the quick and helpful reply!

It has come up before, so it wasn't hard to figure out. :)

> Luis

Let us know if you run into any more trouble.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: 3d plotting (jmol) problems

[mabshoff]

On Apr 2, 7:17 pm, Jaap Spies <[EMAIL PROTECTED]> wrote:
> mabshoff wrote:

> > Yes. Note that Debian's gcj from testing works with jmol, so it is
> > probably some small stuff that needs to be fixed on jmol's end.
>
> I was to lazy to make it a trac ticket. I'm sorry for that. It
> maybe could have been solved a long time ago.

Well, it isn't really a ticket that belongs in our trac.

The problem is listed on the FAQ, so somebody ought to report this
upstream. Once they fix it open a ticket to update jmol to that
version.

> Jaap

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: 3d plotting (jmol) problems

[mabshoff]

On Apr 2, 7:07 pm, Jaap Spies <[EMAIL PROTECTED]> wrote:
> William Stein wrote:
> > On Wed, Apr 2, 2008 at 9:51 AM, Jaap Spies <[EMAIL PROTECTED]> wrote:
> >>  mabshoff wrote:
>
> >>  > On Apr 2, 6:17 pm, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> >>  >> Dear all,
>
> >>  >> I've been having some problems with 3D plots in Sage 2.11.
>
> >>  [...]
>
> >>  > FC6 ships gcj instead of the Sun Java SDK. Recent version of gcj do
> >>  > mostly work with jmol, but FC6's gcj is known to be broken. I would
> >>  > recommend to install the Sun Java SDK. Alternatively you could update
> >>  > the gcj runtime, but that is potentially painful and will likely break
> >>  > other Java related things.
>
> >>  Same for Fedora 7 and Fedora 8! Jmol does work from the command line of 
> >> Sage.
>
> >>  In Fedora 7 I never saw more the a blue rectangle. In Fedora 8 there is
> >>  more to see, but never the intended plot.
>
> > Did you try using the Sun Java SDK?
>
> I have Sun Java SDK as a plugin in my web browsers. So I can use jmol
> from the notebook!
>
> > If jmol never works with gcj, I propose that when calling to jmol
> > (esp. from the command line where we have control), we check
> > if gcj is going to be used, and if so, just give an error and tell
> > people they have to install the Sun Java SDK or should use
> > viewer='tachyon' instead.
>
> That would be better than having hanging processes :)
>
> > Also, we could lean on (or become part of)
> > the jmol developers and ask them to support gcj...
>
> For the ease of use for Fedora users this would be great.

Yes. Note that Debian's gcj from testing works with jmol, so it is
probably some small stuff that needs to be fixed on jmol's end.

> Jaap

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: 3d plotting (jmol) problems

[mabshoff]

On Apr 2, 6:17 pm, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> Dear all,
>
> I've been having some problems with 3D plots in Sage 2.11.
>
> The machine is a dual (64 bit) Xeon with 16GB of RAM, running Fedora
> Core 6.
>
> boole[~]$ uname -a
> Linux boole.math.utk.edu 2.6.22.14-72.fc6 #1 SMP Wed Nov 21 14:10:25
> EST 2007 x86_64 x86_64 x86_64 GNU/Linux
>
> I installed the binary (64bit debian), and was having problems.  When
> I first opened the notebook with a 3D graph, it froze firefox.  So, I
> tried:
>
> sage: var('x y');
> sage: parametric_plot3d((x,y,x^2*y^3),(x,-2,2),(y,-2,2))
>
> from the command line.  "top" showed java taking 100% of one of the
> CPUs for about 40 seconds, the Jmol logo appeared briefly, the java
> process would go away, but no window would come out and I got no error
> messages.  (I tried this a few times with different functions, always
> with the same result.)  "make test" gave only one error (sage -t
> devel/sage-main/sage/stats/test.py).
>
> I decided to build from source, which went succesfully and "make test"
> had no errors.  I have the same problem, but now I get:
>

> Best regards,
>
> Luis

FC6 ships gcj instead of the Sun Java SDK. Recent version of gcj do
mostly work with jmol, but FC6's gcj is known to be broken. I would
recommend to install the Sun Java SDK. Alternatively you could update
the gcj runtime, but that is potentially painful and will likely break
other Java related things.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: SAGE 2.11 and clisp-2.41.p13

[mabshoff]

On Apr 1, 6:12 pm, "John P. Burkett" <[EMAIL PROTECTED]> wrote:
> Hi Michael,
>
> Thanks for your swift response.
>
> mabshoff wrote:
>
> > Please gzip up install.log and post a link here so that I can take a
> > look. My magic power of deduction is in the shop, so I need to take a
> > look at the logs ;)
>
> My install.log is available 
> athttp://www.uri.edu/artsci/ecn/burkett/install.log.gz
>
> Best regards,
> John
>

Hi John,

looking at the log I cannot fine the clisp failure. It seems that you
upgraded from sage-2.8.2. Was that a direct upgrade from 2.8.2. or did
you do intermediate upgrades?

If you did upgrade straight up from 2.8.2 I would highly recommend to
build Sage from scratch in a different directory. We do test upgrades
from each release the the next one, but jumping over 15+ releases
tends to make things a little unpredictable. The rule of thumb is that
x.y.z->x.y.z+n works, but x.y.?->x.y+n.? has a tendency to cause
trouble.

>
>
> >> Near the end of my lengthy
> >> /usr/local/sage-2.8.2-debian32-i686-Linux/install.log
> >> the following, possibly relevant, lines appear:
> >> "Setting RHOMES to  ['/usr/lib/R']
> >> Traceback (most recent call last):
> >>    File "setup.py", line 106, in 
> >>      RVERSION = rpy_tools.get_R_VERSION(RHOME, force_exec=True)
> >>    File
> >> "/usr/local/sage-2.8.2-debian32-i686-Linux/spkg/build/rpy-1.0.1.p1/src/rpy_tools.py",
> >> line 99, in get_R_VERSION
> >>      " `%s'.\n" % rexec )
> >> RuntimeError: Couldn't execute the R interpreter `/usr/lib/R/bin/R'.
> >> Error building RPY -- Python interface to R.
> >> real    0m0.221s
> >> user    0m0.066s
> >> sys     0m0.018s
> >> sage: An error occurred while installing rpy-1.0.1.p1"
>
> >> The file /usr/lib/R/bin/R exists.  When I type "R", R starts.  I don't
> >> why the system "couldn't execute the R interpreter."

Things somehow go wrong here:

RHOMES= []
DEBUG= True
Setting RHOMES to  ['/usr/lib/R']
Traceback (most recent call last):
  File "setup.py", line 106, in 
RVERSION = rpy_tools.get_R_VERSION(RHOME, force_exec=True)
  File "/usr/local/sage-2.8.2-debian32-i686-Linux/spkg/build/
rpy-1.0.1.p1/src/rpy_tools.py", line 99, in get_R_VERSION
" `%s'.\n" % rexec )
RuntimeError: Couldn't execute the R interpreter `/usr/lib/R/bin/R'.

Sage's R interpreter wouldn't be in /usr/lib. I am clueless to what is
going on here. Can you check if you shell environment contains any
reference to R? Let me think about this problem for a little while,
maybe I will come up with something.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: SAGE 2.11 and clisp-2.41.p13

[mabshoff]

On Apr 1, 5:13 pm, "John P. Burkett" <[EMAIL PROTECTED]> wrote:

Hi John

> Trying to upgrade to SAGE 2.11 on a Dell with an
> i686 Intel Pentium D chip and a CPU running at 2.80 GH under Gentoo
> Linux,  I (as root) did "sage -upgrade".  Near the end of the process, a
> message appeared saying that the clisp executable was not installed. The
> message also stated the the following:
> "If you already have clisp, you can type touch spkg/installed/clisp-2.38
> (or whatever the current version is) from SAGE_ROOT, and continue the
> install." I tried doing "touch spkg/installed/clisp-2.41" and repeating
> "sage -upgrade" but the problem persisted.  The message that appeared on
> this second attempt, similar or identical to the first message, reads in
> part as follows:
> "/usr/bin/install: cannot stat `clisp.pdf': No such file or directory
> make[1]: *** [install-man] Error 1
> make[1]: Leaving directory
> "/usr/local/sage-2.8.2-debian32-i686-Linux/spkg/build/clisp-2.41.p13/src/src'
> Something went wrong -- the clisp executable was not installed.
> If you already have clisp, you can type touch spkg/installed/clisp-2.38
> (or whatever the current version is) from SAGE_ROOT, and continue the
> install.  This tells SAGE that you already have clisp-2.38 installed.
> Clisp is only used by the Maxima component of SAGE.
> real    3m56.837s
> user    3m25.190s
> sys     0m30.153s
> sage: An error occurred while installing clisp-2.41.p13"

Please gzip up install.log and post a link here so that I can take a
look. My magic power of deduction is in the shop, so I need to take a
look at the logs ;)

> Near the end of my lengthy
> /usr/local/sage-2.8.2-debian32-i686-Linux/install.log
> the following, possibly relevant, lines appear:
> "Setting RHOMES to  ['/usr/lib/R']
> Traceback (most recent call last):
>    File "setup.py", line 106, in 
>      RVERSION = rpy_tools.get_R_VERSION(RHOME, force_exec=True)
>    File
> "/usr/local/sage-2.8.2-debian32-i686-Linux/spkg/build/rpy-1.0.1.p1/src/rpy_tools.py",
> line 99, in get_R_VERSION
>      " `%s'.\n" % rexec )
> RuntimeError: Couldn't execute the R interpreter `/usr/lib/R/bin/R'.
> Error building RPY -- Python interface to R.
> real    0m0.221s
> user    0m0.066s
> sys     0m0.018s
> sage: An error occurred while installing rpy-1.0.1.p1"
>
> The file /usr/lib/R/bin/R exists.  When I type "R", R starts.  I don't
> why the system "couldn't execute the R interpreter."
>
> Any help in resolving this problem would be greatly appreciated.

R isn't easily relocatable. I did fix the issue on OSX since I was
under the impression that it works on Linux. I can fix it in an analog
way on Linux.

> Best regards,
> John

Cheers,

Michael

> --
> John P. Burkett
> Department of Environmental and Natural Resource Economics
> and Department of Economics
> University of Rhode Island
> Kingston, RI 02881-0808
> USA
>
> phone (401) 874-9195
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage on Mac - an Odyssey

[mabshoff]

On Mar 31, 7:43 pm, dwb <[EMAIL PROTECTED]> wrote:
> I'm so sorry - my big mistake was to not continue reading in the
> readme file. I was so stuck on step 3 that I never noticed that step 4
> was exactly what I needed to do. Not only that, but just now I didn't
> even make it all the way through step 4 - I browsed a bit for
> Terminal, then used the search button to find Terminal instead of
> following the steps in step 4 to find it.

Ok, I am glad that your problem turned out to be easily solvable.

> Watching the video again, I noticed that it does eventually pan to the
> whole dialog and briefly mention it - I don't know how I missed that
> before. I never got this dialog, perhaps it does not appear on OS X.4.
> However, the operator pushes the Open button on the dialog and
> immediately gets a Terminal window running sage - this probably
> happens because he has run sage on the machine before, so it has
> already been configured to run with Terminal. This is what was so
> misleading to the first time user. Before redoing the video, the
> operator should clear the association between sage and Terminal so
> that the dialog mentioned in step 4 will appear.

Yeah, that seems like a rather important fact :) I wasn't involved in
making the video and I have actually never seen it [I don't do
flash ;)]. So whoever did, please consider redoing it.

> dwb
>

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage on Mac - an Odyssey

[mabshoff]

Hi,

rereading your initial email and thinking about the problem is the
following:

When I double-click on the
sage icon, instead of getting the dialog only half visible and glossed
over in the video, I get one that says: There is no default
application specified to open the document "sage". (Cancel) (Choose
Applicaton...)

It clearly states in the readme:

3) Use finder to visit the sage folder you just copied it and double
click on the "sage" icon.
4) Select to run it with "Terminal":
 Choose Applications, then select "All Applications" in the
 "Enable:" drop down.  Change the "Applications" drop down
 to "Utilities".  On the left, scroll and select "Terminal".
 Click "Open", then in the next dialog select "Update".

Sage will not run properly if you do not do that. That is exactly the
steps you have to take on OSX 10.4 as well as OSX 10.5 - I just
verified it.

On Mar 31, 6:59 pm, dwb <[EMAIL PROTECTED]> wrote:
> Perhaps the video only applies to OS X.5, in which case it should be
> so labeled.

Well, the differences with those dialogs between OSX 10.4 and 10.5 are
miniscule and limited to the new default look in OSX 10.5.

> Perhaps the dialog box which appears in the video should be brought
> fully into view and discussed rather than glossed over.
> Perhaps the video could be made more encompassing, since currently it
> seems to assume things will just work, which they may under OS X.5,
> but not in my case.

They generally do work. The majority of Sage users do develop on Apple
hardware. If the readme were wrong I am sure somebody would have
pointed that out by now.

> Perhaps you would like to actually test the OS X.4 binary on OS X.4 in
> order to update the video and readme.

*Every* Sage release is tested on OSX 10.4. How do you think we create
the binaries.

> Perhaps you would like to try to fix the specific problem I
> encountered, for which I stll have no direct solution, that the icon
> doesn't seem to be runnable.
> Perhaps the video and readme just need a clarification for OS X.4.
> Perhaps as a newbie user of OS X and very rusty unix user I merely
> forgot about something that should have been in the video and readme.
> Perhaps an effort could be made to document what has not been tested
> yet so that volunteers could know what needs to be tested, and newbies
> could steer clear.

Well, we are talking about a computer algebra system here, not a some
little program so that grandma can print pictures of her
grandchildren. A certain amount of computer knowledge is required.

> Perhaps older, but more stable and more fully tested versions could be
> made available as options to those less able or willing to debug.

The is no such thing as older, more tested release. While the quality
of new releases does vary it is generally less buggy than the previous
version. We test very, very extensively (50,000+ test inputs) on
eighteen different configurations and every bug fixed is actually
verified to work in the release. No patch is merged without the
accompanying test case.

> From my limited experience, I would add to the readme: "If when you
> click the sage icon, it doesn't know how to run, instead search for
> unix in the Finder help in order to bring up a Terminal window, into
> which type: /applications/sage/sage". But, is that really what you
> want to say to Mac users? I mean, it's a work around, but isn't there
> a fix?
>
> A recent post to this list about creating an icon on the desktop that
> would run sage did for me create a desktop icon, but the icon does not
> directly run sage.
> I used TextEdit to see that the XML of the icon does contain the
> string /applications/sage/sage, but for some reason it doesn't launch
> sage.
> It does however give me easy access to a Terminal window, into which I
> can type: /applications/sage/sage, which does work.
>
> Also, my wife in an effort to help tried to install each of the two
> binaries currently available for her Power PC based Mac, and got a
> corrupted file error each time.

Well, that isn't really out fault?

> dwb
>

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage on Mac - an Odyssey

[mabshoff]

Hi,

> >  It is unclear to me what you want us to do? Could you clarify?
>
> I think he is suggesting to whoever creates the Mac binaries to modify
> the README file
> to make it easier to use for the first-time user. Maybe the poster could be 
> more
> specific about the exact words he would suggest?
>

Yes, I got that much. But what *specifically* would you suggest to
change or add?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: how to set coding to utf-8 in sage script

[mabshoff]

On Mar 31, 12:30 pm, vakaras <[EMAIL PROTECTED]> wrote:
> > The patch has been merged in Sage 2.11 which should be out today.
> > Alternatively you can download  the patch, cd into $SAGE_ROOT/devel/
> > sage, do an "hg import ../foo/bar.patch", cd back into $SAGE_ROOT and
> > do a "./sage -b". Assuming you have a compiler installed it should
> > work, especially if you compiled from scratch yourself.
>
> Now in file sage/devel/sage/sage/misc/interpreter.py line 375 is
> out.write('# -*- encoding: utf-8 -8-\n')
> but this doesn't helped.
>

Ok, I think I figured out what the problem is:

a) print "klaidelė" from inside Sage 2.10.4 or higher via command
line: works

b) running "sage foo.py" where foo.py consits of 'print "klaidelė"':

  File "foo.py", line 1
SyntaxError: Non-ASCII character '\xc4' in file utf8.py on line 1, but
no encoding declared; see http://www.python.org/peps/pep-0263.html for
details
sage: Error running utf8.py using Python
[EMAIL PROTECTED]:/scratch/mabshoff/release-cycle/sage-3.0.alpha0$ sage
utf8.py
  File "utf8.py", line 1
SyntaxError: Non-ASCII character '\xc4' in file utf8.py on line 1, but
no encoding declared; see http://www.python.org/peps/pep-0263.html for
details
sage: Error running utf8.py using Python

c) running foo.py as in_
[start foo.py]
# -*- encoding: utf-8 -8-
print "klaidelė"
[end foo.py]
works:
sage foo.py
klaidelė


So, is (c) a bug in Sage or shut it be the users job to set the
encoding? Alternatively we could check if any encoding is set and if
not default to utf-8.

Thought?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage on Mac - an Odyssey

[mabshoff]

On Mar 30, 5:07 pm, dwb <[EMAIL PROTECTED]> wrote:
> I'm using a Mac Book running OS X.4.11. I downloaded sage-2.10.3-
> osx10.4-intel-i386-Darwin.dmg, then followed the instructions in both
> the readme file and the video tutorial. When I double-click on the
> sage icon, instead of getting the dialog only half visible and glossed
> over in the video, I get one that says: There is no default
> application specified to open the document "sage". (Cancel) (Choose
> Applicaton...)
>
> After searching through the forums, I tried renaming the icon to
> sage.bin, which caused stuffit to estimate hundreds of thousands of
> hours to unpack it, then realize it couldn't do anything when I tried
> to move the dialog box. I tried to run sage from the Terminal, but it
> just says "-bash: sage: command not found". I tried renaming the icon
> to sage.app - still no luck. I dredged up some ancient unix knowledge
> and tried chmod, man chmod, got stuck in less, finally got out with
> the idea that h might give me help, noticed somewhere that ls had an
> option to show flags, and was able to list the flags on the sage icon
> to show all the executable flags are set. Then I noticed somewhere
> that an explicit path placed before an executable caused it to be run
> from that path, so I tried: . sage
> **
> You must compile SAGE first using 'make' in the SAGE root directory.
> (If you have already compiled SAGE, you must set the SAGE_ROOT
> variable in
> the file '-bash').
> **
> [Process exited - exit code 1]
>
> At this point I could not figure out how to get control of the
> terminal window, and so had to dismiss it.
>
> I searched in vane for the file -bash.
>
> I normally run Windows XP on this Mac using VMWare Fusion, and had
> first considered installing the Windows version of Sage, but on
> reading that it would be running under VMWare, it seemed to me better
> to run it directly under OS X. At this point, I was beginning to
> reconsider.
>
> Going to sage-support, I didn't find anything close to this problem,
> so I started writing this message.
> After recreating the above examples, I launched a new Terminal window
> and tried: ./sage
> --
> | SAGE Version 2.10.3, Release Date: 2008-03-11                      |
> | Type notebook() for the GUI, and license() for information.        |
> --
> The SAGE install tree may have moved.
> Regenerating Python.pyo and .pyc files that hardcode the install PATH
> (please wait at most a few minutes)...
> Please do not interrupt this.
> Setting permissions of DOT_SAGE directory so only you can read and
> write it.
>
> sage:
>
> Please investigate and update the tutorial video and readme file so
> others won't be so encumbered.

It is unclear to me what you want us to do? Could you clarify?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: SAGE-2.10.4 Install Problems

[mabshoff]

On Mar 31, 7:24 am, Daryl Hammond <[EMAIL PROTECTED]> wrote:

> Problem 3 Documentation build
>
> Finally I build the documentation with:
> cd sage-2.10.4/devel/doc
> ./rebuild
>
> No errors are indicated in the build process.  However, the special
> images are missing from the documetation when viewed with Firefox.
> Looking at the folders, the image files for /const, /ref, and /tut are
> named img**.old when img**.png is expected.
>
> Issuing  rename .old .png *.old
> for the files in /const, /ref, and /tut circumvents this.
>
> 

I is generally unnecessary to build the documentation since it is
build before releasing the source tarball. We are not aware of any
problem related to the images [for example 
http://sagemath.org/doc/html/const/node14.html
is just fine] as Mike Hansen pointed out, so what TeX distrubution do
you use? Are you sure rebuild did actually work and not die before
finishing? At least with TeX Live has some problems with left
over .aux and .toc cache files from the TeTeX distribution we use -
see http://trac.sagemath.org/sage_trac/ticket/2679

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: how to set coding to utf-8 in sage script

[mabshoff]

On Mar 30, 5:10 pm, Lars Fischer <[EMAIL PROTECTED]>
wrote:
> Hello,
>
> On Mar 30, 4:28 pm, vakaras <[EMAIL PROTECTED]> wrote:> Hello,
>
> > How to set coding in "*.sage" script to utf-8 ?
> > I have a script:
> > # coding: utf-8
> > print "klaidelė";
>
> there is a patch:http://trac.sagemath.org/sage_trac/ticket/2593
>
> and sage should be able to import this patch somehow. I tried it like:
>
> sage: hg_sage.import_patch("http://trac.sagemath.org/sage_trac/
> attachment/ticket/2593/trac_2593.patch")
> Attempting to load remote 
> file:http://trac.sagemath.org/sage_trac/attachment/ticket/2593/trac_2593.p...
> Loading: [.applying /home/lf/.sage/temp/velocity/11285/tmp_0.patch
> abort: no diffs found
> 
>
> But it didn't work. Probaply the others will tell you how to apply the
> patch to your sage version. Perhabs you have to use the linux patch
> command.

