On Dec 21, 2007 11:27 AM, Robert Miller <[EMAIL PROTECTED]> wrote:
>
> > > What needs to happen to go from 2.9.1.alpha2 to a release
> > > candidate and a release?
>
> Also, we *really* need to update COPYING.txt to reflect the new
> packages.
Yes, for ATLAS and R.
Anyway, I built alpha2 on a bunch of machines and here are the results:
---------------------------------------------------------------------------------
* My LAPTOP-- a macbook pro core 2 duo running osx 10.5.1:
The build completes fine, but about 50 doctest files fail with:
rank4:sage-2.9.1.alpha2 was$ ./sage -t devel/sage-main/sage/gsl/dwt.pyx
sage -t devel/sage-main/sage/gsl/dwt.pyx
------------------------------------------------------------
Unhandled SIGBUS: A bus error occured in SAGE.
This probably occured because a *compiled* component
of SAGE has a bug in it (typically accessing invalid memory)
or is not properly wrapped with _sig_on, _sig_off.
You might want to run SAGE under gdb with 'sage -gdb' to debug this.
SAGE will now terminate (sorry).
------------------------------------------------------------
This failure occurs when I try to do *any* plotting at all (that is
what causes all the failures above):
sage: show(plot(sin))
------------------------------------------------------------
Unhandled SIGBUS: A bus error occured in SAGE.
This probably occured because a *compiled* component
of SAGE has a bug in it (typically accessing invalid memory)
or is not properly wrapped with _sig_on, _sig_off.
You might want to run SAGE under gdb with 'sage -gdb' to debug this.
SAGE will now terminate (sorry).
------------------------------------------------------------
Under gdb we get, so the problem must have to do with freetype.
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x00000020
0x06980709 in std::codecvt<char, char, __mbstate_t>::do_unshift
(this=0x6a1b048, __to=0x0, [EMAIL PROTECTED]) at
../../../../gcc-4.2-20061007/libstdc++-v3/src/codecvt.cc:77
77 ../../../../gcc-4.2-20061007/libstdc++-v3/src/codecvt.cc: No such
file or directory.
in ../../../../gcc-4.2-20061007/libstdc++-v3/src/codecvt.cc
Current language: auto; currently c++
(gdb) bt
#0 0x06980709 in std::codecvt<char, char, __mbstate_t>::do_unshift
(this=0x6a1b048, __to=0x0, [EMAIL PROTECTED]) at
../../../../gcc-4.2-20061007/libstdc++-v3/src/codecvt.cc:77
#1 0x069bec61 in std::ostream::_M_insert<long> (this=0xbfff70b4,
__v=81) at
/Users/admin/build3/i386-apple-darwin8.8.1/libstdc++-v3/include/bits/locale_facets.h:2341
#2 0x0725ef2b in FT2Font::FT2Font ()
#3 0x07262d73 in ft2font_module::new_ft2font ()
#4 0x0726c88b in Py::ExtensionModule<ft2font_module>::invoke_method_varargs ()
#5 0x07273b9f in method_varargs_call_handler ()
#6 0x0009c970 in PyEval_EvalFrameEx ()
#7 0x0009e51a in PyEval_EvalCodeEx ()
I just tried my new matplotlib-0.91.1.spkg to see if this fixes the
problem -- but there are
a number of changes I have to make to plot/* to get sage to work with
this new version
of matplotlib. I'm changing
http://sagetrac.org/sage_trac/ticket/924
to a blocker and trying to make all necessary fixes to Sage's plotting
code right now.
---------------------------------------------------------------------------------
2. FERMAT -- OSX 10.4 PowerPC
Two files have failures. The first is:
sage -t devel/sage-main/sage/calculus/calculus.py
**********************************************************************
File "calculus.py", line 2005:
sage: f = f.nintegral(x,0,1,1e-14)
Expected:
Traceback (most recent call last):
...
ValueError: Maxima (via quadpack) cannot compute the integral to
that precision
Got:
Traceback (most recent call last):
File "/Users/was/build/sage-2.9.1.alpha2/local/lib/python2.5/doctest.py",
line 1212, in __run
compileflags, 1) in test.globs
File "<doctest __main__.example_40[1]>", line 1, in <module>
f = f.nintegral(x,Integer(0),Integer(1),RealNumber('1e-14'))###line
2005:
sage: f = f.nintegral(x,0,1,1e-14)
File
"/Users/was/build/sage-2.9.1.alpha2/local/lib/python2.5/site-packages/sage/calculus/calculus.py",
line 2068, in nintegral
raise TypeError, err
TypeError: Error executing code in Maxima
CODE:
sage356 : quad_qags(sage6,sage6,sage352,sage353,sage354,sage355)$
Maxima ERROR:
^M
Maxima encountered a Lisp error:^M
^M
^M
WRITE-CHAR on #<CLOSED OUTPUT BUFFERED FILE-STREAM CHARACTER
#P"/dev/fd/1"> is illegal^M
^M
Automatically continuing.^M
To reenable the Lisp debugger set *debugger-hook* to nil.^M
<BLANKLINE>
**********************************************************************
1 items had failures:
1 of 14 in __main__.example_40
***Test Failed*** 1 failures.
For whitespace errors, see the file .doctest_calculus.py
[61.0 s]
----- This is related to Mike Hansen's precision detection patch -- it
doesn't work right on
OSX PPC, evidently.
The other failure, mentioned in the thread above is:
sage -t devel/sage-main/sage/graphs/graph.py
**********************************************************************
File "graph.py", line 4150:
sage: E[1][0]
Expected:
Vector space of degree 5 and dimension 1 over Real Double Field
User basis matrix:
[ 0.632455532034 -0.632455532034 -0.4472135955 -0.013900198608
0.0738411279702]
Got:
Vector space of degree 5 and dimension 1 over Real Double Field
User basis matrix:
[ 0.632455532034 -0.632455532034 -0.4472135955 0.223603825523
0.0439447748974]
**********************************************************************
----------------------------------------------------------------------------------
3. 32-bit Intel Debian and RHEL5:
Only the graph.py test fails. Everything else works.
----------------------------------------------------------------------------------
4. 32-bit Intel Mac Pro OSX 10.5.1 on Xeon:
Exactly one failure -- the calculus.py failure mentioned above on OSX10.4 PPC.
----------------------------------------------------------------------------------
5. Intel OSX 10.4:
Exactly one failure -- the calculus.py failure mentioned above on OSX10.4 PPC.
-- William
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---