Hmm, it seem that the apply patch from trac patch is broken. Can
somebody verify this?

The patch has been merged in Sage 2.11 which should be out today.
Alternatively you can download  the patch, cd into $SAGE_ROOT/devel/
sage, do an "hg import ../foo/bar.patch", cd back into $SAGE_ROOT and
do a "./sage -b". Assuming you have a compiler installed it should
work, especially if you compiled from scratch yourself.

Cheers,

Michael

> With best regards,
> Lars Fischer
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Trac Guidelines are now in the Wiki

[mabshoff]

Hello folks,

there have been a large, nebulous set of rules regarding how things
are done in trac, patch review and merging and the Sage development
process in general. Now I finally took the time to clear those up and
I put a *draft* of the guidelines up at

http://wiki.sagemath.org/TracGuidelines

What I wrote there is obviously not the final version and none of the
rules are written in stone. I would actually like a discussion about
"the rules" to clear any potential issue where things are either wrong
or unclear. I consider what I wrote  down the consensus from having
done releases for the last four months, but even I do make mistakes ;)

I am working on additional sections on workflow and so on. I believe
that eventually the content of that Wiki page should go into the
documentation or that at least a link from the official doc should
point to that wiki page.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage in Windows

[mabshoff]

On Mar 28, 5:26 pm, "Fabio Tonti" <[EMAIL PROTECTED]> wrote:
> I really don't know much about vmware, but could this URL 
> (http://communities.vmware.com/thread/55242) help? It describes sharing files
> with vmware using simple network folders.

Yeah, shared folders should work. Another possibility would be to use
something like a virtual sftp directory. One commercial offering is

http://www.southrivertech.com/products/webdrive/index.html

Another one is http://www.sftpdrive.com/

Since I do not really use Windows these days there might be something
equivalent from the OS scene.

Cheers,

Michael

> Another possible way: use a usb-pendrive. Could this work?
>
> On Fri, Mar 28, 2008 at 5:16 PM, Chris Kurth <[EMAIL PROTECTED]> wrote:
>
> > I'm trying to use the current version of Sage in Windows (I was using
> > an older version in cygwin). I'm using the VMware image and putty to
> > access it with ssh. Everything works fine except I'd like to be able
> > to edit my code directly from Windows (I'm using IDLE), and I don't
> > know an easy way to transmit text files to the virtual machine. Is
> > there a simple automated way to do this? I figured out how to use wget
> > to download files from my webspace, but that was awkward.
>
> > I realize my question isn't about Sage, per se...
>
> > -Chris Kurth
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: HELLO

[mabshoff]

On Mar 28, 11:19 am, khresta <[EMAIL PROTECTED]> wrote:
> My install.log

Hi,

what you pasted below isn't your install.log, but it seems to be the
configure log from GAP. I also don't see what your question is. Did
GAP fail to build?



Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: no sage.bin

[mabshoff]

On Mar 28, 5:04 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Thu, Mar 27, 2008 at 8:36 PM, toothpaste <[EMAIL PROTECTED]> wrote:
>
> >  Please find below links to the install.log and the output of uname -a
>
> >  http://syntaxthug.syntaxpolice.org/~ralf/install.log
> >  http://syntaxthug.syntaxpolice.org/~ralf/uname.txt
>
> >  I think that I have enough room for an install.
>
> >  The first install failed for lack of space and then I wrote to the
> >  host and the granted me a temporary surplus of memory.
>
> Hi Toothpaste,
>
> Can you try a 100% complete fresh install from source with at
> least 2GB free disk space?  Then report back about whether
> or not that works?
>
> William

Hi Toothpaste,

According to the log it isn't disk space, but memory:

virtual memory exhausted: Cannot allocate memory
make[3]: *** [htconst_n.o] Error 1
make[3]: Leaving directory `/home/ralf/sage/sage-src/sage-2.10.4/spkg/
build/eclib-20080127.p1/src/qcurves'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/ralf/sage/sage-src/sage-2.10.4/spkg/
build/eclib-20080127.p1/src'
E

Note that building Sage requires about 700 MB of RAM (physical+swap)
since gcc and templates eats copious amounts of RAM. The high point in
RAM usage is the LinBox wrapper.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mysterious problem.

[mabshoff]

On Mar 27, 4:55 pm, Michel <[EMAIL PROTECTED]> wrote:
> Hi Michael,

Hi Michael,

> I feel embarrassed causing so much trouble While answering your
> email I finally solved the problem
> (which was my fault of course). I turned out that I had a stray libstdc
> ++.so.6 in my home directory which somehow
> Singular picked up (don't know why).

Ahh, that is the best result :) - I am just a little surprised
LD_LIBRARY_PATH does include your home directory.

> The reason why I had this stray libstdc++.so.6 was that earlier I had
> experienced a similar problem with Singular
> while installing an Athlon sage binary. Googling pointed me to a reply
> of William to someome experiencing the
> same problem. William suggested trying an Athlon specific version of
> libstdc++.so.6. This did not work but I forgot to delete the
> downloaded libstdc++.so.6

Ok. Many times it is something simple like that.  And if it is any
conciliation you aren't the first peson on the list who made that
mistake. In the end we all learn from our mistakes ;)

> Thanks again for your help,
> Michel

Anytime.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mysterious problem.

[mabshoff]

On Mar 27, 7:03 am, Michel <[EMAIL PROTECTED]> wrote:
> "which sage" gives /usr/bin/sage which is the usual script which sets
> SAGE_ROOT (/usr/local/share/sage-2.10.4 in this case) and
> calls /usr/local/share/sage-2.10.4/local/bin/sage-sage.
>
> Things like
>  env SAGE_ROOT=/usr/local/share/sage-2.10.4 /usr/local/share/
> sage-2.10.4/local/bin/sage-sage -singular
> give the error message reported earlier

You should run "cd SAGE_INSTALL_DIR; source local/bin/sage-env"

Also:  [snip]/sage-2.10.4/local/bin/sage-sage -singular does not work.
Call it via the sage executable.

> Singular-3-0-4: /lib/libgcc_s.so.1: version `GCC_4.2.0' not found
> (required by l ibstdc++.so.6)

What compiler do you use? Is that the default compiler shipped with
the system and if not is it in $PATH and $LD_LIBRARY_PATH.

Also: run "ldd local/bin/Singular-3-0-4" and post the output here
please.


> unless executed from /usr/local/share/sage-2.10.4. Afterwards the
> command continues to work in in the same shell, even in another
> directory.
>
> How does sage finds it .so files? Because I think that is the problem.

Via LD_LIBRARY_PATH, but that is not set permanently unless you source
local/bin/sage-env from $SAGE_ROOT.

> Michel

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.10.4 on opensuse 10.3

[mabshoff]

On Mar 26, 5:18 pm, Florin <[EMAIL PROTECTED]> wrote:
> Thanks everybody for your help.
>
> Following the advice here, these are the solutions that worked out for
> me:
>
> 1,2. Eliminating the Intel compilers from the path resulted in a clean
> build of sage, all the tests were successful..

Ok, the issue has been fixed in 2.11 and should no longer happen. I
consider that an SCons bug. Amazingly the SCons release from Dec. 2007
does not support gfortran nor g95!

> 3. removing the .aux and .toc cache files from the documentation area
> solved the pdf/html problems.

ok, we ought to make sure that we remove all those temp files before
packaging the documentation.

> Florin

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: no sage.bin

[mabshoff]

On Mar 26, 4:39 pm, toothpaste <[EMAIL PROTECTED]> wrote:
> I have successfully installed sage on my debian box and I am so
> excited to use it.

What exactly did you do? Install from source? Install a binary? If so
which binary? What arch are you on?

> But it can't seem to find the sage.bin.  Here are the errors I got
> when I tried to run sage.
>
> /usr/bin/env: sage.bin: No such file or directory
> /home/ralf/sage/sage-src/sage-2.10.4/local/bin/sage-sage: /home/ralf/
> sage/sage-src/sage-2.10.4/local/bin/sage-ipython: sage.bin

Does that happen when you run "./sage" also?

> Also, when I type "which sage.bin" at the prompt, I get no response.
> Please advise.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Handling Octave functions with multiple return values?

[mabshoff]

On Mar 24, 4:26 pm, Michael <[EMAIL PROTECTED]> wrote:
> What is the recommended way to handle Octave functions with multiple
> return values in Sage?
>
> Like Magma, Octave functions can return multiple values.  Working with
> such Magma functions is documented in the Sage documentation (http://
> modular.math.washington.edu/sage/doc/html/ref/node125.html), but I
> cannot find similar documentation for Octave.
>
> Here is how it works in Octave -- foo is a function that returns two
> values:
>
> octave:4> function [a b] = foo() a=1; b=2; endfunction
> octave:5> foo()
> ans = 1
> octave:6> [c,d] = foo()
> c = 1
> d = 2
>
> This code transliterated into Sage is shown below.   What is the
> correct way to do this?

I am not sure, but we currently do not test the Octave interface
regularly since it is optional. What we need is to extend the optional
doctest framework to recognize when a given optional package is
installed and then test that code. It is very likely that I will run "-
long -optional" in the future instead of "-optional" now.

To get back to your original problem: We need more Octave users that
report issues with the current interface. That might be a little
painful in the short term, but in the long term the result should be
much better.

Cheers,

Michael

>
> --
> | SAGE Version 2.10.3, Release Date: 2008-03-11                      |
> | Type notebook() for the GUI, and license() for information.        |
> --
>
> sage: octave.eval("function [a b] = foo() a=1; b=2; endfunction")
> ''
> sage: octave.foo()
>  1
> sage: [c, d] = octave.foo()
> ---
>    Traceback (most recent call
> last)
>
> /home/login/ in ()
>
> /usr/local/sage/local/lib/python2.5/site-packages/sage/interfaces/
> expect.py in __iter__(self)
>    1013
>    1014     def __iter__(self):
> -> 1015         for i in range(1, len(self)+1):
>    1016             yield self[i]
>    1017
>
> /usr/local/sage/local/lib/python2.5/site-packages/sage/interfaces/
> expect.py in __len__(self)
>    1017
>    1018     def __len__(self):
> -> 1019         raise NotImplementedError
>    1020
>    1021     def __reduce__(self):
>
> :
> sage:
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.10.4 on opensuse 10.3

[mabshoff]

On Mar 24, 1:38 pm, Florin <[EMAIL PROTECTED]> wrote:
> Hi,

Hi Florin,

> Several problems I encountered while building sage-2.10.4 on opensuse
> 10.3:
>
> 1. make fails because the scons tool is hardwired to use -no_archive
> option for ld.
> This option is not valid for the version of ld in openSuSE 10.3.
> Simple workaround: do a first build that fails, remove -no_archive
> flag
> from local/lib/scons-0.97/SCons/Tool/ifort.py, then redo the build
> that
> will finish successfully.

The issue here is that you have the Intel Fortran compiler installed.
For some stupid reason SCons selects that as the default fortran
compiler regardless whether gfortran or g95 is installed. I thing it
is time that we fix that issue once and for all by teaching SCons not
to use the Intel compiler.  That is #1618. I will likely fix that
today since it has come up over and over again.

> 2. "make test" fails on most tests with the error:
>  ImportError: /usr/local/sage-2.10.4/local/lib/libcsage.so: undefined
> symbol: __dso_handle

This is likely caused by the Intel Fortran compiler.

> 3. "make html" and "make pdf" fail because [EMAIL PROTECTED] cannot be 
> handled.

Got any more details?

> I understand that some of these errors may be because of various bugs
> or unimplemented features in opensuse 10.3, still some decent
> workarounds would be highly appreciated.
> Thank you very much,

Not sure what the problem could be in this case, but my guess is that
some essential TeX macro package is missing. I remember the days when
SuSE installed a decent TeX per default, but those days are long gone
and it has been a while since TeX is installed per default at all.

> Florin Manolache

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mixed 32/64 bit .o files under Linux PowerPC

[mabshoff]

On Mar 9, 2:43 am, Sameer <[EMAIL PROTECTED]> wrote:
> William, Michael,
>   I re-installed gap by compiling it in 32 bit mode and that got rid
> of the error.

Good, that is a workaround - at least until someone wants to run GAP
in 64 bit mode :)

> gap, it seems doesn't support 64 bit mode. Now, I am
> dealing with some startup errors such as:
>  /usr/local/PET/src/build/64/sage-2.10.2/local/lib/liblinboxwrap.so.0:
> undefined symbol: _gfortran_st_write

This looks like libgfortran.so is not in $SAGE_LOCAL/lib

> and if I can manage to get rid of this, I see others such as:
> import sage.matrix.matrix_space
>   File "/usr/local/PET/src/build/64/sage-2.10.2/local/lib/python2.5/
> site-packages/sage/matrix/matrix_space.py", line 42, in 
> import matrix_integer_dense
> ImportError: /usr/local/PET/src/build/64/sage-2.10.2/local/lib/
> libiml.so.0: undefined symbol: cblas_dswap
>

Did the ATLAS build work? GSL itself might not be enough.

>
> Is there any site specify file where I can turn off the default
> loading of some of these libraries so I can see what I need to re-
> compile? The problem with ld being used in several packages is that it
> expects 32 bit code. I couldn't find a ld -m64 option to generate
> shared objects in 64 bits. So, I ended up using gcc -m64 -shared -o
> lib or xlc_r -q64 -G -o lib...  instead. And some of the
> dependencies need to be resolved better.
>
> Thanks,
> - Sameer

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mixed 32/64 bit .o files under Linux PowerPC

[mabshoff]

On Mar 9, 1:48 am, Sameer <[EMAIL PROTECTED]> wrote:
> William,
>Thanks a lot for the
> sage -br
> command. I found that I needed to re-install singular, fplll and then
> it needed qd so I got qd 2.1 initially but it didn't work, so I got qd
> 2.3.5 from LBL and it compiled everything. At the end, sage gave an



> So, it gave me a segfault. This is on an IBM p690 (32x Power4 cpus)
> SuSe Linux system using gfortran/gcc in 64 bits. It has installed
> gap-4.4.10.p2.
> What would you recommend at this stage? Re-install gap since it
> segfaults?

Try recompiling GAP with "-O0". GAP works fine with gcc 4.1 on 32 bit
PPC, otherwise we need to look more deeply into the problem.

Cheers,

Michael



--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mixed 32/64 bit .o files under Linux PowerPC

[mabshoff]

On Mar 8, 9:23 pm, Sameer <[EMAIL PROTECTED]> wrote:
> Hi,
>I made scripts for gcc/g++/gfortran and put them in my ~/bin
> directory. These scripts add -m64 argument to the compilers and so I
> was able to generate 64 bit binaries. After several errors and
> manually intervening (building shared objects etc.) I think it has
> compiled SAGE completely on ppc64 Linux. It said:
>
> Successfully installed jmol-11.5.2.p1
> Now cleaning up tmp files.
> Making SAGE/Python scripts relocatable...
> Making script relocatable
> make[1]: Leaving directory `/usr/local/PET/src/build/64/sage-2.10.2/
> spkg'
>
> real21m30.674s
> user19m19.231s
> sys 1m41.583s
> To install gap, gp, singular, etc., scripts
> in a standard bin directory, start sage and
> type something like
>sage: install_scripts('/usr/local/bin')
> at the SAGE command prompt.
>
> SAGE build/upgrade complete!
>
> But when I try to run it, it says:
> sage-2.10.2> ./
> sage
> --
> | SAGE Version 2.10.2, Release Date: 2008-02-22  |
> | Type notebook() for the GUI, and license() for information.|
> --
> ---
>Traceback (most recent call
> last)
>
> /usr/local/PET/src/build/64/sage-2.10.2/local/bin/ in
> ()
>
> : No module named
> sage.misc.preparser_ipython
> WARNING: Failure executing code: 'import sage.misc.preparser_ipython;
> sage.misc.preparser_ipython.magma_colon_equals=True'
> ---
>Traceback (most recent call
> last)
>
> /usr/local/PET/src/build/64/sage-2.10.2/local/bin/ in
> ()
>
> : No module named sage.misc.misc
> 
>
> Now, I see the files:
> ./devel/sage-main/.hg/store/data/sage/misc/preparser__ipython.py.i
> ./devel/sage-main/sage/misc/preparser_ipython.py
> ./devel/sage-main/build/lib.linux-ppc64-2.5/sage/misc/
> preparser_ipython.py
>
> I tried seeting my PYTHONPATH environment variable to include these
> directories, but it didn't work. How should I proceed?

Without looking at the build log it is next to impossible to determine
what went wrong. You could put those up somewhere on the net and post
a link in this thread. It would be best if it was a build log that
didn't include all the previous attempts.


> I appreciate all your help!

What hardware and distribution are you building this on?

> Thanks!
> - Sameer


Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Mixed 32/64 bit .o files under Linux PowerPC

[mabshoff]

On Mar 6, 1:47 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Wed, Mar 5, 2008 at 4:40 PM, Sameer <[EMAIL PROTECTED]> wrote:
>
> >  William,
> >   Thanks a lot for your suggestion. I had not set it. Some more
> >  problems with the compile on powerpc Linux. I get:
> >  ./lisp.run -on-error appease -B . -N locale -E 1:1 -Efile UTF-8 -
> >  Eterminal UTF-8 -norc -m 1800KW -x "(and (load \"init.lisp\") (sys::
> >  %saveinitmem) (ext::exit)) (ext::exit t)"
> >   i i i i i i i   ooooo   o   o
> >   I I I I I I I  8 8   8   8 8 o  88
> >   I  \ `+' /  I  8 8   8 888
> >\  `-+-'  /   8 8   8  o   8
> > `-__|__-'8 8   8   8  8
> > |8 o   8   8 o 8  8
> >   --+--   o8oo  ooo8ooo   o   8
>
> >  Copyright (c) Bruno Haible, Michael Stoll 1992, 1993
> >  Copyright (c) Bruno Haible, Marcus Daniels 1994-1997
> >  Copyright (c) Bruno Haible, Pierpaolo Bernardi, Sam Steingold 1998
> >  Copyright (c) Bruno Haible, Sam Steingold 1999-2000
> >  Copyright (c) Sam Steingold, Bruno Haible 2001-2006
>
> >  ;; Loading file defseq.lisp ...
> >  ;; Loaded file defseq.lisp
> >  ;; Loading file backquote.lisp ...
> >  ;; Loaded file backquote.lispmake[2]: *** [interpreted.mem]
> >  Segmentation fault
>
> > make[2]: Leaving directory `/usr/local/PET/src/build/sage-2.10.2/spkg/
> >  build/clisp-2.41.p12/src/src'
> >  Silly permissions error with first make of clisp.
> >  Do a 'make' again, since second 'make' works.
> >  Error building clisp.
>
> >  real30m10.186s
> >  user29m17.033s
> >  sys 0m42.184s
> >  sage: An error occurred while installing clisp-2.41.p12
>
> > Please email sage-develhttp://groups.google.com/group/sage-devel
> >  explaining the problem and send the relevant part of
> >  of /usr/local/PET/src/build/sage-2.10.2/install.log.  Describe your
> >  computer, operating system, etc.
> >  If you want to try to fix the problem, yourself *don't* just cd to
> >  /usr/local/PET/src/build/sage-2.10.2/spkg/build/clisp-2.41.p12 and
>
> > type 'make'.
> >  Instead type "/usr/local/PET/src/build/sage-2.10.2/sage -sh"
> >  in order to set all environment variables correctly, then cd to
> >  /usr/local/PET/src/build/sage-2.10.2/spkg/build/clisp-2.41.p12
>
> > (When you are done debugging, you can type "exit" to leave the
> >  subshell.)
> >  make[1]: *** [installed/clisp-2.41.p12] Error 1
>
> > make[1]: Leaving directory `/usr/local/PET/src/build/sage-2.10.2/spkg'
>
> >  real98m24.666s
> >  user95m9.979s
> >  sys 2m30.678s
> >  + '[' all = all -a 2 = 0 ']'
>
> It's a bummer clisp didn't build.   We only use it for maxima, so if
> you can install any version of clisp on your system (e.g., using
> your normal package tools),

I suspect a compiler incompability. An upgrade to clisp 2.44.1 should
fix most if not all of those.

> then sage will use that to build
> Maxima.  There is no binary linking between sage and maxima/clisp,
> however you won't get far in Sage without having maxima on your
> system.
>

Yep.

>
> > build/sage-2.10.2>
>
> >  next. If I try and execute the sage executable at this stage, I get:
> >  build/sage-2.10.2> /usr/local/PET/src/build/sage-2.10.2/sage
> >  --
> >  | SAGE Version 2.10.2, Release Date: 2008-02-22  |
> >  | Type notebook() for the GUI, and license() for information.|
> >  --
> >  Setting permissions of DOT_SAGE directory so only you can read and
> >  write it.
>
> >  init2.c:37:  assertion failed: ((64 - 0)+0) == (((64 - 0)+0)/8) * 8 &&
> >  sizeof(mp_limb_t) == (((64 - 0)+0)/8)
> >  /usr/local/PET/src/build/sage-2.10.2/local/bin/sage-sage: line 212:
> >  13965 Segmentation fault  sage-ipython -c "$SAGE_STARTUP_COMMAND;"
> >  "$@"
>
> That's gmp, since init2.c is:
>
>   gmp-4.2.1/mpz/init2.c
>
> It seems that something is miscompiled about your gcc.  I'm sure this has
> something to do with 64/32-bit issues.

Well, the problem is that the gcc on the system produces 32 bit
binaries per default while GMP usually defaults to the arch with the
largest wordsize. All the issues are (easily) fixable, but it just
takes time to get fixes like that merged for a rather uncommon
platform. Since I can only build on a PS3 things also tend to build
slowly, so if anybody has an account on a 64 bit PPC box let me know.

> Which Linux distribution are you using?   Maybe you could switch to Debian,
> which might be saner than whatever you are currently using.
>

I disagree here :)

Cheers,

Michael

>
> > build/sage-2.10.2>
>
> >Thanks,
> >- Sameer
>
> >  On Mar 5, 2:42 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
>
> > > On Wed, Mar 5, 2008 at 1:45 PM, Sameer <[EMAIL PROT

[sage-support] Re: FreeBSD Linux Emulation

[mabshoff]

Hi,

On Mar 6, 5:31 am, "bryan newbold" <[EMAIL PROTECTED]> wrote:
> I've had some luck running the sage-2.10.2-linux-32bit-debian-i686-Linux
> binaries under FreeBSD 6.3-RELEASE (manually upgraded) using linux emulation
> (linux_base-fc-4_10 = fedora core 4.10 distribution files); basic calculus
> and arithmetic work from the interactive shell, the notebook runs with 3D
> and Tachyon plotting, jsmath, etc.

Good to know.

> What I did:
>
>  * make sure linux kernel module was compiled into FreeBSD and is loaded
>with kldstat; I also had linprocfs but i'm not sure it's required.
>
>  * install emulators/linux_base
>
>  * as root symbolically link /usr/bin/bash to /bin/bash in freeBSD
>
>  * unzip/tar sage-2.10.2-linux-32bit-debian-i686-Linux.tar.gz, enter dir,
>   edit 'sage' to set SAGE_ROOT
>
>  * run ./sage
>
> I didn't even have to use brandelf!

Good to know. Can you add the above steps to the FreeBSD port page in
the wiki as a work around until all the fixes are merged?

> The biggest problem i'm having now is the same as Petr Muzikar from an old
> sage-newbie 
> post:http://groups.google.com/group/sage-newbie/browse_thread/thread/8bb27...
> 1c08
>
> Basically when trying to plot() from shell or notebook matplotlib throws an
> integer literal error (trace at end of email). I haven't really tried
> debugging this yet; is there a known solution or should I dig in?
>
> I also tried compiling sage-2.10.{1,2} from sources following the wiki
> (http://wiki.sagemath.org/freebsd) and had a variety of issues (atlas,
> lapack, ...); I tried swapping in the lapack i'd compiled previously for
> scipy but had no love. I don't have good debug info about that now, i'll be
> more precise when I have more time (those ATLAS compiles are
> piinful).

That isn't so much a BSD issue, but the case when ATLAS mis-detects or
doesn't know your CPU.

> Would it be appropriate for me to start a freebsd wiki page on the trac?
> Should I submit new tickets for compilation problems/patches/workarounds or
> just glom on to #1638?

Nope, that is a separate issue and just happens to hit us mostly on
*BSD - it can also happen on Solaris for example. The rule in trac is
that one ticket is only about on issue (Except META ticket, but that
is a different story). I have fixes for all the FreeBSD issues and
just need to merge them. My plan for the rest of the month of March is
to get build fixes for Solaris, Linux PPC 64, Intel 10.5 64 bit,
Cygwin and FreeBSD in the tree.

> -bryan
>



I am not sure about the plotting issue, but I assume a natively
compiled version will work.

Cheers,

Michael

--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage 2.10.2: 18575 Alarm clock

[mabshoff]

Hi,

> >  Come again: On Ondrej's box it take more than *6 minutes* to run that
> >  code? Somthing is seriously FUBAR on his box then. He did mention it a
> >  while back in IRC but I cannot believe it is still a problem and he
> >  didn't fix it :)
>
> Evidently.  sage-location opens and closes a *lot* of files, so on a
> slot computer
> it could take a long time.   Maybe he's using NFS over 10Mbps ethernet? :-)

He is using XFS locally IIRC, which boggles the mind.

>  -- William

Cheers,

Michael


--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage 2.10.2: 18575 Alarm clock

[mabshoff]

On Feb 26, 5:43 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Mon, Feb 25, 2008 at 8:39 PM, mabshoff
>
>
>
> <[EMAIL PROTECTED]> wrote:
>
> >  On Feb 26, 5:24 am, "Ondrej Certik" <[EMAIL PROTECTED]> wrote:
> >  > Hi,
>
> >  > I have 64bit Debian and I did:
>
> >  > $ 
> > wgethttp://sagemath.org/SAGEbin/linux/64bit/sage-2.10.2-debian-64bit-inte...
>
> > > $ tar xzf sage-2.10.2-debian-64bit-intel-x86_64-Linux.tar.gz
> >  > $ ln -s sage-2.10.2-debian-64bit-intel-x86_64-Linux/ sage
> >  > $ cd sage
> >  > $ ./sage
> >  > --
> >  > | SAGE Version 2.10.2, Release Date: 2008-02-22  |
> >  > | Type notebook() for the GUI, and license() for information.|
> >  > --
> >  > The SAGE install tree may have moved.
> >  > Regenerating Python.pyo and .pyc files that hardcode the install PATH
> >  > (please wait less than a minute)...
> >  > Please do not interrupt this.
> >  > /home/ondra/ext/sage/local/bin/sage-sage: line 149: 18575 Alarm clock
> >  >"$SAGE_ROOT/local/bin/"sage-location
>
> >  I just grepped through my tree and there is no "Alarm clock" string
> >  anywhere. I also checked the binary you tried and there is no string
> >  in there either. The log in SAGE_LOCAL/bin doesn't indicate any
> >  commits that I don't have and the repo is clean, i.e. no outstanding
> >  changes.
>
> >  So: can anybody else reproduce this?
>
> This is in sage-location:
>
> import signal
> signal.alarm(360)
>
> This is *stupid*.  It's doing a process that really really better not
> get interrupted, and it's
> interrupting itself if it takes more than 6 minutes.  I can't imagine
> what idiot wrote such
> bad code

Come again: On Ondrej's box it take more than *6 minutes* to run that
code? Somthing is seriously FUBAR on his box then. He did mention it a
while back in IRC but I cannot believe it is still a problem and he
didn't fix it :)

> ... oh wait, that was me.  I think that was in there right when that
> code was first written,
> mainly for debugging purposes when I was doing some testing.  Anyway...  
> didn't
> David Harvey say something in irc about all my old code being
> "impatient" -- I guess
> this is a prime example of that!
>
> Anyway, fix up at trac #2311:
>  http://trac.sagemath.org/sage_trac/ticket/2311
>

Cheers,

Michael

>
>
> >  > Setting permissions of DOT_SAGE directory so only you can read and write 
> > it.
>
> >  > sage:
> >  > Exiting SAGE (CPU time 0m0.00s, Wall time 2m2.75s).
> >  > Exiting spawned Gap process.
>
> >  > Notice the "Alarm clock"  line. What does that mean?
>
> >  > Ondrej
>
> >  Cheers,
>
> >  Michael
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage 2.10.2: 18575 Alarm clock

[mabshoff]

On Feb 26, 5:24 am, "Ondrej Certik" <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I have 64bit Debian and I did:
>
> $ wgethttp://sagemath.org/SAGEbin/linux/64bit/sage-2.10.2-debian-64bit-inte...
> $ tar xzf sage-2.10.2-debian-64bit-intel-x86_64-Linux.tar.gz
> $ ln -s sage-2.10.2-debian-64bit-intel-x86_64-Linux/ sage
> $ cd sage
> $ ./sage
> --
> | SAGE Version 2.10.2, Release Date: 2008-02-22  |
> | Type notebook() for the GUI, and license() for information.|
> --
> The SAGE install tree may have moved.
> Regenerating Python.pyo and .pyc files that hardcode the install PATH
> (please wait less than a minute)...
> Please do not interrupt this.
> /home/ondra/ext/sage/local/bin/sage-sage: line 149: 18575 Alarm clock
>"$SAGE_ROOT/local/bin/"sage-location

I just grepped through my tree and there is no "Alarm clock" string
anywhere. I also checked the binary you tried and there is no string
in there either. The log in SAGE_LOCAL/bin doesn't indicate any
commits that I don't have and the repo is clean, i.e. no outstanding
changes.

So: can anybody else reproduce this?

> Setting permissions of DOT_SAGE directory so only you can read and write it.
>
> sage:
> Exiting SAGE (CPU time 0m0.00s, Wall time 2m2.75s).
> Exiting spawned Gap process.
>
> Notice the "Alarm clock"  line. What does that mean?
>
> Ondrej

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: 2.10.2 bessel_I(1,1,"scipy") bug

[mabshoff]

On Feb 25, 9:53 pm, Kate <[EMAIL PROTECTED]> wrote:
> When I build 2.10.2 from source on my
> x86-Linux box (pentium4-fc6) using gcc-4.2.3,
> I get a 'make check' failure at
>
> ./sage -t devel/sage-main/sage/functions/special.py
>
> sh: line 1:  3345 Illegal instruction /home/kate/sage/sage-2.10.2-
> x86-Linux/local/bin/python .doctest_special.py >.doctest/out
> 2>.doctest/err
>
> A mysterious error (perphaps a memory error?) occurred, which may have
> crashed doctest.
>  [3.6 s]
> exit code: 256
>
> The offending line seems to be
>
> sage: bessel_I(1,1,"scipy")
> /home/kate/sage/sage-2.10.2-x86-Linux/local/bin/sage-sage: line 212:
> 3484 Illegal instruction sage-ipython -c "$SAGE_STARTUP_COMMAND;"
> "$@"
>
> Kate

Hi Kate,

this is now #2303. I will check out if I can reproduce this anywhere
and we will hopefully have a fix soon. It is a blocker for 2.10.3.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: error while compiling sage-2.10.2

[mabshoff]

On Feb 24, 5:38 pm, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> On Feb 24, 12:05 pm, JMD <[EMAIL PROTECTED]> wrote:
>
>
>
> > on a linux-x86_64 machine
>
> > Installing c_lib
> > scons: `install' is up to date.
> > Traceback (most recent call last):
> >   File "setup.py", line 1217, in 
> > deps = create_deps(ext_modules)
> >   File "setup.py", line 1208, in create_deps
> > deps_graph(deps, f, visited)
> >   File "setup.py", line 1175, in deps_graph
> > this_deps = search_all_includes(f)
> >   File "setup.py", line 1099, in search_all_includes
> > S = open(filename).readlines()
> > IOError: [Errno 2] No such file or directory: 'sage-main/sage/modules/
> > free_module_element.pyx'
> > sage: There was an error installing modified sage library code.
>
> > ERROR installing SAGE
>
> Hi,
>
> that is cause by #2180. There is a patch for it, but the short
> solution is to touch the pyrex file in question. Then everything is
> back to normal.
>
> Cheers,
>
> Michael

Burcin did pop up in IRC a little while ago and reported the same
error. Neither touching the file, not a "sage -ba" nor applying the
patch from #2180 fixes the problem, so I am fresh out of ideas. It
only seems to happen on updates, i.e. in the other cases I saw the
issue a touching of a single file was sufficient. So there is maybe
more than one bug lurking in that code. Since I have been unable to
reproduce the harder problem yet any input (or even better solution)
is welcome :)

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: error while compiling sage-2.10.2

[mabshoff]

On Feb 24, 12:05 pm, JMD <[EMAIL PROTECTED]> wrote:
> on a linux-x86_64 machine
>
> Installing c_lib
> scons: `install' is up to date.
> Traceback (most recent call last):
>   File "setup.py", line 1217, in 
> deps = create_deps(ext_modules)
>   File "setup.py", line 1208, in create_deps
> deps_graph(deps, f, visited)
>   File "setup.py", line 1175, in deps_graph
> this_deps = search_all_includes(f)
>   File "setup.py", line 1099, in search_all_includes
> S = open(filename).readlines()
> IOError: [Errno 2] No such file or directory: 'sage-main/sage/modules/
> free_module_element.pyx'
> sage: There was an error installing modified sage library code.
>
> ERROR installing SAGE

Hi,

that is cause by #2180. There is a patch for it, but the short
solution is to touch the pyrex file in question. Then everything is
back to normal.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: contour_plot bug

[mabshoff]

On Feb 21, 12:51 am, "David Joyner" <[EMAIL PROTECTED]> wrote:
> I can verify this on an intel mac *and* that
>
> sage: 
> contour_plot(y-1,(-10,10),(-10,10),fill=False,contours=2,plot_points=100)
> sage: 
> contour_plot(y-1,(-10,10),(-10,10),fill=False,contours=1,plot_points=100)
>
> yield the same plot.
>
> On Wed, Feb 20, 2008 at 4:47 PM, Kate <[EMAIL PROTECTED]> wrote:
>
> >  In 2.10.1,
>
> >   sage:  R. = PolynomialRing(QQ,2)
> >   sage:  contour_plot(y-1,(-10,10),
> >  (-10,10),fill=False,contours=1,plot_points=100)
>
> >  plots the line y = -9 rather than the line y=1
>
> >   sage:  contour_plot(-y+1,(-10,10),
> >  (-10,10),fill=False,contours=1,plot_points=100)
>
> >  correctly plots y = 1.
>
> >  Kate

Ok, this is now #2243.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Graph isomorphism problem.

[mabshoff]

On Feb 18, 9:49 pm, Robert Miller <[EMAIL PROTECTED]> wrote:
> Carlo,
>
> Thanks again for reporting this. During Bug Days 10, I found the cause
  ^^^

Cool, email from the future :)

By the way: Jason did mention in IRC today that he was giving a
tentative positive review to #1961. Since a lot of tickets depend on
that could you give a list of ticket that need review and depend on
that ticket? There certainly are 2184-2186.

> of the bug, and posted a patch fixing the problem (there are a few
> dependencies, but once everything gets merged, your example will
> work):
>
> >http://trac.sagemath.org/sage_trac/ticket/1961
>
> -- Robert M

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Error building a new branch

[mabshoff]

On Feb 18, 5:29 pm, kcrisman <[EMAIL PROTECTED]> wrote:
> Thanks again for the helpful comments.  Jason's in particular helped
> me realize a large part of the problem is that I usually cannot do
> serious work during a time frame when I am online, and I didn't know
> to look in the chapter 7 of the developer's tutorial (though there is
> not that much more there than in the mercurial-howto.txt).  I think
> I'll be able to figure it out, though, with some patience. If I have
> specific questions or suggestions to make the process more
> straightforward - I think Jason is right that it does presuppose
> having used a control system before.  But whatever.
>
> Instead, I now think my specific problem is elsewhere, and I would
> really appreciate some help with it if anyone can decipher what
> follows. The immediate result of attempting to make a new branch is
> that I get a message which ends as follows after trying to use sage
> (in that branch, of course):
>
> /Applications/sage/sage
> --
> | SAGE Version 2.10.1, Release Date: 2008-02-02  |
> | Type notebook() for the GUI, and license() for information.|
> --
> ---
>Traceback (most recent call
> last)
>
> 
>
> Not having a prompt is a pretty big issue.  The particular file which
> raised the error message is libcsage.dylib, but presumably this is the
> symptom, not the cause.  Any thoughts on what went wrong?  Now I am
> scared to go back to my main branch, because the last time I tried
> this it ate that branch too, and I would rather not download Sage yet
> again just for this.
>
> -kcrisman
>
> ***
> Error message when cloning:
>
> Now cloning the current SAGE library branch...
> hg clone  sage sage-hackbranch
> 1426 files updated, 0 files merged, 0 files removed, 0 files
> unresolved
> Copying over all Pyrex auto-generated .c, .cpp and .h files...
> Copying over build directory.
> cp -r sage/build sage-hackbranch
> Building sage-hackbranch
> sage -b hackbranch
>
> --
> sage: Building and installing modified SAGE library files.
>
> Installing c_lib
> gcc -o src/convert.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/convert.c
> gcc -o src/interrupt.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/interrupt.c
> gcc -o src/mpn_pylong.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/mpn_pylong.c
> gcc -o src/mpz_pylong.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/mpz_pylong.c
> gcc -o src/stdsage.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/stdsage.c
> gcc -o src/gmp_globals.os -c -O2 -g -fPIC -I/Applications/sage/local/
> include -I/Applications/sage/local/include/python2.5 -I/Applications/
> sage/local/include/NTL -Iinclude src/gmp_globals.c
> g++ -o src/ZZ_pylong.os -c -fPIC -I/Applications/sage/local/include -I/
> Applications/sage/local/include/python2.5 -I/Applications/sage/local/
> include/NTL -Iinclude src/ZZ_pylong.cpp
> g++ -o src/ntl_wrap.os -c -fPIC -I/Applications/sage/local/include -I/
> Applications/sage/local/include/python2.5 -I/Applications/sage/local/
> include/NTL -Iinclude src/ntl_wrap.cpp
> g++ -o libcsage.dylib -single_module -flat_namespace -undefined
> dynamic_lookup -dynamiclib src/convert.os src/interrupt.os src/
> mpn_pylong.os src/mpz_pylong.os src/stdsage.os src/gmp_globals.os src/
> ZZ_pylong.os src/ntl_wrap.os -L/Applications/sage/local/lib -lntl -
> lgmp -lpari
> ld: flag: -undefined dynamic_lookup can't be used with
> MACOSX_DEPLOYMENT_TARGET environment variable set to: 10.1
> /usr/bin/libtool: internal link edit command failed
> scons: *** [libcsage.dylib] Error 1
> ERROR: There was an error building c_lib.

What version of XCode are you using? It is highly recommended to
install the latest supported XCode for 10.4 and XCode 3 for 10.5. and
build Sage from sources if you want to do development. Once you did
the full build you can usually upgrade in place. Even with a binary it
is highly recommended to do a "sage -ba" before cloning. If you did
build from source something went wrong, so we should be given access
to the build logs, i.e. gzip them and upload them somewhere, then post
a link here.

The error about MACOSX_DEPLOYMENT_TARGET being set to 10.1 d

[sage-support] Re: upgrade vs version

[mabshoff]

[EMAIL PROTECTED] wrote:
> Hi,
>
> I have installed sage 2.9.  I ran  "sage -upgrade" on tuesday,
> february the 12. Do I now have the current version, 2.10.1 (dated from
> february the 02)?

Unless the upgrade failed somehow the banner now should say

--
| SAGE Version 2.10.1, Release Date: 2008-02-02  |
| Type notebook() for the GUI, and license() for information.|
--

If it doesn't something went wrong.


> Thx,
>
> JPD

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: symbol lookup error: /usr/lib/libxml2.so.2: undefined symbol: gzopen64

[mabshoff]

Hi,

> Matthias,
>
> I saw the problem too on my Ubuntu 7.10 computers too.  Thanks for the
> motivation to make the time to fix it :).

I do remember complaining about the fact that the notebook is launched
with its own code instead of the new infrastructure. I never filed a
bug report, but I am sure I can find it in my IRC log.

> There are two patches up at the above URL that fix the problem for me
> and should fix the problem for you.  If you have experience applying
> patches and doing development with mercurial, you could try the patches
> and tell us if it fixes the problem for you (one patch is for
> SAGE_ROOT/local/bin/sage-env and the other is for
> SAGE_ROOT/devel/sage/sage/server/misc.py).  If you don't have experience
> with patches and development or don't want to somehow mess up your
> current sage installation, I suggest doing your workaround (opening the
> browser first, then starting the notebook) until 2.10.2.  I think the
> fixes will make it into 2.10.2.

It has been merged and will be in Sage 2.10.2.alpha1, which should be
out "soon", i.e. by Sunday.

> Thanks,
>
> Jason

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Extreme Newbie Development Questions

[mabshoff]

On Feb 16, 11:34 pm, "dean moore" <[EMAIL PROTECTED]> wrote:
> OK, there seems an interest in us relatively new folk contributing code for
> SAGE.
> Admirable.  More than I get from M*cr*s*ft.  Or Apple.  Thanks.  I mean it.
>
> Surfing Wikipedia, I came upon the animated graph at
> , and thought, "Sounds like
> a good project.  Bet I can do that in SAGE" (see attached .gif, not the
> same constants as the Wikipedia image).
>
> I coded, swore, googled, shook my fist at the computer, and posted a few
> questions
> to the Newbie Group that doesn't seem to exist anymore.

We want to channel traffic into two mailing lists: sage-support for
everything that isn't an developer issue and sage-devel for developers
issue. Many issues on sage-newbie didn't get the attention they needed
because too few people were reading it. Your chances to get replies
are much better on sage-support.

>  And spent a lot
> of time documenting the monstrosity, removing magic numbers & generally
> cleaning up.
>
> Glanced at , indented my
> loops
> an extra space, -- obvious -- but unsure of other stuff. Added the
>
> -  Copyright (C) 2007 Dean Moore < [EMAIL PROTECTED] > ...
>
> boilerplate to the top -- perhaps a bit presumptuous on my part.  Some
> of the other coding conventions were confusing to me -- I wrote my
> doctoral thesis in AMS-TeX, but the LaTex stuff left me lost.
>
> But is this just a nice learning project on my part, or is there a demand
> for such
> functionality in SAGE?  Would people investigating, say, Ptolemaic
> astronomy,
> high school, college students/teachers find such worthwhile?  Or would my
> talents be better used elsewhere than animating relations, which is perhaps
> some SAGE equivalent of picking flowers on a spring day?

Many people use Sage in an educational setting and your contribution
in that area (as well as other stuff) is very welcome. If you plan to
contribute regularly you can ask off-list for a trac account (either
William or me).

> The code is such a monstrosity I don't care to here include it, -- someone
> said
> something about doing this --  but "published" it under the name
> "Hypotrochoid"
> at .
>
> Another question, I somehow thought that contributors had to be running
> Linux.  While
> I plan to do my M*cr*s*ft system a dual boot here in a bit, this is in
> abeyance.

You can perfectly develop in Linux, OSX and Windows. On Windows you
need to use the VMWare image, but there is an effort under way to
reanimate the Cygwin port and also get a "more native" port up and
running. So Windows will become a developer platform with equal
footing to the others, but it will a while for the more native version
to arrive. I am  hoping that the Cygwin port with work again in a
months or two since I have been spending some time on getting
everything to build again.

> Thanks for any answers or ideas.

You could also point out any shortcomings in the documentation and
also add to it. Not every developer must write code.

> Dean

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Bessel argument order

[mabshoff]

On Feb 16, 11:38 pm, "David Joyner" <[EMAIL PROTECTED]> wrote:

Hi David,

> Micheal suggested replacing all "#random's" by "..." and
> William seconded this. Then William suggested adding the scip option to
> the functions implemented. This has been done as well.
> The patch passes "sage -t" has some examples added and some
> docstring typos fixed. It can be found 
> at:http://sage.math.washington.edu/home/wdj/patches/special_16-02-2008.hg
> Should I make a ticket for this?

Yep, everything you work on (and plan to contribute to Sage) is a
ticket ;)



Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: symbol lookup error: /usr/lib/libxml2.so.2: undefined symbol: gzopen64

[mabshoff]

On Feb 16, 10:38 pm, Matthias Hillenbrand <[EMAIL PROTECTED]>
wrote:
> Hi Michael,

Hi Matthias,

> Thank you very much for your help. I am looking forward to the next
> release of SAGE.

I didn't really do much, I just ended up writing a couple emails :).
We are tracking the issue at

http://trac.sagemath.org/sage_trac/ticket/2182

> Matthias

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: symbol lookup error: /usr/lib/libxml2.so.2: undefined symbol: gzopen64

[mabshoff]

On Feb 16, 6:06 pm, Matthias Hillenbrand <[EMAIL PROTECTED]>
wrote:
> There seems to be no difference between the output form 'env' before
> and after starting the notebook.

Hi Matthias,

Jason Grout has figured out what the problem is and will probably post
a patch in the next 24 hours or so that should make it into 2.10.2.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: symbol lookup error: /usr/lib/libxml2.so.2: undefined symbol: gzopen64

[mabshoff]

On Feb 16, 5:31 pm, Matthias Hillenbrand <[EMAIL PROTECTED]>
wrote:
> Hello,

Hi Matthias,

> Since I have upgraded to SAGE 2.10.x I get the following error message
> when starting the notebook:
>
> gnome-www-browser: symbol lookup error: /usr/lib/libxml2.so.2:
> undefined symbol: gzopen64

The issue you mentioned has happened in the past and been fixed by
resetting LD_LIBRARY_PATH to the original version when launching
applications from Sage that are not in $SAGE_LOCAL/bin. So the problem
should not happen, unless you have SAGE_LOCAL/lib in LD_LIBRARY_PATH
somehow.

> I get this error message on two different computers (one running
> Ubuntu 7.10 the other Debian Testing, which I installed today) and
> with any browser (e.g. Firefox as standard browser). If I open a
> session of the browser before starting SAGE, the error won't occur.
>
> I searched for this error message with google and it seems to occur,
> if different versions of libz.so.1 are involved. Till now I found a
> libz.so.1 in the SAGE dir /local/lib, which differs from the one
> included in my Debian installation.
>
> The Version of the installed libxml2 is 2.6.31.dfsg-1 and my system
> has the following dependencies:
>
> ldd /usr/lib/libxml2.so.2
> linux-gate.so.1 =>  (0xe000)
> libdl.so.2 => /lib/i686/cmov/libdl.so.2 (0xb7def000)
> libz.so.1 => /usr/lib/libz.so.1 (0xb7dda000)
> libm.so.6 => /lib/i686/cmov/libm.so.6 (0xb7db4000)
> libc.so.6 => /lib/i686/cmov/libc.so.6 (0xb7c67000)
> /lib/ld-linux.so.2 (0x8000)
>
> ls -l /usr/lib/libz*
> lrwxrwxrwx 1 root root15 2008-02-16 12:56 /usr/lib/libz.so.1 ->
> libz.so.1.2.3.3
> -rw-r--r-- 1 root root 81012 2008-01-20 00:22 /usr/lib/libz.so.1.2.3.3
>
> Thanks in advance!

Could you post the output from "env" from a clean session and also one
where the launch fails? Since there are potentially large a diff of
those two would also be good.

What was the last working version of Sage before you upgraded?

> Matthias

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Extraction problem - Sage Windows bin

[mabshoff]

On Feb 16, 3:49 am, tangent <[EMAIL PROTECTED]> wrote:
> Hi
>
> I am having trouble decompressing sage-vmware-2.10.1.7z I have now
> downloaded it twice as I originally thought the file had been
> corrupted during the download.
>
> Using the 7z file manager, it returns an error message for each file
> in the compressed package. Using Winzip, the 7z file is not recognised
> as a valid archive (though this software has used 7z compression
> previously).

WinZip indeed seems to support 7z, but I would recommend that you try
the original extraction utility from Fwww.7-zip.org.

> I noticed this is a new file and that somebody else is having problems
> with the Linux version. Is it possible there is a problem with the
> original file? Has anybody successfully installed v2.10.1 for windows?

The problem was the the downloaded file was incomplete. I personally
did not install on Windows, but I am sure the general brokenness of
the VMWare image would have come up in the last two weeks :)

We should *really* provide checksums (sfv for Windows users, md5 for
the others) since the issue has come up repeatedly. We should make it
scripted, so that when somebody is adding (i.e. William only at the
moment) all that he needs to do is to run a script. I would suggest to
make it part of "mirror" since that script gets run for sure :)

> thanks, Jen

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: optional_packages() error

[mabshoff]

On Feb 16, 7:43 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Feb 15, 2008 10:16 PM, Ted Kosan <[EMAIL PROTECTED]> wrote:
>
>
>
> > When I execute the optional_packages() function in a worksheet, I
> > receive the following error:
>
> > 
> > Traceback (most recent call last):
> >   File "/opt/sage-2.10-debian32-i686-Linux/local/bin/sage-list-packages",
> > line 20, in 
> > os.makedirs("%s/tmp"%SAGE_ROOT)
> >   File "/opt/sage-2.10-debian32-i686-Linux/local/lib/python/os.py",
> > line 172, in makedirs
> > mkdir(name, mode)
> > OSError: [Errno 13] Permission denied: 
> > '/opt/sage-2.10-debian32-i686-Linux/tmp'
> > Using SAGE Serverhttp://www.sagemath.org//packages
>
> Does the user running sage have write permissions to
>
> /opt/sage-2.10-debian32-i686-Linux/tmp?

It very much looks like a permission issue and I reported the issue at
#961. We will more than likely hit the issue more often in the future
as people use shared installs, so maybe be ought to start using $TMP
instead of the local tmp as directory for temporary files.

>  -- William
>

Cheers,

Michael

>
> > Optional package list (shown above) appears to be currently not
> > available or corrupted (network error?).
> > ([], [])
> > 
>
> > I am using SAGE Version 2.10.  Does anyone have any thoughts on what
> > is causing this error?
>
> > Thanks,
>
> > Ted
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Odd behaviour of parametric_plot

[mabshoff]

> Still no response to this problem, either
>
> Bill

Generally speaking: Sage Days somewhere (IPAM in this case) tend to
drop the amount of people that deal with problems on the mailing list
before, during and after the work shop because people have to travel,
get ready for the workshop and afterwards deal with the things that
got left undone during their absence. Just look at the number of
messages to sage-devel this month and SD7 is the clear culprit.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage Installation - Problem with untar-ing Linux Binaries

[mabshoff]

On Feb 16, 2:46 am, Kishore <[EMAIL PROTECTED]> wrote:
> The size of the tar ball is less than what you mentioned
>
> -rw---  1 kishore pointcount 236711936 Feb 12 21:01 sage-
> sage-2.10.1-linux-debian64-intel-x86_64-Linux.tar.gz

You are missing about 50 mb from the tarball, so you will need to
redownload it again. You might want to use a download manager,
depending on the connection you use.

> Nevertheless, I proceeded continuing and when I typed the command
> 'sage' , I got an error mentioning that
> "You must compile SAGE first using 'make' in the SAGE root directory.
> (If you have already compiled SAGE, you must set the SAGE_ROOT
> variable in the file sage)"
>
> I tried setting the SAGE_ROOT variable in 'sage' file to my bin
> directory where it's installed and still after  doing so, I faced this
> issue

That won't work since you are missing quite a bit of Sage.

> "/local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/sage.bin: /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-
> Linux/local/bin/sage.bin: cannot execute binary file
> /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/python: /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-
> Linux/local/bin/python: cannot execute binary file
> /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/sage-ipython: line 6:
> SAGE IPython startup script.
>
> : command not found
> /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/sage-ipython: line 8: __SAGE_AUTOINJECT__: command not found
> import: unable to open X server `'.
> Can't open IPython.CrashHandler: No such file or directory at /eci/
> share/bin/from line 57.
> Can't open import: No such file or directory at /eci/share/bin/from
> line 57.
> Can't open CrashHandler: No such file or directory at /eci/share/bin/
> from line 57.
> /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/sage-ipython: line 13: syntax error near unexpected token `('
> /local/bin/sage-sage-2.10.1-linux-debian64-intel-x86_64-Linux/local/
> bin/sage-ipython: line 13: `sage = IPython.Shell.start()'
> "
>
> Please let me know if I am missing something to complete my
> installation successfully.

Download a full archive ;)

> Regards
> Kishore

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-newbie list and sage-support list

[mabshoff]

>
> > How exactly do you close a google group?
>
> I just changed the list settings so that only "managers can post", which is 
> just
> you and me.   That should be good enough.

We should also edit the welcome messages for those two groups to point
people to -support and -devel. While someone is at it it might also be
a good idea to homogenize those welcome messages. Not that I am
volunteering ...

> William

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-newbie list and sage-support list

[mabshoff]

On Feb 15, 10:59 pm, "John Cremona" <[EMAIL PROTECTED]> wrote:
> I agree.  It would be different if Sage were centralised with a boss
> who could assign certain people the task of looking after newbies.
>
> Did you reply to User 1 trying to convince him/her to try again but to
> use sage-support?  As Sage does actually provide amazingly good
> support )via this list for example) it would be a great pity if false
> rumours to the contrary were to spread.

Yep, we should also evaluate if we need a separate sage-forum - it is
unclear to me what is different about that group.

> John

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Sage 2.10.1 released

[mabshoff]

Hello folks,

Sage 2.10.1 has been released on February 2nd, 2008. It is available
at

   http://sagemath.org/download.html

* About Sage (http://www.sagemath.org)

Sage is developed by volunteers and combines over 75 open source
packages.  It is available for download from sagemath.org and its
mirrors in source or binary form. If you have any questions and/or
problems please report them to the google groups sage-devel,
sage-support, sage-forum or sage-newbie. You can also drop by in
#sage-devel on irc.freenode.net.

-

At least the following 27 developers contributed to this release:

 * Michael Abshoff
 * Martin Albrecht
 * Nick Alexander
 * Robert Bradshaw
 * Michael Brickenstein
 * Craig Citro
 * Timothy Clemans
 * John Cremona
 * Burcin Erocal
 * Alex Ghitza
 * Jason Grout
 * Mike Hansen
 * David Harvey
 * David Joyner
 * Josh Kantor
 * Simon King
 * David Kohel
 * Robert Miller
 * Joel Mohler
 * Willem Jan Palenstijn
 * Yi Qiang
 * Denis Simone
 * William Stein
 * Steve Vonn
 * Ralf-Philip Weinmann
 * Carl Witty
 * Paul Zimmermann

Cheers,

Michael Abshoff (release chair), William Stein

* Major Features, New Spkgs and Bugfixes

 * Updated to eclib.spkg from cremona.spkg.
 * Update to FLINT 1.0.6
 * Updated to MPFR 2.3.1 which contains an important fix that
   avoids smashing the stack with large precisions
 * Improved ATLAS with pre-tuned settings for more CPUs. This
   greatly reduced the compile time on Pentium M and Athlon CPUS.
 * Massive improvements on the doctesting coverage.
 * Fix security issue with the notebook (found by T. Clemans).
   It is *highly* recommended that you upgrade.
 * Fixed various memleaks in PolyBoRi. There are still some
   issues in the PolyBoRi wrapper.
 * Progress on the Solaris port as well as the 64-bit OS X port.
   Neither port is yet ready for prime time.

* Known Issues with 2.10.1:

 * GNUTLS might fail to build on OS X if guile is present. For a
   workaround use the updated GNUTLS spkg linked from ticket #2003
 * You need to update the optional valgrind.spkg to the 3.3.0
   release because the interface has been changed in 2.10.1 and it
   isn't backward compatible with earlier releases.

* Bug Statistics

We closed 186 tickets. For details see

http://trac.sagemath.org/sage_trac/milestone/sage-2.10.1

* Upcoming Releases

The next release will be 2.10.2, chaired by Michael Abshoff. The
release
is planned post Sage Days 7, but details are sketchy at the moment.
Potential features include:

 * David Roe's unramified and Eisenstein extensions of Qp and Zp
 * John Voight's code for enumerating totally real fields [with
   generalizations for not-totally real fields]
 * OS X 10.5 support in 64-bit mode [this will depend on fixing
   a vexing segfault in libSingular and getting disutils to
   play nice with numpy]
 * GCC 4.3 support out of the box
 * FreeBSD support out of the box
 * Michael Abshoff's improved experimental Solaris support
 * William Stein's rewrite of the mwrank Cython interface

* Doctesting Coverage

For 2.10 we had:
Overall weighted coverage score:  35.4%
Total number of functions:  18296

We increased coverage by 2.9% in 2.10.1 while removing 6 functions:
Overall weighted coverage score:  38.3%
Total number of functions:  18290

* Closed Tickets:

Merged in rc5:

#1547: Michael Abshoff: Add pre-tuned settings for ATLAS for
   certain CPUs
#1886: Michael Abshoff: add ATLAS pretuning for AMD Athlon MP
   processors
#1977: Michael Abshoff: valgrind 3.3.0 no longer appends $PID per
   default to the logs
#1986: William Stein: Numerical noise in fast evaluation code.
#1999: Michael Abshoff, William Stein: delete mpl3d code from sage
#2019: David Joyner: problem with gap_packages-4.4.10_3
#2021: Jason Grout: Only start one maxima instance
#2026: William Stein, Jason Grout: matrix.eigenspaces doctest
   description is misleading
#2027: Ralf-Philip Weinmann, Michael Brickenstein: update PolyBoRi
   to current CVS: fix memleak and remove joke patent claim
#2029: Michael Abshoff: adjust TIMEOUT for long and valgrinded
   doctests
#2035: William Stein: transform.pyx is now broken (probably caused
   by other fixes)

Merged in rc4:

#1926: Burcin Erocal: fixes for the maple interface
   [this is another patch to fix some more doctest issues]
#1960: Martin Albrecht: bug when reducing Gröbner basis
#1967: William Stein: fix matplotlib locale-related bugs once and for
   all
#1968: Timothy Clemans and William Stein: notebook -- remove
   capability of using live tutorial for users not signed in to
   the notebook server; this was a potential security risk.
#1973: Jason Grout, Mike Hansen: native partition_associated function
#1983: William Stein: change 0^0, for 0 a Sage integer, to evaluate
   to 1 for consistency with Python, PARI, Magma, Maple, MPFR,
   GMP, etc.
#1985: Steve Vonn, Michael Abshoff: is_pseudoprime docstring

[sage-support] Sage 2.10.1.rc5 released!

[mabshoff]

Here we go with Sage 2.10.1.rc5. This is the *final* rc and
it will be identical with the final 2.10.1 release modulo
any critical bug fixes that will pop up over the next 12
hours. The release is planned for the tonight (PST).

We finally fixed the ATLAS build issue that would cause very
long build times. If your ATLAS build takes very long please
let us know and we can provide you with the needed information
to create pre-tuned settings that we will then include in the
next release.

Please report *any* build issues and doctest failures.

The tarball [205MB] is available at the usual place:

http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10.1/sage-2.10.1.rc5.tar

Cheers,

Michael

Merged in rc5:

#1547: Michael Abshoff: Add pre-tuned settings for ATLAS for
   certain CPUs
#1886: Michael Abshoff: add atlas pretuning for AMD Athlon MP
   processors
#1977: Michael Abshoff: valgrind 3.3.0 no longer appends $PID per
   default to the logs
#1986: William Stein: Numerical noise in fast evaluation code.
#1999: Michael Abshoff, William Stein: delete mpl3d code from sage
#2019: David Joyener: problem with gap_packages-4.4.10_3
#2021: Jason Grout: Only start one maxima instance
#2026: William Stein, Jason Grout:matrix.eigenspaces doctest
   description is misleading
#2027: Ralf-Philip Weinmann, Michael Brickenstein: update PolyBoRi
   to current CVS: fix memleak and remove joke patent claim
#2029: Michael Abshoff. adjust TIMEOUT for long and valgrinded
   doctests

Merged in rc4:

#1926: Burcin Erocal: fixes for the maple interface
   [this is another patch to fix some more doctest issues]
#1960: Martin Albrecht: bug when reducing Gröbner basis
#1967: William Stein: fix matplotlib local-related bugs once and for
   all
#1968: William Stein: notebook -- remove capability of using live
   tutorial for users not signed in to the notebook server
#1973: Jason Grout, Mike Hansen: native partition_associated function
#1983: William Stein: change 0^0, for 0 a Sage integer, to evaluate
   to 1 for consistency with Python, PARI, Magma, Maple, MPFR,
   GMP, etc.
#1985: Steve Vonn, Michael Abshoff: is_pseudoprime docstring
   doesn't wrap
#1988: Craig Citro: doctest fallout -- osx ppc 10.4 overflow error
#1989: Craig Citro: doctest fallout -- osx ppc 10.4 matrix index out
   of range
#1996: Craig Citro: doctest failure on 2.10.1.rc3: libecm.pyx
#2002: Jason Grout: creating a fresh new notebook in sage-2.10.1.rc3
   is broken
#2005: William Stein: The .sage directory -- adjust permissions
#2006: Michael Abshoff: crap -- gap contains a binary
#2008: Michael Abshoff: crap -- cython contains a stupid OSX file
#2009: Michael Abshoff: crap -- networkx spkg in sage-2.10.1.rc3
   contains a bunch of .svn directories
#2010: Michael Abshoff: crap -- libpng contains lots and lots of
   weird (OS X?) temp or meta files
#2011: Michael Abshoff: crap -- mpfi ships with two precompiled
   binaries
#2012: Michael Abshoff: crap -- jmol contains some os x files
   that shouldn't be there
#2013: Michael Abshoff: crap -- f2c spkg contains precompiled binary
#2014: Michael Abshoff: crap -- matplotlib contains an annoying OS
   X temp file
#2015: Michael Abshoff: crap -- twisted contains a bunch of ._ osx
   crap
#2016: Michael Abshoff: crap -- there is a precompiled object
   file in the SAGE_ROOT/examples directory
#2018: Michael Abshoff: crap -- scipy sandbox has a bunch of .svn
   directories
#2022: William Stein: bug with load and loads when data is not
   compressed during save

Merged in rc3:

#740: John Cremona: Implement standard transformations for elliptic
  curves & points
#1328: Michael Abshoff: 2.8.14/Solaris: partitions_c.h compile fix
#1329: Michael Abshoff: 2.8.14/Solaris: real_rqdf.pyx compile fixes
#1554: Mike Hansen: issues with calling symbolic expressions
#1637: Michael Abshoff: Update to mpfr 2.3.1 - fix small issues in
   the spkg
#1887: Robert Miller: graph isomorphism for multiple edges
#1923: Jason Grout: Make an obvious checkbox or menu to switch on
   pretty printing in the notebook
#1938: Robert Bradshaw: Fast (double) function evaluation for
   plotting, etc.
#1958: Craig Citro: fix problems with ANSI codes in sage0.py
#1966: Josh Kantor: inline_fortran broke on OSX
#1970: Willaim Stein, Martin Albrecht: gnutls should not be needed
   if you're running the notebook insecurely, so don't require it
#1972: William Stein: programming guide -- fix a confusing typo
#1978: Yi Qiang, Michael Abshoff: update python-gnutls
#1979: John, Cremona, Michael Abshoff: doctest fall out from #740
   in tut.tex

Merged in rc2:

#571: John Cremona, Denis Simone, William Stein: weird doctest
  failure in simon_two_descent [to quote William: "John and
  Denis Simone in fact did greatly update this code and it
  is now in Sage. And this particularly doctest runs almost
 

[sage-support] Re: Error calling plots.

[mabshoff]

On Feb 2, 3:49 am, adrianmatematico <[EMAIL PROTECTED]>
wrote:
> [EMAIL PROTECTED]:~$ cat /proc/cpuinfo
> processor   : 0
> vendor_id   : CentaurHauls
> cpu family  : 6
> model   : 9
> model name  : VIA Nehemiah
> stepping: 8
> cpu MHz : 1197.000
> cache size  : 64 KB
> fdiv_bug: no
> hlt_bug : no
> f00f_bug: no
> coma_bug: no
> fpu : yes
> fpu_exception   : yes
> cpuid level : 1
> wp  : yes
> flags   : fpu vme de pse tsc msr cx8 mtrr pge cmov pat mmx
> fxsr sse up rng rng_en ace ace_en
> bogomips: 2404.28
> clflush size: 32

Ok, that seems to indicate that it isn't a problem with ATLAS since
that one tends to pop up right at startup. I would consider gmp
another possibility.

> I will try to build from source, then.  Thanks a bunch!!!

Ok, let us know about any problems you see.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Error calling plots.

[mabshoff]

On Feb 2, 3:29 am, adrianmatematico <[EMAIL PROTECTED]>
wrote:
> I tried to install a couple of things and now sage cannot plot.  I
> tried to install visual-python, along with the libraries.
>
> Sage used to work, and still does in my other computer.  But when I do
>
> sage: plot(cos(x),-3,3)
> /usr/local/src/sage-2.10-ubuntu32-i686-Linux/local/bin/sage-sage: line
> 210:  9119 Illegal instruction (core dumped) sage-ipython -c
> "$SAGE_STARTUP_COMMAND;" "$@"
>
> I tried to do sage -testall   and it failed with a similar message in
> the file
>
> sage -t  devel/sage-main/sage/modular/modform/cuspidal_submodule.pysh:
> line 1:  9380 Illegal instruction (core dumped) /usr/local/src/
> sage-2.10-ubuntu32-i686-Linux/local/bin/
> python .doctest_cuspidal_submodule.py > .doctest/out 2> .doctest/err
>
> A mysterious error (perphaps a memory error?) occurred, which may have
> crashed doctest.
>  [5.3 s]
>
> I have no idea.
>

>
> Any help is appreciated.

Hmm, what is your exact CPU architecture? I.e. the output from "cat /
proc/cpuinfo"? Reasons for the crash might be

a) a corrupted download (unlikely)
b) ATLAS vs. your CPU issue

> Adrian.

Probably the easiest thing to get rid of the problem would be to
compile from sources. You can also switch to a system ATLAS by
following the FAQ entry "QUESTION: I downloaded a Sage binary and it
crashes on startup with Illegal instruction. What can I do?" - see
http://wiki.sagemath.org/faq

Let us know if that works.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Fwd: Problems with sage 2.10 binary

[mabshoff]

On Feb 1, 7:03 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> -- Forwarded message --
> From: Terry Duell <[EMAIL PROTECTED]>
> Date: Feb 1, 2008 1:25 AM
> Subject: Problems with sage 2.10 binary
> To: [EMAIL PROTECTED]
>
> Hullo,

Hi Terry,

> I tried to send this to sage-support, but need to be subscribed. I would like
> to see if sage is what I need before subscribing to yet another mailing list,
> hopefully you may be able to help me get started.

You can easily unsubscribe from Google groups or use the web interface
to post and read messages. There is also options not to receive email
or get them in digest mode and so on.

> I have just downloaded the binary for sage 2.10. I got the rhel5 version and
> am trying to use it on an x86 running Fedora Core 6.
> When I try to run sage for the first time, this is the output to the
> terminal...
>
> [EMAIL PROTECTED] sage]$ sage
> --
> | SAGE Version 2.10, Release Date: 2008-01-18|
> | Type notebook() for the GUI, and license() for information.|
> --
> The SAGE install tree may have moved.
> Regenerating Python.pyo and .pyc files that hardcode the install PATH
> (please wait less than a minute)...
> Please do not interrupt this.
>
> /home/terry/Packages/sage/local/bin/sage-sage: line 210:  3233 Illegal
> instruction sage-ipython -c "$SAGE_STARTUP_COMMAND;" "$@"
> [EMAIL PROTECTED] sage]$
>
> Subsequent attempts to run sage all result in the same error message.
> It doesn't seem to like me :-)
>
> Helpful advice will be gratefully accepted.

The "Illegal instruction" indicates that the CPU you run Sage on
doesn't have the needed instruction set [SSE2 or 3] the CPU the binary
was compiled on used. For a solution see the FAQ at

  http://wiki.sagemath.org/faq

specifically

  "QUESTION: I downloaded a Sage binary and it crashes on startup with
Illegal instruction. What can I do?"

> Cheers,
> --
> Regards,
> Terry Duell

Cheers,

Michael

> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: bug

[mabshoff]

On Jan 30, 10:36 am, "steve vonn" <[EMAIL PROTECTED]> wrote:
> is_pseudoprime?
> The help entry for (is_pseudoprime) has a nonwrapping text bug
>
> INPUT:
> flag -- int
> 0 (default): checks whether x is a
> Baillie-Pomerance-Selfridge-Wagstaff pseudo prime (strong Rabin-Miller
> pseudo prime for base 2, followed by strong Lucas test for the
> sequence (P,-1), P smallest positive integer such that P^2 - 4 is not
> a square mod x).
> > 0: checks whether x is a strong Miller-Rabin pseudo
> prime for flag randomly chosen bases (with end-matching to catch
> square roots of -1).

hi Steve,

thanks for reporting this buglet. It is now #1985 and should be fixed
in the Sage 2.10.1 release.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Sage 2.10.1.rc3 released!

[mabshoff]

Here we go: Sage 2.10.1.rc3 is hopefully the next to last rc.
The release is planned in the next 24 hours. We finally fixed
the GNUTLS issues, elliptic curve improvements (#740), some
Solaris build issues and various bug fixes. What needs to be
done before the release is #1968 (a security fix), also an
ATLAS update (actually five different tickets) and some more
small bug fixes. Anything else will be reviewed on a case by
case basis, but since the next release isn't far away we will
be rather conservative due to SD7.

The tarball [202MB] is available at the usual place:

http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10.1/sage-2.10.1.rc3.tar

Please report all issues. If you do not have Maple installed please
apply
the patch at

http://trac.sagemath.org/sage_trac/attachment/ticket/1926/1926-make_doctests_optional.patch

to avoid that doctest failure.

Cheers,

Michael

###

Must be merged/resolved before Sage-2.10.1.final:

#1495: Terminate ATLAS build on Linux when power management is
   enabled
#1547: Add pre-tuned settings for ATLAS for certain CPUs
#1641: Make ATLAS restart build on tolerance error
#1767: Add SAGE_ATLAS_LIB, SAGE_ATLAS_TARGZ flags to allow use
   of external atlas
#1886: add atlas pretuning for AMD Athlon MP processors
#1926: fixes for the maple interface
#1968: notebook -- remove capability of using live tutorial
   for users not signed in to the notebook server

Need additional reviews: #1978, #1979

###

To Do:

* reviews for ticket with spkg/patches/bundles
* disable mandatory FLINT check
* update documentation for new ATLAS options
* #1689: update documentation for new SAGE_FORTRAN_LIB
* #1884: leak in coercion [problematic, probably bumped to 2.10.2
  due to segfaults triggered by #1337]
* #1874: PolyBoRi leak: fix it or get somebody else to do it
* remove opencdk since it is included in gnutls 2.2.1 [2.10.2?]
* add two open valgrind issues to track: Valgrind issues in 2.9.2

sage-memcheck.3056: Conditional jump or move depends on uninitialised
value(s), Invalid read of size 1
doctest calculus.py
sage-memcheck.19097:
doctest: real_rqdf.py

sage-memcheck.24275: longest_match - libz: Conditional jump or move
depends on uninitialised value(s)

###

spkg audit:

* ensure that "make install" is non-parallel, i.e. do a "export
MAKE=make"
* SAGE_LOCAL check (#633)
* add spkg-check (#299)
* add proper SPKG.txt to all packages
* /usr/bin/env bash (#1638)
* add md5sums for spkgs (#329)
* check for SAGE_DEBUG mode, i.e CFLAGS="-O0 -g" (#475)
* Debian build support

###

Bugfixes that need to be pushed upstream:

 * gsl_isinf: Bug on OSX where the system's isinf always returns "1"
 * libfplll Solaris build fixes
 * GNUTLS Cygwin build fixes
 * rubiks.spkg: dik solver has cc -> $(CC) fix
 * python_gnutls fix for OSX and other gnutls versions

###

Reopened:

#1973: Jason Grout, Mike Hansen: native partition_associated function
   [causes import errors on Sage startup after a "sage -ba"]
#1926: Burcin Erocal, Mike Hansen: fixes for the maple interface
   [patch available to make maple doctests optional again]

To Review/merge que:

Merged in rc3:

#740: John Cremona: Implement standard transformations for elliptic
  curves & points
#1328: Michael Abshoff: 2.8.14/Solaris: partitions_c.h compile fix
#1329: Michael Abshoff: 2.8.14/Solaris: real_rqdf.pyx compile fixes
#1554: Mike Hansen: issues with calling symbolic expressions
#1637: Michael Abshoff: Update to mpfr 2.3.1 - fix small issues in
   the spkg
#1887: Robert Miller: graph isomorphism for multiple edges
#1923: Jason Grout: Make an obvious checkbox or menu to switch on
   pretty printing in the notebook
#1938: Robert Bradshaw: Fast (double) function evaluation for
   plotting, etc.
#1958: Craig Citro: fix problems with ANSI codes in sage0.py
#1966: Josh Kantor: inline_fortran broke on OSX
#1970: Willaim Stein, Martin Albrecht: gnutls should not be needed
   if you're running the notebook insecurely, so don't require it
#1972: William Stein: programming guide -- fix a confusing typo
#1978: Yi Qiang, Michael Abshoff: update python-gnutls
#1979: Michael Abshoff: doctest fall out from #740 in tut.tex

Merged in rc2:

#571: John Cremona, Denis Simone, William Stein: weird doctest
  failure in simon_two_descent [to quote William: "John and
  Denis Simone in fact did greatly update this code and it
  is now in Sage. And this particularly doctest runs almost
  instantly now, so remove #long"]
#1058: John Cremona, William Stein: the mwrank interface barfs on
   bad input

[sage-support] Re: lost worksheets

[mabshoff]

Hi Marshall,

On Jan 28, 10:27 pm, Marshall Hampton <[EMAIL PROTECTED]> wrote:
> I now have 14 students actively using Sage on 3 different servers, and
> I have run across a mysterious problem I haven't seen reported here
> before.  Three of my students had worksheets suddenly go missing,
> without a deliberate delete.  I looked in the notebook files and the
> worksheets were still there, just not visible on the list of
> worksheets.  Any guesses on what happened?  As a stopgap, I manually
> pasted in the worksheet text to a new copy, but if this happens a lot
> it will really aggravate them.

Are those worksheets on servers you host or were they on the public
notebook?

> -M. Hampton

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage 2.10 released!

[mabshoff]

On Jan 23, 10:15 pm, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> On Jan 23, 10:04 pm, Kate <[EMAIL PROTECTED]> wrote:



> Hi Kate,
>
> this is now #1898. You can expect a patch for this in the next hour or
> two.
>
> Cheers,
>
> Michael

Hi Kate,

there is now a patch up at #1898. Could you please test and report if
it works for you?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.10.1.alpha0/1: notebook and jmol don't work for me

[mabshoff]

On Jan 23, 7:49 pm, "Yi Qiang" <[EMAIL PROTECTED]> wrote:
> I can comment on the python-gnutls issue. The problem is that the API
> for gnutls has changed and is NOT backwards compatible with the
> previous releases. Hence, python-gnutls will not work since it is
> simply a package that wraps the gnutls library and those wrappings
> were automatically generated by gcc-xml and py2xml.  Below is the
> email I got from the python-gnutls author about what needs to be done
> to get python-gnutls to work.

Hi Yi,

thanks for tracking this down. I will try to get it to work as
described in the email.

Cheers,

Michael

> show details Jan 11 (12 days ago)
>
> Reply
>
> On Friday 11 January 2008, you wrote:
>
> > Hi Dan,
> > We've been using your python-gnutls wrappers for a while now for the
> > SAGE [1] project with great success.  However we'd like to update to
> > the latest gnutls release (2.2.0 ATM) and I noticed that they've made
> > quite a few changes to the OpenPGP parts which causes
> > incompatibilities with the wrapper. Do you have any plans to update
> > python-gnutls to work with gnutls-2.2.0?
>
> I will when the distribution I use will have it packaged. Until then I
> stick with what I have since there is no reason for me to upgrade.
>
> Otherwise, openpgp is not really used, just made available. So unless you
> actively use the low level openpgp functions yourself it should not be a
> problem. Otherwise, you can regenerate the low level libraries yourself
> to match the latest gnutls. You just have to install the code generator
> fromhttp://starship.python.net/crew/theller/ctypes/(xml2py) and gcc-xml
> With them you can run make clean; make before you run setup.py build to
> regenerate the low level library files to match the latest gnutls header
> files.
>
> --
> Dan
>
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage 2.10 released!

[mabshoff]

On Jan 23, 10:04 pm, Kate <[EMAIL PROTECTED]> wrote:
> Michael,
>
> Sage 2.10 fails one test on ia64-Linux:
>
> sage -t  devel/sage-main/sage/rings/polynomial/
> polynomial_element.pyx**
> File "polynomial_element.pyx", line 2669:
> sage: p.roots(ring=CIF)
> Expected:
> [([-1.4142135623730952 .. -1.4142135623730949], 1),
> ([1.4142135623730949 ..
> 1.4142135623730952], 1), ([-1.2146389322441827 .. -1.2146389322441821]
> - [0.1414250525823937... .. 0.1414250525823939...]*I, 2),
> ([-0.141425052582393... .. -0.1414250525823937...] +
> [1.2146389322441821 .. 1.2146389322441827]*I, 2),
> ([0.141425052582393... .. 0.141425052582393...] -
> [1.2146389322441821 .. 1.2146389322441827]*I, 2),
> ([1.2146389322441821 .. 1.2146389322441827] + [0.14142505258239376 ..
> 0.14142505258239399]*I, 2)]
> Got:
> [([-1.4142135623730952 .. -1.4142135623730949], 1),
> ([1.4142135623730949 ..
> 1.4142135623730952], 1), ([-1.2146389322441827 .. -1.2146389322441821]
> - [0.14142505258239371 .. 0.14142505258239397]*I, 2),
> ([-0.14142505258239397 .. -0.14142505258239376] +
> [1.2146389322441821 .. 1.2146389322441829]*I, 2),
> ([0.14142505258239373 .. 0.14142505258239394] - [1.2146389322441821 ..
> 1.2146389322441829]*I, 2), ([1.2146389322441821 .. 1.2146389322441827]
> + [0.14142505258239376 .. 0.14142505258239399]*I, 2)]
> **
>
> Kate

Hi Kate,

this is now #1898. You can expect a patch for this in the next hour or
two.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: int / Integer question

[mabshoff]

On Jan 23, 2:29 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Jan 22, 2008 11:48 PM, Paul Zimmermann <[EMAIL PROTECTED]> wrote:



> > I guess 'long' is based on GMP too, does it make sense to have two 
> > concurrent
> > interfaces to GMP integers?
> > Paul Zimmermann
>
> long is _not_ based on GMP.   long is implemented from scratch in C by the
> Python developers.   It is completely independent from GMP, and has much
> different performance characteristics, storage formats, etc.

FYI: The arbitrary precision arithmetic in Python 3K will be based on
GMP. Somebody posted prototype bindings on one of the GMP mailing
lists about 2 months ago.



Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.10.1.alpha0/1: notebook and jmol don't work for me

[mabshoff]

On Jan 23, 11:06 am, Simon King <[EMAIL PROTECTED]> wrote:
> Dear Michael
>
> On Jan 23, 1:30 am, mabshoff <[EMAIL PROTECTED]
>
> dortmund.de> wrote:
> > On Jan 22, 11:34 pm, Simon King <[EMAIL PROTECTED]> wrote:
> 
> > > I mean: saying "./sage -notebook", i got a lot of error messages and
> > > eventually an "Unhandled SIGSEGV".
>
> > I can reproduce it.
>
> Slightly relieving for me :-)
>
> > A couple remarks:
>
> > a) import gnutls works for 2.10.1.alpah2
> > b) python_gnutls in the way we ship it is outdated. There is a 1.1.4
> > at the cheese shop which compiles fine. But it doesn't make a
> > difference.
> > c) We might just me missing the asn1 librariy, one which can be found
> > atftp://ftp.gnutls.org/pub/gnutls/libtasn1/libtasn1-1.2.tar.gz
>
> So does it mean i only have to wait until 2.10.1.alpha2 is available?
> Or may i have done something wrong when i installed gnutls (using
> yast)?

I downgraded to the last known working gnutls.spkg and that didn't fix
the problem. So I am currently trying to figure out what exactly is
wrong. William started revisiting various alternatives and it seems
likely now that we might downgrade for 2.10.1 and revisit the issue
again in 2.10.2, unless we come up with some alternative
implementation.

> 
>
> > What JDK are you running?
>
> What is a JDK?
>
> > jmol seems to have problems with non-Sun
> > ones, so installing that ought to solve your problem in case it isn't
> > GNUTLS related.
>
> jmol had no problems with my machine so far. When i use sage-2.10 that
> i obtained from building sage-2.8.6 and successive upgrade, plot3d-ing
> works fine.
>
> And i don't know if it is GNUTLS related, because there was no error
> message. Simply "S.show()" had the same visible effect as saying
> "none".

See the other thread in sage-devel :)

> Cheers
>   Simon

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.10.1.alpha0/1: notebook and jmol don't work for me

[mabshoff]

On Jan 22, 11:34 pm, Simon King <[EMAIL PROTECTED]> wrote:
> Dear Sage team,
>
> in reply to myself, in the hope to clearify things:
>
> > After installing gnutls, building was successful. However, I was
> > unable to run the notebook in the alpha-versions.
>
> I mean: saying "./sage -notebook", i got a lot of error messages and
> eventually an "Unhandled SIGSEGV".
>
> Seehttp://sage.math.washington.edu/home/SimonKing/nb_error.txt

I can reproduce it. A couple remarks:

a) import gnutls works for 2.10.1.alpah2
b) python_gnutls in the way we ship it is outdated. There is a 1.1.4
at the cheese shop which compiles fine. But it doesn't make a
difference.
c) We might just me missing the asn1 librariy, one which can be found
at ftp://ftp.gnutls.org/pub/gnutls/libtasn1/libtasn1-1.2.tar.gz

Since I have never looked in that area of the code some debugging help
is appreciated. I am taking this over to sage-devel.

> > When i tried to display something, the jmol applet simply didn't show
> > up.
>
> Here, i successfully did "./sage" and successfully defined a 3d-
> object, but i couldn't show it:
>
> --
> | SAGE Version 2.10.1.alpha1, Release Date: 2008-01-21   |
> | Type notebook() for the GUI, and license() for information.|
> --
>
> sage: import sage.plot.plot3d.all as p
> sage: S=p.sphere(color='red')
> sage: S
> // No picture was showing up!
>
> sage: S.show()
> // again, nothing happened.
> sage: S.__str__()
> 'Graphics3d Object'
> // so S is what it should be
> sage: 2+2
> 4
> // At least there is *something* going in a good way :-)

What JDK are you running? jmol seems to have problems with non-Sun
ones, so installing that ought to solve your problem in case it isn't
GNUTLS related.

> > Please tell me what data i should provide (install.log? The lengthy
> > error message after trying to open the notebook?) to track this down.
>
> I told myself that providing install.log can't hurt -- the built was
> successful, but perhaps there is a hint in the install.log about the
> other problems. 
> Seehttp://sage.math.washington.edu/home/SimonKing/install.log.gz,
> which is the install log for alpha1.
>
> By the way:
> uname -a says
>   Linux mpc739 2.6.18.8-0.3-default #1 SMP Tue Apr 17 08:42:35 UTC
> 2007 x86_64 x86_64 x86_64 GNU/Linux
> and i use
>   gcc (GCC) 4.2.1
>
> Yours
>Simon

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: File Functions and Win path

[mabshoff]

On Jan 18, 6:42 pm, mark h <[EMAIL PROTECTED]> wrote:
> i want to write some actuarial specifications in Sage. i want to
> create test cases in external files, so that they can also be used for
> the implementation testing.
>
> can i read/write data from external files ?
> can Sage process text ?

Yes and yes. Since Sage is written in Python you can use all of
Python's file functions. See http://www.diveintopython.org/toc/index.html
and specifically http://www.diveintopython.org/file_handling/file_objects.html

> i'm running on win platform. how do i construct path to file to load ?

VMware offers a shared folder - see 
http://www.vmware.com/support/ws5/doc/ws_running_shared_folders.html
for details and how to set that up. That way you can access, create
and work with files that are on the Windows side transparently from
Sage. Note that in Sage you need to give the Linux path, not the
Windows one.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Sage 2.10 released!

[mabshoff]

On Jan 19, 5:36 am, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> Hello folks,
>
> Sage 2.10 has been released on January 18th, 2008. It is available at
>
>http://sagemath.org/download.html
>
> * About Sage (http://www.sagemath.org)
>
> Sage is developed by volunteers and combines 71 open source packages.
> It is available for download from sagemath.org and its mirrors in
> source or binary form. If you have any questions and/or problems
> please report them to the google groups sage-devel, sage-support,
> sage-forum or sage-newbie. You can also drop by in #sage-devel in
> freenode.
>
> -
>
> We did not achieve all the goals we had set ourselves for Sage 2.10
> and also decided to release slightly before the planned date.  The
> reason was one hand the much better than expected stability of
> 2.10.alpha4 [it build and tested perfectly on all out test platforms]
> and on the other hand the need to have a stable release for Bug Day 9
> planned for this Saturday.
>
> The following people contributed to this release:
>
>  * Michael Abshoff
>  * Martin Albrecht
>  * Bill Allombert
* Benjamin Antieau
>  * Robert Bradshaw
>  * Craig Citro
>  * Francis Clarke
>  * Burcin Erocal
>  * Alex Ghitza
>  * Jason Grout
>  * Peter Jipsen
>  * David Joyner
>  * Mike Hansen
>  * Bill Hart
>  * David Harvey
>  * Josh Kantor
>  * Kiran Kedlaya
>  * David Kohel
>  * Ted Kosan
>  * Willem Jan Palenstijn
>  * Robert Miller
>  * Bobby Moretti
>  * Rich Morin
>  * Minh Nguyen
>  * Jaap Spies
>  * William Stein
>  * Carl Witty
>  * Paul Zimmermann
>
> Cheers,
>
> Michael Abshoff (release chair), William Stein
>
> * Major Features, New Spkgs and Bugfixes
>
>  * Python is now built with ucs4
>  * FLINT was updated to the 1.0.5 release
>  * Many bug fixes and also a couple of memory leak fixes
>  * Integrate a fix to the mpfr library so we no longer smash
>the stack with high precisions
>  * Fix a long standing, hard to hit bug in the maxima interface
>
> * Known Issues with 2.10:
>
>  * The mpfr fix for smashing the stack causes speed regressions,
>especially for multiplications. This ought to be fixed for the
>upcoming mpfr 2.3.1 release.
>  * Any non-standard Python packages that build native components
>need to be rebuilt due to the switch from ucs2 to ucs4
>
> We closed 83 tickets.  For details see
>
>http://trac.sagemath.org/sage_trac/milestone/sage-2.10
>
> * Upcoming Releases
>
> The next release will be 2.10.1, chaired by Michael Abshoff. The
> release
> is planned for the 25th of January, 2008. Potential features include:
>
>  * OSX 10.5 support in 64 bit mode [this will depend on fixing
>a vexing segfault in libSingular and getting disutils to
>play nice with numpy]
>  * gcc 4.3 support [depends on the update to eclib.spkg]
>  * FreeBSD support out of the box
>
> * Doctesting Coverage
>
> For 2.9.3 we had:
> Overall weighted coverage score:  35.2%
> Total number of functions:  18100
>
> We increased coverage by 0.2% in 2.10 while adding 196 extra
> functions:
> Overall weighted coverage score:  35.4%
> Total number of functions:  18296
>
> * Closed Tickets:
>
> Merged in alpha4:
>
> #1266: Mike Hansen: sage.functions.piecewise is missing from
>reference manual
> #1289: Mike Hansen: serious problems with how ceil and floor
> #1640: Alex Ghitza: missing documentation Elliptic Curve for
>heegner_discriminants
> #1705: Martin Albrecht: factorization of polynomials over
>non-prime finite fields is TOTALLY BROKEN in Sage
>[we now throw a not implemented error on this case and
>will fix this once upstream, i.e. Singular fixes this]
> #1727: Mike Hansen: truth value of inequalities not as expected
> #1734: Robert Bradshaw, William Stein: variable jmol translucency
> #1736: Alex Ghitza: sturm_bound() not working on spaces of cusp
>forms


 #1745: Benjamin Antieau: bug in point enumeration mod p.

The above patch was wrongly attributed to William, so this email
should set the record straight. We take credit very seriously, so
please inform us of any mistakes we make in the release notes. I have
already updated the official release notes on the web. Once more
apologies to Benjamin.

Cheers,

Michael


> #1747: Martin Albrecht: speed improvement for
>mq.SR.polynomial_system
> #1774: Bill Allombert: potential very serious problems with
>SEA and PARI-2.3.3
> #1783: Mike Hansen: fix latex errors with fraction field elements
> #1786: Mike Hansen: coercion error with fraction field of
&g

[sage-support] Re: Sage 2.10 released!

[mabshoff]

On Jan 19, 12:06 pm, bill purvis <[EMAIL PROTECTED]> wrote:
> On Saturday 19 January 2008, mabshoff wrote:> Hello folks,
>
> > Sage 2.10 has been released on January 18th, 2008. It is available at
>
> >http://sagemath.org/download.html
>
> > * About Sage (http://www.sagemath.org)

Hi Bill,

> As a relative newbie to sage-devel can I ask:
> Is it safe to assume that 2.10 is essentially the same as 2.10.alpha4?
> If not can I simply apply a few fixes to save the effort of downloading,
> compiling and testing everything?

You can run ./sage -upgrade which will only download the latest bits
and pieces and compile those. From alpha4 it is only the
sage-2.10.spkg (and also some other smaller spkgs like ext). There is
no tricial way to get and apply the patches [it is possible, just not
tricial] since this time we did some merging post alpha4 that got
attributed to alpha4.

> Bill

Cheers,

Michael

> +---+
> | Bill Purvis, Amateur Mathematician|
> |  email: [EMAIL PROTECTED]  |
> |  http://bil.members.beeb.net |
> +---+
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Sage 2.10 released!

[mabshoff]

Hello folks,

Sage 2.10 has been released on January 18th, 2008. It is available at

   http://sagemath.org/download.html

* About Sage (http://www.sagemath.org)

Sage is developed by volunteers and combines 71 open source packages.
It is available for download from sagemath.org and its mirrors in
source or binary form. If you have any questions and/or problems
please report them to the google groups sage-devel, sage-support,
sage-forum or sage-newbie. You can also drop by in #sage-devel in
freenode.

-

We did not achieve all the goals we had set ourselves for Sage 2.10
and also decided to release slightly before the planned date.  The
reason was one hand the much better than expected stability of
2.10.alpha4 [it build and tested perfectly on all out test platforms]
and on the other hand the need to have a stable release for Bug Day 9
planned for this Saturday.

The following people contributed to this release:

 * Michael Abshoff
 * Martin Albrecht
 * Bill Allombert
 * Robert Bradshaw
 * Craig Citro
 * Francis Clarke
 * Burcin Erocal
 * Alex Ghitza
 * Jason Grout
 * Peter Jipsen
 * David Joyner
 * Mike Hansen
 * Bill Hart
 * David Harvey
 * Josh Kantor
 * Kiran Kedlaya
 * David Kohel
 * Ted Kosan
 * Willem Jan Palenstijn
 * Robert Miller
 * Bobby Moretti
 * Rich Morin
 * Minh Nguyen
 * Jaap Spies
 * William Stein
 * Carl Witty
 * Paul Zimmermann

Cheers,

Michael Abshoff (release chair), William Stein

* Major Features, New Spkgs and Bugfixes

 * Python is now built with ucs4
 * FLINT was updated to the 1.0.5 release
 * Many bug fixes and also a couple of memory leak fixes
 * Integrate a fix to the mpfr library so we no longer smash
   the stack with high precisions
 * Fix a long standing, hard to hit bug in the maxima interface

* Known Issues with 2.10:

 * The mpfr fix for smashing the stack causes speed regressions,
   especially for multiplications. This ought to be fixed for the
   upcoming mpfr 2.3.1 release.
 * Any non-standard Python packages that build native components
   need to be rebuilt due to the switch from ucs2 to ucs4

We closed 83 tickets.  For details see

http://trac.sagemath.org/sage_trac/milestone/sage-2.10

* Upcoming Releases

The next release will be 2.10.1, chaired by Michael Abshoff. The
release
is planned for the 25th of January, 2008. Potential features include:

 * OSX 10.5 support in 64 bit mode [this will depend on fixing
   a vexing segfault in libSingular and getting disutils to
   play nice with numpy]
 * gcc 4.3 support [depends on the update to eclib.spkg]
 * FreeBSD support out of the box

* Doctesting Coverage

For 2.9.3 we had:
Overall weighted coverage score:  35.2%
Total number of functions:  18100

We increased coverage by 0.2% in 2.10 while adding 196 extra
functions:
Overall weighted coverage score:  35.4%
Total number of functions:  18296

* Closed Tickets:

Merged in alpha4:

#1266: Mike Hansen: sage.functions.piecewise is missing from
   reference manual
#1289: Mike Hansen: serious problems with how ceil and floor
#1640: Alex Ghitza: missing documentation Elliptic Curve for
   heegner_discriminants
#1705: Martin Albrecht: factorization of polynomials over
   non-prime finite fields is TOTALLY BROKEN in Sage
   [we now throw a not implemented error on this case and
   will fix this once upstream, i.e. Singular fixes this]
#1727: Mike Hansen: truth value of inequalities not as expected
#1734: Robert Bradshaw, William Stein: variable jmol translucency
#1736: Alex Ghitza: sturm_bound() not working on spaces of cusp
   forms
#1745: William Stein: bug in point enumeration mod p.
#1747: Martin Albrecht: speed improvement for
   mq.SR.polynomial_system
#1774: Bill Allombert: potential very serious problems with
   SEA and PARI-2.3.3
#1783: Mike Hansen: fix latex errors with fraction field elements
#1786: Mike Hansen: coercion error with fraction field of
   multivariate polynomials
#1788: Robert Bradshaw: cython annotation in notebook
#1790: Michael Abshoff: numerical noise doctest failure with
   gcc 4.2.2/x86-64
#1791: Michael Abshoff: numerical noise doctest failure with
   gcc 4.2.2/x86
#1797: Mike Hansen: add another doctest for expansion of
   elementary symmetric functions
#1799: Carl Witty: fix spelling errors in qqbar.py
#1803: Michael Abshoff: Add HISTORY.txt to $SAGE_ROOT
#1806: Michael Abshoff: bump various spkgs due to Sage 2.10
   upgrade (python ucs4 related)
#1807: Mike Hansen: fix combinatorics section of the reference
   manual
#1808: Martin Albrecht: three simple LaTeX fixes
#1825: William Stein: rewrite the script in devel/doc/ref
   so that anybody can add new auto-generated files to
   the reference manual via hg.
#1827: William Stein: plot3d/transform.pyx test failure
#1828: William Stein: add 3d graphics to the reference manual

Merged in alpha3:

#1428: Mike Hansen: add SVD method to matrix_compl

[sage-support] Re: Problems with standalone python/sage scripts

[mabshoff]

On Jan 18, 9:09 pm, "Georg Grafendorfer"
<[EMAIL PROTECTED]> wrote:
> Hi Robert, excuse me for the response delay, i did not notice your question,
>
> Does
>
>
>
> > #!/path/to/sage_root/local/bin/sage-python
>
> > work?
>
> no, the output is in both cases (sage -python as well as sage-python):
> bash: ./exp1.sage: /home/georg/Daten/.System/bin/sage/local/bin: bad
> interpreter: Keine Berechtigung
> where "Keine Berechtigung = no permission"
> it doesn't even work as root, but the file is executable...

What file system is home one? What file system in  /home/georg/Daten
on?  How do you mount them, i.e. with what options? Check /proc/mounts
for that. Does any script work from that particular location?

Can you run the script with "strace -f ./BMV.sage" and post the
output? That might be long, so please consider tarring up the result
and posting a link.

> Georg

Cheers,

Michael

--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: can one modify how xdvi executes when "show" is invoked?

[mabshoff]

Hi,

Rick Pember wrote:
> Hi,
>
> I'm running sage remotely (see previous post to see why) and "show"
> makes a somewhat blurry
> image. I found that using xdvi on a .dvi file gave me the same problem
> -- so the problem is with
> my version of xdvi, not sage -- but that if I modifed my execution of
> xdvi (e.g.,  "xdvi -density 15
>  -nogrey -s 8" )  I then could get a clear image.

You probably need to turn on anti alias with xdvi.

> Is there some way to tell sage to execute xdvi in this way when show
> is invoked? I've tried
> putting a shell script named xdvi in my path (ahead of /usr/bin)  that
> executes
> "xdvi -density 15 -nogrey -s 8" , but that seems only to confuse sage
> -- "show" then

The script should look something like

#!/bin/sh
/path/to/xdvi -density 15 -nogrey -s 8 $@

> appears to no-op.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: show doesn't work under Ubuntu 7.10

[mabshoff]

On Jan 18, 6:03 am, Rick Pember <[EMAIL PROTECTED]> wrote:
> Hi,

Hi Rick,

> I am a Sage "newbie" but I'm convinced this is not a newbie issue:
>
> Just today, I both built version 2.9.3 and downloaded binary versions
> 2.9.1 and 2.9.2 to
> on an Intel Core Duo box running  Ubuntu 7.10.
>
> The symptom is that a command like
>
> show(g)
>
> simply returns with no action occurring. sage otherwise appears to
> work.
>
> I saw that the same problem was reported under sage-devel a couple of
> months ago
>
> http://groups.google.com/group/sage-devel/browse_thread/thread/78f8b6...
>
> but did not see any resolution, other than a ticket number being
> assigned
> (How do tickets work? I couldn't find any info.)
>
> I was able to download sage on a remote machine and successfully
> execute "show" remotely, so I'm convinced the problem is not just an X
> display problem or a problem with my installation procedure.
> These other machines are not running Ubuntu.

The issue has been fixed in Sage 2.10 which should be out in the next
24 hours. The issue was tracker at #975 which you can find at

http://trac.sagemath.org/sage_trac/ticket/975

Over there you will also find all the other open issues and with the
navigation bar you can play around with Sage's trac installation.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://www.sagemath.org
-~--~~~~--~~--~--~---

[sage-support] Re: Elementary symmetric function expansion (bug?)

[mabshoff]

On Jan 16, 10:30 pm, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> On Jan 16, 10:23 pm, BFJ <[EMAIL PROTECTED]> wrote:
>
> > Hi,
>
> > First of all, I don't know whether to post about this here, or file a
> > bug report, or something else. I'm new to community developed
> > software. Let me know if this isn't the right place to report the
> > following.
>
> > I'm using sage 2.9.2 and 2.9.3 on different (both PPC) machines, both
> > running OS X 10.4.
> > Consider the following commands which should expand an elementary
> > symmetric function as a polynomial in 3 variables:
>
> 
>
> Hi BJF,
>
> I can reproduce the problem with 2.9.3, but it is fixed in
> 2.10.alpha3:
>
> --
> | SAGE Version 2.10.alpha3, Release Date: 2008-01-14 |
> | Type notebook() for the GUI, and license() for information.|
> --
>
> sage: sage: k=SFAElementary(QQ)
> sage: sage: f=k([2])
> sage: sage: f.expand(3)
> x0*x1 + x0*x2 + x1*x2
> sage:
>
> Mike Hansen did post a patch that was merged early on in 2.10.alpha2
> or so. He might be able to tell you if the bug was fixed "on purpose"
> or "by accident". 2.10 should be out by the weekend or maybe Monday.
> Alpha3 is available at
>
> http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/sag...
>
> Mike: could you add a doctest that tests this behavior?

Oops, looking at the actual failure this is a libSingular bug. I can't
find the exact ticket number at the moment, but I remember merging
that fix.

Sorry for the double post.

> > Similar commands, but on objects from SFASchur, SFAPower, SFAMonomial,
> > etc.., work perfectly. It seems to be only SFAElementary that has a
> > problem.
> > Thanks,
>
> > BFJ
>
> Cheers,
>
> Michael

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Elementary symmetric function expansion (bug?)

[mabshoff]

On Jan 16, 10:23 pm, BFJ <[EMAIL PROTECTED]> wrote:
> Hi,
>
> First of all, I don't know whether to post about this here, or file a
> bug report, or something else. I'm new to community developed
> software. Let me know if this isn't the right place to report the
> following.
>
> I'm using sage 2.9.2 and 2.9.3 on different (both PPC) machines, both
> running OS X 10.4.
> Consider the following commands which should expand an elementary
> symmetric function as a polynomial in 3 variables:



Hi BJF,

I can reproduce the problem with 2.9.3, but it is fixed in
2.10.alpha3:

--
| SAGE Version 2.10.alpha3, Release Date: 2008-01-14 |
| Type notebook() for the GUI, and license() for information.|
--

sage: sage: k=SFAElementary(QQ)
sage: sage: f=k([2])
sage: sage: f.expand(3)
x0*x1 + x0*x2 + x1*x2
sage:

Mike Hansen did post a patch that was merged early on in 2.10.alpha2
or so. He might be able to tell you if the bug was fixed "on purpose"
or "by accident". 2.10 should be out by the weekend or maybe Monday.
Alpha3 is available at

http://sage.math.washington.edu/home/mabshoff/release-cycles-2.10/sage-2.10.alpha3.tar

Mike: could you add a doctest that tests this behavior?

> Similar commands, but on objects from SFASchur, SFAPower, SFAMonomial,
> etc.., work perfectly. It seems to be only SFAElementary that has a
> problem.

> Thanks,
>
> BFJ

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: jmol and Camino

[mabshoff]

On Jan 16, 8:55 pm, "Hector Villafuerte" <[EMAIL PROTECTED]> wrote:
> Hi,

Hi Hector

> I get this error on Camino (... don't ask):

I am afraid to ask why :)

Which camino release are you running? The latest was released on
December 5, 2007, so I must say that I am suprised that  the Camino
project is still alive. What java runtime are you using? Could it be
that there are incompatibilities between those two, since Apple
release a new JDK a while back? Which OSX release are you running?

> ReferenceError: _jmolInitCheck is not defined
>
> While trying to plot this:
>
> var('v')
> f1 = (x, x*sin(v), x*cos(v))
> parametric_plot3d(f1, (x,0,5), (v,0,2*pi))
>
> I must say that this works great in Firefox and Safari, I just thought
> you would like to know.

:)


> Best,
> --
>  Hector

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Doc Day 1 announcement: January 17th, 2008, 9am-5pm PST

[mabshoff]

Hello folks,

after doing 8 Bug Days the time has come to spend some more time on
the documentation, which also needs a lot of work. So after some
discussion in IRC we decided to get together in IRC at the above date
and time and work on the documentation. Since we have never done a doc
day it isn't 100% clear to me how the whole thing will go down, but I
assume we will just go with the flow.

Thoughts? Suggestions?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Problems with standalone python/sage scripts

[mabshoff]

On Jan 16, 12:40 am, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Jan 15, 2008 3:00 PM, Georg Grafendorfer
>
>
>
> <[EMAIL PROTECTED]> wrote:
> > Hi Robert,
>
> > > 1. Is this a clean, from-source build of sage-2.9.3?
>
> > Yes, i tried it out on two different systems now, Athlon XP, and Core Duo,
> > both running on Debian Etch, and both show the same behaviour
>
> > > 2. What is the output of /usr/bin/env for you? Mine (Intel OS X
> > > 10.5.1) doesn't mention sage at all, although mysteriously things are
> > > working for me.
>
> > the  SAGE_ROOT directory is included in my path, i even defined a global
> > variable with this name, i.e.  SAGE_ROOT.
>
> > > > #!/usr/bin/env sage -python
> > > > import sys
>
> > > For me, I get
> > > $ ./BMV.sage
> > > $
>
> > you response encouraged me to try out some more things, and if i change the
> > first line to
>
> > #!/usr/bin/env sage-python
> > instead of
>
> > #!/usr/bin/env sage -python
> >  (note that there is no space anymore) things work as excepted (at least
> > import sys and print "Hello World"), seems like my /usr/bin/env does not
> > like the space between (#!/usr/bin/env "sage -python" does not work either)
>
> > but however, using
>
> > #!/path/to/sage_root/sage-python
> > import sys
>
> > still does not work and shows the same strange "mouse behaviour" as
> > described in my original posting, on both systems!!
> >  no idea why the second one does not work on my systems!!
>
> > > However, when I run the sage itself, it looks like it's pointing at
> > > some weird version:
> > > $ /usr/bin/env sage
> > > --
> > > | SAGE Version 2.9.2, Release Date: 2008-01-05   |
> > > | Type notebook() for the GUI, and license() for information.|
> > > --
> > > Loading SAGE library. Current Mercurial branch is: demo
>
> > $/usr/bin/env sage
>
> > brings me to the sage prompt as expected
>
> > > Hopefully this helps, although I have a feeling this thread isn't
> > > over...
>
> > Anyway, at least i found a partial solution to carry on with, but it seems
> > as there are still some things to clarify, especially the mouse thing
> > concerning the import sys
>
> > Thank you very much,
> >  Georg
>
> I consider all this a bug, and it's definitely a problem on numerous _linux_
> systems.  This problem doesn't happen at all on OSX.  I've made it
>http://trac.sagemath.org/sage_trac/ticket/1789

I thought I have seen this before and the issue was a buggy env, i.e.

#!/usr/bin/env sage -python

while

#!/usr/bin/env sage

should work. There was a thread about this in one of the sage-* Google
groups. It boiled down to that env should work fine with multi
arguments while some buggy env can't handle it. Maybe we can add some
script that does the equivalent of "sage -python", maybe "local/bin/
sage-python" does that job, but I assume the env isn't set up
properly. In that case we should change the documentation since it has
been popping up over and over these days.

> William

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Computations with Lie algebras?

[mabshoff]

On Jan 14, 7:09 pm, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:

Hi,

> I just looked at the GAP docs, and I don't quite see how to do what I
> want (decomposing some random thing into irreducibles, or at least
> getting the multiplicity of the trivial representation), but I'm sure
> it's possible.
>
> What I do know is that Magma has some of this functionality,
> implemented in a fashion that I can understand. So that could be one
> possible interface model if we decide to wrap this functionality from
> GAP, LiE, or elsewhere.
>
> Kiran

We already have an optional LiE spkg at 
http://sagemath.org/packages/optional/lie-2.2.2.p3.spkg
- and I think Mike Hansen did work at some point to wrap
functionality. Somebody please correct me if I got that somehow mixed
up.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Sage classroom server using vmware

[mabshoff]

Hi Jason,

On Jan 14, 6:51 pm, Jason Grout <[EMAIL PROTECTED]> wrote:
> I'd like to set up a sage notebook for my class to use.  At the AMS
> meetings, William mentioned that the easiest way to do this was to use
> the vmware appliance to run the notebook (even if I plan to run the
> appliance under linux).  I presume the reasons have to do with security.
>   I have the vmware appliance running now, so the initial setup is all
> good.  However, the address is a 192.168.x.x address, which I presume is
> not visible outside of my network.  Is there a way to get the sage
> vmware appliance visible to the world so students could use it from
> anywhere?  When I installed vmware, I just used the defaults everywhere.
>   Should I have chosen something different when answering questions
> about the networking?

You need to set up bridged networking instead of NAT.

> Are there any security issues I should be aware of?  I don't care if the
> vmware image gets messed up---I assume I can just reset it.  I can back
> up my notebook worksheets so I don't lose work.  I assume that people
> escaping the vmware appliance and getting into my computer is a
> non-issue; is that correct?

It is theoretically possible, but the skill set needed to do that
would make you want to turn off your computer and hide it somewhere in
a shielded room, so I wouldn't worry about it too much.

> In return for answers, I'll write an entry in the FAQ or a new page in
> the wiki :).

Nice. John Voight is running a bunch of VMWare images on top of
Windows with DSage, so he might have some input on the setup if it
goes beyond enabling bridged networking.

> Thanks,
>
> Jason


Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Problems with 2.9.3, jmol applet, jsMath fonts

[mabshoff]

On Jan 14, 6:17 pm, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> Dear Michael,
>
> On Jan 14, 2008 11:58 AM, mabshoff
>
> <[EMAIL PROTECTED]> wrote:
>
> (...)
>
> > > But, so far, I haven't had any problems while running my old stuff...
>
> > The doctest that fails communicates with another Sage instance. I have
> > seen this fail on occasion with systems with low memory. Since your
> > gcc is tuned for "i486-linux-gnu" could you enlighten us how much
> > memory your system has available? Maybe "i486-linux-gnu" is the
> > default gcc build target on Debian - I no longer have access to a 32
> > bit Debian box.

Hi Luis,

> It has 1Gb...  Is that already considered "low memory"?

I wouldn't think so, the problem happened with 256 MB or so. What
might be an issue is something else (i.e. the tp_new issue), which
seems to be finally fixed upstream in python. Please report if the
problem persists in the the next release. Unless I can come up with a
reproducible test case on my end I am reluctant to open a ticket. If
anybody else can reproduce this please speak up.

> :-) As I
> said, so far it hasn't given me any trouble, although I use Sage at
> home less than at work.  About gcc, I don't know much about it at all.
> I just did "apt-get install gcc" or something like that...

ok, somebody else might know this hopefully.

> Thanks!
>
> Luis

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Problems with 2.9.3, jmol applet, jsMath fonts

[mabshoff]

On Jan 13, 12:01 am, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> Dear Michael (and Ted, and all),
>
> Thanks for your reply!
>
> On Jan 11, 2008 10:56 AM, mabshoff
>
>
>
> <[EMAIL PROTECTED]> wrote:
> > > Here is the error:
>
> > > --
> > >  **
> > > File "sage0.py", line 47:
> > > sage: a^3
> > > Expected:
> > > 8
> > > Got:
> > > 
> > > **
> > > File "sage0.py", line 52:
> > > sage: V.gens()
> > > Expected:
> > > ((1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1))
> > > Got:
> > > (...)
>
> > Hmm, is that the only error? Could you post the exact output from gcc -
> > v please?
>
> Yes, that is the only error given by "make test".  And, I get:

Mh, so far I haven't seen anything odd about that doctest. I checked
the valgrind logs from that session from my 2.9.2 test run and nothing
popped up.

> debian[~]$ gcc -v
> Using built-in specs.
> Target: i486-linux-gnu
> Configured with: ../src/configure -v
> --enable-languages=c,c++,fortran,objc,obj-c++,treelang --prefix=/usr
> --enable-shared --with-system-zlib --libexecdir=/usr/lib
> --without-included-gettext --enable-threads=posix --enable-nls
> --program-suffix=-4.1 --enable-__cxa_atexit --enable-clocale=gnu
> --enable-libstdcxx-debug --enable-mpfr --with-tune=i686
> --enable-checking=release i486-linux-gnu
> Thread model: posix
> gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)
>
> But, so far, I haven't had any problems while running my old stuff...

The doctest that fails communicates with another Sage instance. I have
seen this fail on occasion with systems with low memory. Since your
gcc is tuned for "i486-linux-gnu" could you enlighten us how much
memory your system has available? Maybe "i486-linux-gnu" is the
default gcc build target on Debian - I no longer have access to a 32
bit Debian box.

Cheers,

Michael

> > > Then I tried to check the new 3D plots.  (It worked in my office
> > > computer, and it looks great!  Kudos!)  When evaluating the Moebius
> > > band sent by D. Joyner, I get a empty rectangle where the graph should
> > > appear, and the lower left (status) bar of Firefox shows "Downloading
> > > JmolApplet..."  But nothing happens.  (...)
>
> > I have no idea about jmol, somebody else needs to figure that one out.
> > A shot in the dark: Which java do you have installed? I believe the
> > version from gcc causes trouble, so installing a "proper" JDK 1.5 or
> > 1.6 from Sun *might* solve the problem.
>
> OK, I had installed (manually) java 1.5 from Sun.  But, although I
> installed the plugins manually, the system did not see the binaries.
> It used gij.  It took me a while to find that out.  What fixed it was
> installing java with APT and telling the system to use it.
>
>   $ sudo apt-get install sun-java5-bin sun-java5-doc \
> sun-java5-fonts sun-java5-jre sun-java5-plugin
>
>   $ sudo update-alternatives --config java
>
>   (Choose "java", not "gij??".)
>
> Now it works!  Thanks for your help!
>
> > > Finally, (and this has been a "problem" for a long time, also in my
> > > laptop, running Debian Etch again), I have jsMath fonts installed, but
> > > sage (or Firefox) does not find it...
>
> > I believe the current version of jsmath is 3.4, so that might be the
> > cause.
>
> I didn't try that since Ted's suggestion worked for me!
>
> Ted wrote:
> > You may try installing the fonts manually by creating a .fonts
> > directory (note the period in front of the f) in your home directory
> > and then unzipping the TeX-fonts-25.zip into it.  The fonts can be
> > obtained from
> > (http://www.math.union.edu/~dpvc/jsmath/download/jsMath-fonts.html).
>
> Does the new version of jsMath come with those??  I am just surprised
> that I did not have to do that with my office computer running Fedora
> Core 6.  In fact, now that I've checked it, it seems that my office
> computer doesn't even have jsMath installed at all (except what comes
> with Sage).  I am bit puzzled, but it's working for me...
>
> Thanks Michael and Ted for your help!
>
> Luis
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Problems with 2.9.3, jmol applet, jsMath fonts

[mabshoff]

On Jan 11, 5:15 am, "Luis Finotti" <[EMAIL PROTECTED]> wrote:
> Dear all,

Hi Luis,

> I've been having problems running Sage 2.9 in my home computer: Athlon
> XP 2800+, running Debian Etch, with kernel 2.6.18-4-k7.
>
> I tried the pre-compiled versions, but they game many errors when
> testing, which I assumed it was due to an old CPU (as I've seen
> mentioned before).  So, I compiled 2.9.3 from the source (which took a
> long, long time, also as mentioned by someone else before).
>
> Then, "make test" fails for "sage -t
> devel/sage-main/sage/interfaces/sage0.py" (only).
>
> Here is the error:
>
> --
>  **
> File "sage0.py", line 47:
> sage: a^3
> Expected:
> 8
> Got:
> 
> **
> File "sage0.py", line 52:
> sage: V.gens()
> Expected:
> ((1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0), (0, 0, 0, 1))
> Got:
>
> **
> File "sage0.py", line 62:
> sage: g = V.0;  g
> Expected:
> (1, 0, 0, 0)
> Got:
>
> **
> File "sage0.py", line 69:
> sage: s('%s.parent()'%g.name())
> Expected:
> Vector space of dimension 4 over Rational Field
> Got:
>
> **
> File "sage0.py", line 74:
> sage: s('x = 5')
> Expected:
> 5
> Got:
>
> **
> File "sage0.py", line 78:
> sage: s('x')
> Expected:
> 5
> Got:
>
> **
> File "sage0.py", line 85:
> sage: a
> Expected:
> 10
> Got:
>
> **
> File "sage0.py", line 91:
> sage: s3('"x"')
> Expected:
> 8
> Got:
>
> **
> File "sage0.py", line 93:
> sage: s('x')
> Expected:
> 5
> Got:
>
> **
> 1 items had failures:
>9 of  21 in __main__.example_1
> ***Test Failed*** 9 failures.
> For whitespace errors, see the file .doctest_sage0.py
>  [3.9 s]
> exit code: 256
>
> --
> The following tests failed:
>
> sage -t  devel/sage-main/sage/interfaces/sage0.py
> Total time for all tests: 3.9 seconds
> 

Hmm, is that the only error? Could you post the exact output from gcc -
v please?

> I am not sure what this affects, but many of my old computations still work.
>
> Then I tried to check the new 3D plots.  (It worked in my office
> computer, and it looks great!  Kudos!)  When evaluating the Moebius
> band sent by D. Joyner, I get a empty rectangle where the graph should
> appear, and the lower left (status) bar of Firefox shows "Downloading
> JmolApplet..."  But nothing happens.  Opening the Java Console (by
> right clicking at the space for the plot), I get:
>
> 
> Java Plug-in 1.5.0_02
> Using JRE version 1.5.0_02 Java HotSpot(TM) Client VM
> User home directory = /home/finotti
>
> 
> c:   clear console window
> f:   finalize objects on finalization queue
> g:   garbage collect
> h:   display this help message
> l:   dump classloader list
> m:   print memory usage
> o:   trigger logging
> p:   reload proxy configuration
> q:   hide console
> r:   reload policy configuration
> s:   dump system and deployment properties
> t:   dump thread list
> v:   dump thread stack
> x:   clear classloader cache
> 0-5: set trace level to 
> 
>
> load: class JmolApplet not found.
> java.lang.ClassNotFoundException: JmolApplet
> at sun.applet.AppletClassLoader.findClass(Unknown Source)
> at java.lang.ClassLoader.loadClass(Unknown Source)
> at sun.applet.AppletClassLoader.loadClass(Unknown Source)
> at java.lang.ClassLoader.loadClass(Unknown Source)
> at sun.applet.AppletClassLoader.loadCode(Unknown Source)
> at sun.applet.AppletPanel.createApplet(Unknown Source)
> at sun.plugin.AppletViewer.createApplet(Unknown Source)
> at sun.applet.AppletPanel.runLoader(Unknown Source)
> at sun.applet.AppletPanel.run(Unknown Source)
> at java.lang.Thread.run(Unknown Source)
> Caused by: java.io.IOException: open HTTP connection failed.
> at sun.applet.AppletClassLoader.getBytes(Unknown Source)
> at sun.applet.AppletClassLoader.access$100(Unknown Source)
> at sun.applet.AppletClassLoader$1.run(Unknown Source)
> at java.security.AccessController.doPrivileged(Native Method)
> ... 10 more
> load: class JmolApplet not found.
> java.lang.ClassNotFoun

[sage-support] Re: unable to install from tarballs in fedora7

[mabshoff]

On Jan 10, 1:06 pm, esdc <[EMAIL PROTECTED]> wrote:
> Hi friends,

Hi E,

> I am trying to install sage on an acer travelmate4000 with fedora7, I
> have tried with the tar files of version 2.9.2
>
> I have tried the debian one and the rhel5 one, but both gave the same
> error, here's what I did:
>
> moved the tar file to /usr/local/
>
> unpacked
>
> run sage
>
> and this is what I obtained:
>



> : /usr/local/sage-2.9.2-rhel5-
> linux32bit-i686-Linux/local/lib/libpari-gmp.so.2: cannot restore
> segment prot after reloc: Permission denied
> 
>
> please, can anyone help??

You have SELinux enabled. Disabling it will solve the problem. See
http://wiki.sagemath.org/faq

This is a bug on our end and I am looking into solving this. What is
needed is a relabeling of the Sage files. If you come up with a script
to do that please let us know.

> thanks a lot!
>
> E

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: very long compilation

[mabshoff]

On Jan 7, 10:48 am, JMD <[EMAIL PROTECTED]> wrote:
> SAGE-2.9.3 compilation time is very long on on my Dell laptop :
> inspiron 9300 with a Pentium 4m 1.73 Ghz.
> More than 4 hours
> Most of the time is taken by the package ATLAS
> During compilation I see "-DATL_ARCH_CoreDuo" which seems to indicate
> that th Pentium 4m is recognized as a CoreDuo.

Hi JMD,

this is indeed the case a known ATLAS issue that certain Pentium M are
misidentified as CoreDuo. We are hoping that this issue will be
resolved upstream soon. The long compilation time is also cause by
certain CPUs not having a profile present in ATLAS, which we plan to
resolve soon.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Upgrade failed

[mabshoff]

On Jan 6, 11:21 pm, Simon King <[EMAIL PROTECTED]> wrote:
> Dear Sage-Support team,
>
> when i did sage -upgrade, it finished with the following error:
>
> Finished extraction
> There is no spkg-install script, no setup.py, and no configure script,
> so I do not know how to install /usr/local/sage-2.8.6/spkg/standard/
> doc-2.9.3.spkg.
> make: *** [installed/doc-2.9.3] Error 1
>
> What happened, and what can i do?

Hi Simon,

you might have had bad timing and updated just as William released
2.9.3. So try deleteing the old doc-2.9.3.spkg and run -upgrade again.
If that fails again please post the md5sum of the doc-2.9.3.spkg you
downloaded.

>
> Yours
>Simon

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: parametric_plot3d changes ignored

[mabshoff]

On Jan 6, 8:44 pm, tclendenen <[EMAIL PROTECTED]> wrote:

Hi Tim,

> I was trying the parametric_plot3d backed by Jmol by putting the
> following into a worksheet cell:
>
> var('t')
> c = 2*pi
> parametric_plot3d( (cos(t), sin(t), t), (0, c), plot_points=100,
> color="red")
>
> It worked as expected, however when I changed the second line to c =
> 3*pi then did shift+enter, it still plots the graph using the first
> value of c (2*pi).  Even after saving the worksheet and restarting the
> virtual machine, the change was not recognized.
>
> I'm using the Sage 2.9.1 virtual appliance that has been upgraded to
> 2.9.3.
>
> Thanks,
> Tim

The patch at http://trac.sagemath.org/sage_trac/ticket/1693 ought to
fix that. Can you report back if it solves the problem for you?

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: SIGILL with sage 2.9.1 on Fedora 8, AthlonXP2000

[mabshoff]

On Jan 5, 11:41 am, jara <[EMAIL PROTECTED]> wrote:



Hi jara,

> Hello,
> I replaced the libraries with ones from my distribution and sage now
> works :-)

Good, I will add that solution to the FAQ then shortly.

> But regarding the precompiled downloads there are imho 3 options what
> to do:
> 1. don't mark binaries as "i686" because they aren't i686-compatible.
> Mark them as Pentium4 (i886??) or something.

Good point.

> 2. recompile for i686 and sage will run slowly but it will at least
> run. Place a notice in a README that recompilation from sources is
> highly recommended for the best performance (remember Gentoo ;-)

Well, that has been more or less the way it used to be. We used to
provide binaries as a convenience since the vast majority of people
used to build from source. That has changed now and we need to deal
with that.

> 3. provide several compiled versions of atlas - each compiled for
> different architecture - and select the best one on installation. For
> example Fedora 8 has following packages of atlas:
> atlas.i386   3.6.0-12.fc8   fedora
> atlas-3dnow.i386 3.6.0-12.fc8   fedora
> atlas-sse.i386   3.6.0-12.fc8   fedora
> atlas-sse2.i386  3.6.0-12.fc8   fedora

Well, that would increase download size for everybody which is
something we have been trying to shrink lately. I am not sure how we
would do provide the right ATLAS dynamically - Debian's ATLAS
struggles from the same problem, so if you have any ideas [or even
better code] in that direction I would be very glad to hear about it.

> Thanks for your help,

No problem, that is what we are here for.

> -- Jara

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: package for fedora

[mabshoff]

On Jan 5, 1:18 am, Neal Becker <[EMAIL PROTECTED]> wrote:
> mabshoff wrote:
>
> > On Jan 5, 12:57 am, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:
>
> > Hi,
>
> >> I'm interested in packaging sage for fedora.
>
> > Ok, we have a google group debian-sage that despite its name might be
> > a better place to discuss this.
>
> >> I am a newbie to sage.  One
> >> thing I notice right away, is that sage says it is 'batteries
> >> included'.  I
> >> think that a sage package on fedora should co-operate with existing
> >> packages.  For example, there already are numpy, scipy, octave,
> >> maxima.
> >> Can sage use such existing packages?
>
> > In theory: yes. But in practice we do build all those since using
> > different version numbers can cause trouble. But since we have an
> > extensive test suite you should catch those issues. We are currently
> > on the way to package Sage on Debian and we plan to coordinate with
> > the maintainers of the packages that are already in Debian.
>
> > You might want to make a list of what already is in FC and what is
> > not. Then you should compare version numbers.
>
> Thanks for the info.  Is there a list of what packages sage could make use
> of?  Is this the same thing being planned for debian? (I mean, the idea of
> re-using some existing packages)

Hi Neal,

check out the http://wiki.sagemath.org/DebianSAGE and the pages linked
from there. The idea with Debian is to coordinate with the individual
maintainers which we either know or seem sympathetic to our effort.
Since we do package the a lot of mathematical open source software and
we are usually on the cutting edge they welcome that we want to help
out.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: package for fedora

[mabshoff]

On Jan 5, 12:57 am, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:

Hi,

> I'm interested in packaging sage for fedora.

Ok, we have a google group debian-sage that despite its name might be
a better place to discuss this.

> I am a newbie to sage.  One
> thing I notice right away, is that sage says it is 'batteries
> included'.  I
> think that a sage package on fedora should co-operate with existing
> packages.  For example, there already are numpy, scipy, octave,
> maxima.
> Can sage use such existing packages?

In theory: yes. But in practice we do build all those since using
different version numbers can cause trouble. But since we have an
extensive test suite you should catch those issues. We are currently
on the way to package Sage on Debian and we plan to coordinate with
the maintainers of the packages that are already in Debian.

You might want to make a list of what already is in FC and what is
not. Then you should compare version numbers.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: SIGILL with sage 2.9.1 on Fedora 8, AthlonXP2000

[mabshoff]

On Jan 4, 10:54 pm, jara <[EMAIL PROTECTED]> wrote:
> On Jan 4, 5:38 pm, mabshoff <[EMAIL PROTECTED]
>
> dortmund.de> wrote:
>
> > Hi jara,
>
> > you are correct.The problem is that we build the binaries on a
> > relative modern CPU and hence the binaries cause problems on CPUs like
> > yours. The solution is to build from source. We plan to build on older
> > CPUs in the future, but haven't gotten the needed hardware yet.
>
> > Cheers,
>
> > Michael

Hi,

> Hi,
> I think you just need to pass -march=i686 to the gcc when building the
> precompiled
> "i686" binaries, as imho everyone else does.

the problem is ATLAS which is a tuned BLAS/Lapack library that gets
you maximum performance for a given CPU type. If you have ATLAS
libraries from your distribution you can try replacing the ones in
$SAGE_LOCAL/lib. Those would be

libatlas.a, libatlas.so, libcblas.a, libcblas.so, libf77blas.a and
libf77blas.so

Compiling ATLAS with  -march=i686 wouldn't work and destroy the
performance advantage of using it in the first place. It is a trade
off we did make and hopefully we will have more generic binaries soon.


> Also would it be possible to put back the old versions of sage in the
> download section?
> I used v2.8 and then just today I've installed v2.9. It booted fine so
> I've just deleted the old
> version altogether only to run into the bug in the very next moment.
> Now I am left with no working
> version of sage :-(

Sorry, we do not keep older binaries around due to space limitations.
Maybe somebody else on the list has something lying around and can
help you out. Building from source is surprisingly easy assuming you
have a decent compiler, i.e. gcc 3.4.x or gcc 4.1.x or later.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Sage-2.9

[mabshoff]

On Jan 4, 5:15 pm, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:

Hi Kiran,

> If I describe my setup, it might seem less uncommon than you might
> have previously thought.
>
> Mine is a 64-bit box on a network consisting mostly of 32-bit
> machines. On my machine, /usr/bin is locally mounted, and there is a
> local 64-bit gcc there.
>
> But /usr/local is NFS mounted, and there is a 32-bit gcc in /usr/local/
> bin for the benefit of the benighted masses. Persuading my sysadmin to
> instead install a local 32-bit gcc on every single 32-bit machine in
> the department is not going to happen.

I understand, but providing 32 bit compilers to 64 bit machines seems
a little silly. At least in the setups I have run we had different NFS
exports for 32 and 64 bit Linux boxen.

> So I have to switch rather than fight. My $PATH has /usr/bin ahead of /
> usr/local/bin, so ordinarily (and even in SCons if I force it to
> import my PATH environment variable) this causes no problems.
>
> But SCons defaults not to importing any environment variables, which
> means it has to come up with its own guess for the path. And what it
> comes up with on my box is:
>/usr/local/bin:/opt/bin:/bin:/usr/bin
> I have no idea why. There isn't even a directory /opt/bin on my
> system! This might be some sort of hard-coded default (either in SCons
> or in the RHEL configs somewhere).

SCons is at fault: From local/lib/scons-0.97/SCons/Platform/posix.py

if not env.has_key('ENV'):
env['ENV']= {}
env['ENV']['PATH']= '/usr/local/bin:/opt/bin:/bin:/usr/bin'


> Anyway, I'm able to manually patch the relevant SConstruct files to
> get around this, so it's not a high priority for me to get this fixed.
> But I suspect that sooner or later, someone else will run into this if
> nothing is done about it.

So as a solution I would suggest to import ENV into the PolyBoRi SCons
build system. Feel free to submit a patch for it.

> Kiran

I am also attaching the same into to #1553.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: SIGILL with sage 2.9.1 on Fedora 8, AthlonXP2000

[mabshoff]

On Jan 4, 3:49 pm, jara <[EMAIL PROTECTED]> wrote:
> Hello,
> I am getting a SIGILL with the new sage 2.9.1, precompiled binary
> (Linux-i686):
>
> [EMAIL PROTECTED] ~]$ sage
> --
> | SAGE Version 2.9.1, Release Date: 2007-12-24   |
> | Type notebook() for the GUI, and license() for information.|
> --
>
> sage: plot(sin(x)).show()
> /home/jara/bin/sage-2.9.1-Linux-i686-Linux/local/bin/sage-sage: line
> 210: 29867 Nedovolená instrukce (SIGILL)  sage-ipython -c



>
> The xorpd instruction is imho SSE2 which this CPU does not support:
> processor   : 0
> vendor_id   : AuthenticAMD
> cpu family  : 6
> model   : 6
> model name  : AMD Athlon(tm) XP 2000+
> stepping: 2
> cpu MHz : 1670.560
> cache size  : 256 KB
> fdiv_bug: no
> hlt_bug : no
> f00f_bug: no
> coma_bug: no
> fpu : yes
> fpu_exception   : yes
> cpuid level : 1
> wp  : yes
> flags   : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca
> cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow up ts
> bogomips: 3342.91
> clflush size: 32

Hi jara,

you are correct.The problem is that we build the binaries on a
relative modern CPU and hence the binaries cause problems on CPUs like
yours. The solution is to build from source. We plan to build on older
CPUs in the future, but haven't gotten the needed hardware yet.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Sage-2.9

[mabshoff]

On Dec 29 2007, 1:58 am, "Kiran Kedlaya" <[EMAIL PROTECTED]> wrote:
> I posted an install log (for SAGE 2.9.1.1) here:
>
> http://sage.math.washington.edu/home/kedlaya/install.log
>
> And re my compiler version: the result of gcc -v is:
>
> Using built-in specs.
> Target: x86_64-redhat-linux
> Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
> --infodir=/usr/share/info --enable-shared --enable-threads=posix
> --enable-checking=release --with-system-zlib --enable-__cxa_atexit
> --disable-libunwind-exceptions --enable-libgcj-multifile
> --enable-languages=c,c++,objc,obj-c++,java,fortran,ada
> --enable-java-awt=gtk --disable-dssi --enable-plugin
> --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre
> --with-cpu=generic --host=x86_64-redhat-linux
> Thread model: posix
> gcc version 4.1.2 20070626 (Red Hat 4.1.2-14)
>
> Kiran

Hello,

with Kiran's help we finally tracked this down while in IRC. SCons
picks another gcc that is in $PATH, namely one in /usr/local/bin:

Target: i686-pc-linux-gnu
Configured with: ./configure --prefix=/usr/local/pkg/gcc-4.2.0 --
enable-languages=c,c++,fortran,java,objc,obj-c++
Thread model: posix
gcc version 4.2.0

But that one is a 32 bit target, so things go *boom*.

This should be a very uncommon setup, but we still hit it. So I have
made this #1656.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: sage-2.9.1.1 build problem

[mabshoff]

On Dec 27, 4:39 pm, "Kate Minola" <[EMAIL PROTECTED]> wrote:

Hi Kate,

> When I build sage-2.9.1.1 from source on the architectures
> of interest to me, I get an error when building
>
>  mpfi-1.3.4-cvs20071125.p2
>
> gcc -fPIC -o test_mpfi test_mpfi.o  ../src/.libs/libmpfi.a
> /home/kate/sage/sage-2.9.1.1-x86_64-Linux/local/lib/libmpfr.a -lgmp
> ../src/.libs/libmpfi.a(mpfi_io.o): In function `mpfi_inp_str':
> mpfi_io.c:(.text+0x999): undefined reference to `__gmp_get_memory_functions'
>
> If I look in sage-2.9.1.1/local/lib, I notice that mpfr is built statically,
> but gmp is not.  (Is there a good reason for this?)

Default policy is to build all libraries dynamically so that we do not
need to recompile loads of dependencies. What you encounter is a bug
in mpfi.

>  If I change
> the line in gmp-4.2.1.p12/spkg-install from
>
>SAGE_CONF_OPTS="--enable-shared --disable-static"
>
> to
> SAGE_CONF_OPTS="--enable-shared"
>
> then gmp builds a static version.
>
> If I then change mpfi-1.3.4-cvs20071125.p2/spkg-install
> from
>
> ./configure --prefix="$SAGE_LOCAL" --with-mpfr-dir="$SAGE_LOCAL"
> --with-gmp-incpath="$SAGE_LOCAL"/include CFLAGS="-fPIC"
>
> to
>
> ./configure --prefix="$SAGE_LOCAL" --with-gmp-dir="$SAGE_LOCAL"
> --with-mpfr-dir="$SAGE_LOCAL" CFLAGS="-fPIC"
>
> then mpfi builds fine.

We should add the --with-gmp-dir to the mpfi configure per default and
also fix the the problem so that a dynamic gmp is accepted.

> I suspect you are NOT seeing this problem because your compile
> machines have a system gmp installed that is being picked up.
>
> Naturally, there may be a better way to fix the problem than
> I have given.
>

Cheers,

Michael

> Kate
> --
> Kate Minola
> University of Maryland, College Park
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: libstdc++.so.6: cannot handle TLS data

[mabshoff]

On Dec 25, 2:36 pm, janzo <[EMAIL PROTECTED]> wrote:
> Now the problem is building cremona, I tried to run the debug console
> ("/home/janzo/SAGE/sage-2.9.1/sage -sh") but I don't know what to do
> whit it ¿?

You seem to be running out of luck. Sage needs at least gcc 3.4 since
it requires C99 features. I am not sure what to do except to recommend
that you build gcc 3.4.6 from sources. gcc 4.1 or 4.2 might be better
suited (since we prefer gfortran or g95 over g77), but with the
underlying elderly Linux that might be a problem.

Cheers,

Michael

>
> the error log of cremona:
>
> g++: Internal error: Killed (program cc1plus)

Well, not much to fix here, gcc 3.3 shot itself in the head.

> Please submit a full bug report.
> See http://gcc.gnu.org/bugs.html> for instructions.
> For Debian GNU/Linux specific bug reporting instructions,
> see .


Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: libstdc++.so.6: cannot handle TLS data

[mabshoff]

On Dec 25, 1:29 pm, janzo <[EMAIL PROTECTED]> wrote:
> [orangebang]$ uname -a
> Linux orangebang 2.4.32-grsec+f6b+gr217+nfs+a32+fuse23+tg+++opt
> +c8+gr2b-v6.194 #1 SMP Tue Jun 6 15:52:09 PDT 2006 i686 GNU/Linux

Ok, a heavily patched 2.4 kernel without NPTL.  A 2.6 libstdc++ is
unlikely to ever work.

>
> A debian with 2.4.32 kernell :'(
> (why not 2.6!!? it's fully stable!)

They probably like grsec and the 2.4 code base has way fewer security
issues.

> I'm going to try making it again.

ok, let us know what happens.

Cheers,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: libstdc++.so.6: cannot handle TLS data

[mabshoff]

On Dec 24, 9:42 pm, "William Stein" <[EMAIL PROTECTED]> wrote:
> On Dec 24, 2007 1:27 PM, janzo <[EMAIL PROTECTED]> wrote:
>
>
>
> > Hi!
> > I'm trying to run a sage notebook in my web server (dreamhost). First
> > I get an error message about not finding libstdc++.so.6 but then I
> > found that file to download (in sage repository) and I have put it in
> > sage_path/local/bin and now I get another error log:
>
> You should build from source.
>
> 1. Get sage-2.9.1.tar fromhttp://sagemath.org/dist/src/
> 2. Read the README.txt
>
> William
>
>
>
>
>
> > : libstdc++.so.6: cannot handle TLS
> > data
> > WARNING: Failure executing code: 'import sage.misc.preparser_ipython;
> > sage.misc.preparser_ipython.magma_colon_equals=True'
> > ---
> >Traceback (most recent call
> > last)
>
> > /home/.meta/janzo/sage-2.9-debian-32bit-i686-Linux/local/bin/ > console> in ()
>
> > /home/janzo/sage-2.9-debian-32bit-i686-Linux/local/lib/python2.5/site-
> > packages/sage/misc/misc.py in ()
> > 334 return 0
> > 335
> > --> 336 from sage.misc.misc_c import prod
> > 337
> > 338 # alternative name for prod
>
> > : libstdc++.so.6: cannot handle TLS
> > data

As William suggested above the easiest solution on your end is to
build from source.  TLS stands for "thread local storage" and it is my
impression that your kernel has been compiled without NPTL [=Native
Posix Threads Linux]. So depending on your distribution you should
install libstdc++.so, which should be provided by the package that
offers recent g++ versions, i.e. gcc 4.1.x. But I am afraid that since
your kernel is either misconfigured or rather old that that isn't an
option and even if the is a binary package it might still cause
trouble.

Which Linux distribution is dreamhost offering?

Cheers,

Michael

> > 
>
> > Could you help me? Thanks.
>
> --
> William Stein
> Associate Professor of Mathematics
> University of Washingtonhttp://wstein.org
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Intel Mac Sage -testall errors

[mabshoff]

On Dec 24, 9:53 am, mabshoff <[EMAIL PROTECTED]
dortmund.de> wrote:
> On Dec 24, 8:55 am, tiredwarrior <[EMAIL PROTECTED]> wrote:
>
> > Hi - Thanks to all that replied
>
> Hi,
>
> > I untarred the file from the terminal using tar and the checksum is
> > OK. The error message on untarring unnerved me a bit, and so I think I
> > was too hasty in saying that testall failed. I have since let it run
> > right through and at the end it says "all tests passed!" .
>
> The issue is as Justin described above: lib vs. LIB symlinked on a
> case insensitive file system leads to trouble. We have been bitten by
> the issue before, so we need to resolve the issue somehow. One way in
> my opinion would be to used a dmg instead of a tar.gz as a medium for
> distrubution.
>
> > But it says
> > this despite that early on it shows the following 4 failure messages
> > below. Maybe some are because I do not have Latex installed?
>
> I don't think so, since the doctest extracts all relevant portions
> from the TeX document, it doesn't run latex at all.
>
>
>
> > Andrew
>
> > ---
> > Ran 40 tests in 6.435s
>
> > PASSED (successes=40)
> > Testing of examples currently not implemented.
> > Testing SAGE documentation
> > Testing SAGE tutorial
> > /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> > maketest: line 18: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> > Darwin/devel/doc/tut: No such file or directory
> > ERROR: File ./tut.tex is missing
> > exit code: 1
>
> > --
> > The following tests failed:
>
> > ./tut.tex
> > Total time for all tests: 0.0 seconds
> > Testing SAGE programming guide
> > /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> > maketest: line 22: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> > Darwin/devel/doc/prog: No such file or directory
> > ERROR: File ./prog.tex is missing
> > exit code: 1
>
> > --
> > The following tests failed:
>
> > ./prog.tex
> > Total time for all tests: 0.0 seconds
> > Testing SAGE constructions guide
> > /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> > maketest: line 26: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> > Darwin/devel/doc/const: No such file or directory
> > ERROR: File ./const.tex is missing
> > exit code: 1
>
> > --
> > The following tests failed:
>
> > ./const.tex
> > .
> > .
>
> All the above are due to the lib vs. LIB issue I assume.

I forgot to mention: The issue you see that the failures in *.tex
aren't properly reported in the end is ticket #879 - see
http://trac.sagemath.org/sage_trac/ticket/879

Cheers,

Michael

> > .
>
> > sage -t  devel/sage-main/sage/libs/pari/gen.pyx
> > python(4563) malloc: *** vm_allocate(size=409600) failed (error
> > code=3)
> > python(4563) malloc: *** error: can't allocate region
> > python(4563) malloc: *** set a breakpoint in szone_error to debug
>
> The issue above isn't a doctest failure, it is an error message
> printed by the system and it is exactly what one expects to see in
> that particular doctest since we try to allocate a very large list in
> pari that doesn't fit into physical memory (it it at leasr 100GB in
> site). So OSX tells you that it couldn't allocate a chunk of memory
> roughly 4 GB in  size which is expected since Sage is still a 32 bit
> application on OSX. I am adding the last one to the FAQ right now.
>
> Hope this help,
>
> Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Intel Mac Sage -testall errors

[mabshoff]

On Dec 24, 8:55 am, tiredwarrior <[EMAIL PROTECTED]> wrote:
> Hi - Thanks to all that replied

Hi,

> I untarred the file from the terminal using tar and the checksum is
> OK. The error message on untarring unnerved me a bit, and so I think I
> was too hasty in saying that testall failed. I have since let it run
> right through and at the end it says "all tests passed!" .

The issue is as Justin described above: lib vs. LIB symlinked on a
case insensitive file system leads to trouble. We have been bitten by
the issue before, so we need to resolve the issue somehow. One way in
my opinion would be to used a dmg instead of a tar.gz as a medium for
distrubution.

> But it says
> this despite that early on it shows the following 4 failure messages
> below. Maybe some are because I do not have Latex installed?

I don't think so, since the doctest extracts all relevant portions
from the TeX document, it doesn't run latex at all.

> Andrew
>
> ---
> Ran 40 tests in 6.435s
>
> PASSED (successes=40)
> Testing of examples currently not implemented.
> Testing SAGE documentation
> Testing SAGE tutorial
> /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> maketest: line 18: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> Darwin/devel/doc/tut: No such file or directory
> ERROR: File ./tut.tex is missing
> exit code: 1
>
> --
> The following tests failed:
>
> ./tut.tex
> Total time for all tests: 0.0 seconds
> Testing SAGE programming guide
> /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> maketest: line 22: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> Darwin/devel/doc/prog: No such file or directory
> ERROR: File ./prog.tex is missing
> exit code: 1
>
> --
> The following tests failed:
>
> ./prog.tex
> Total time for all tests: 0.0 seconds
> Testing SAGE constructions guide
> /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> maketest: line 26: cd: /Users/andrewgreen/sage-2.9-osx10.4-intel-i386-
> Darwin/devel/doc/const: No such file or directory
> ERROR: File ./const.tex is missing
> exit code: 1
>
> --
> The following tests failed:
>
> ./const.tex
> .
> .

All the above are due to the lib vs. LIB issue I assume.

> .
>
> sage -t  devel/sage-main/sage/libs/pari/gen.pyx
> python(4563) malloc: *** vm_allocate(size=409600) failed (error
> code=3)
> python(4563) malloc: *** error: can't allocate region
> python(4563) malloc: *** set a breakpoint in szone_error to debug

The issue above isn't a doctest failure, it is an error message
printed by the system and it is exactly what one expects to see in
that particular doctest since we try to allocate a very large list in
pari that doesn't fit into physical memory (it it at leasr 100GB in
site). So OSX tells you that it couldn't allocate a chunk of memory
roughly 4 GB in  size which is expected since Sage is still a 32 bit
application on OSX. I am adding the last one to the FAQ right now.

Hope this help,

Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---

[sage-support] Re: Intel Mac Sage -testall errors

[mabshoff]

On Dec 22, 6:08 pm, William Stein <[EMAIL PROTECTED]> wrote:
> sage -testall tests all optional packages, interfaces to commercial
> software, etc.  use make check from SAGE_ROOT instead.

I don't think so: from sage -advanced:

  -t [-optional] [-verbose] [-long]  -- test examples
in .py, .pyx, .sage or .tex files
   -optional -- include examples with
'optional' and 'package'
   -long -- include lines with the phrase
'long time'
   -verbose  -- print debuging output during
the test
  -tnew ... -- like -t above, but only tests files modified since
last commit
  -testall  -- test all source, docs, and examples in your SAGE
distro.

I use testall all the time on boxen without any of the optional or
commercial packages installed and it works fine.

Cheers,

Michael

>
> - William
>
> (Sent from my iPhone.)
>
> On Dec 22, 2007, at 6:43 AM, mabshoff <[EMAIL PROTECTED]
>
>  > wrote:
>
> > On Dec 22, 6:52 am, Andrew Green <[EMAIL PROTECTED]> wrote:
> >> Hi
>
> > Hi Andrew,
>
> >> I downloaded the sage-2.9-osx10.4-intel-i386-Darwin.tar.gz binary and
> >> untarred it. I got an error similar to another user in that after
> >> exectuting
> >> the tar command it returned after a few seconds...
>
> >> .
> >> .
> >> .
> >> sage-2.9-osx10.4-intel-i386-Darwin/spkg/standard/
> >> scipy_sandbox-20071020.spkg
> >> sage-2.9-osx10.4-intel-i386-Darwin/spkg/standard/README.txt
> >> sage-2.9-osx10.4-intel-i386-Darwin/spkg/gen_html
> >> sage-2.9-osx10.4-intel-i386-Darwin/spkg/install
> >> tar: Error exit delayed from previous errors
>
> >> But I can run Sage, however when I try to execute -testall I get
> >> during the
> >> run...
>
> > I guess you got "just enough" of Sage to make it start up, but the are
> > certainly bits and pieces missing. What are you using to untar the
> > install files? In our experience only the command line works
> > reliably.
>
> > Does the md5sum of the binary you downloaded match?
>
> > afa4a6e82a87dcd877b2fc795692b670  sage-2.9-osx10.4-intel-i386-
> > Darwin.tar.gz
> > d442c41dc69395f57624d5ca52038072  sage-2.9-osx10.5-intel-i386-
> > Darwin.tar.gz
>
> > I think we should really just use a DMG to distribute Sage for OSX.
> > That would certainly eliminate most of the tar issues we have been
> > seeing repeatedly.
>
> >> .
> >> .
> >> .
> >> Ran 40 tests in 5.846s
>
> >> PASSED (successes=40)
> >> Testing of examples currently not implemented.
> >> Testing SAGE documentation
> >> Testing SAGE tutorial
> >> /Users/andrewsage-2.9-osx10.4-intel-i386-Darwin/local/bin/sage-
> >> maketest:
> >> line 18: cd: /Users/andrew/sage-2.9-osx10.4-intel-i386-Darwin/devel/
> >> doc/tut:
> >> No such file or directory
> >> ERROR: File ./tut.tex is missing
> >> exit code: 1
> >> .
> >> .
>
> >> The many remaining .tex tests all fail subsequently with
> >> missing .tex files.
>
> >> Does this mean it has not installed correctly or that I am missing
> >> something
> >> important?
>
> >> I created a Notebook OK and it ran sage commands like integrate(),
> >> but I
> >> tried to implement a plot() command and it gave an error. Perhaps
> >> related to
> >> missing files?
>
> > Probably.
>
> >> If anyone can help it would be appreciated.
>
> >> Thanks and Merry Christmas
>
> >> Andrew
>
> > Cheers,
>
> > Michael
--~--~-~--~~~---~--~~
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URLs: http://sage.math.washington.edu/sage/ and http://sage.scipy.org/sage/
-~--~~~~--~~--~--~---