relocation not working for sage-4.7.1.alpha1

[mariah]

eno% cd sage-4.7.1.alpha1-x86_64-Linux-core2-fc
eno% ./sage
----------------------------------------------------------------------
| Sage Version 4.7.1.alpha1, Release Date: 2011-05-30                |
| Type notebook() for the GUI, and license() for information.        |
----------------------------------------------------------------------
**********************************************************************
*                                                                    *
* Warning: this is a prerelease version, and it may be unstable.     *
*                                                                    *
**********************************************************************
sage: L.<w> = Qp(7).extension(x^2 - 7)
sage: quit;
Exiting Sage (CPU time 0m1.18s, Wall time 0m20.03s).
eno% cd ..
eno% mv sage-4.7.1.alpha1-x86_64-Linux-core2-fc sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved
eno% cd sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved
eno% ./sage
----------------------------------------------------------------------
| Sage Version 4.7.1.alpha1, Release Date: 2011-05-30                |
| Type notebook() for the GUI, and license() for information.        |
----------------------------------------------------------------------
**********************************************************************
*                                                                    *
* Warning: this is a prerelease version, and it may be unstable.     *
*                                                                    *
**********************************************************************
The Sage install tree may have moved
(from /home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc to /home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved)
Changing various hardcoded paths
(please wait at most a few minutes)...
Do not interrupt this.
Done resetting paths
sage: L.<w> = Qp(7).extension(x^2 - 7)
ERROR: An unexpected error occurred while tokenizing input
The following traceback may be corrupted or invalid
The error message is: ('EOF in multi-line statement', (1178, 0))

---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/<ipython console> in <module>()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/rings/padics/padic_generic.pyc in extension(self, modulus, prec, names, print_mode, halt, **kwds)
    463                     else:
    464                         print_mode[option] = self._printer.dict()[option]
--> 465         return ExtensionFactory(base=self, premodulus=modulus, prec=prec, halt=halt, names=names, check = True, **print_mode)
    466 
    467 def local_print_mode(obj, print_options, pos = None, ram_name = None):

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/structure/factory.so in sage.structure.factory.UniqueFactory.__call__ (sage/structure/factory.c:886)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/rings/padics/factory.pyc in create_key_and_extra_args(self, base, premodulus, prec, print_mode, halt, names, var_name, res_name, unram_name, ram_name, print_pos, print_sep, print_alphabet, print_max_ram_terms, print_max_unram_terms, print_max_terse_terms, check, unram)
   2206             if is_Expression(premodulus):
   2207                 # Here we assume that the output of coeffs is sorted in increasing order by exponent:
-> 2208                 coeffs = premodulus.coeffs()
   2209                 preseed = premodulus / coeffs[-1][0]
   2210                 preseed -= preseed.variables()[0]**coeffs[-1][1]

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/symbolic/expression.so in sage.symbolic.expression.Expression.coefficients (sage/symbolic/expression.cpp:18648)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/symbolic/expression.so in sage.symbolic.expression.Expression._maxima_ (sage/symbolic/expression.cpp:3674)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/structure/sage_object.so in sage.structure.sage_object.SageObject._interface_ (sage/structure/sage_object.c:3451)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/misc/lazy_import.so in sage.misc.lazy_import.LazyImport.__getattr__ (sage/misc/lazy_import.c:1353)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/misc/lazy_import.so in sage.misc.lazy_import.LazyImport._get_object (sage/misc/lazy_import.c:1023)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/interfaces/maxima_lib.py in <module>()
    150 init_code.append('nolabels : true')
    151 for l in init_code:
--> 152     ecl_eval("#$%s$"%l)
    153 ## To get more debug information uncomment the next line
    154 ## should allow to do this through a method

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/libs/ecl.so in sage.libs.ecl.ecl_eval (sage/libs/ecl.c:6301)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/libs/ecl.so in sage.libs.ecl.ecl_eval (sage/libs/ecl.c:6252)()

/home/mariah/sage/sage-4.7.1.alpha1-x86_64-Linux-core2-fc-moved/local/lib/python2.6/site-packages/sage/libs/ecl.so in sage.libs.ecl.ecl_safe_eval (sage/libs/ecl.c:2544)()

RuntimeError: ECL says: THROW: The catch MACSYMA-QUIT is undefined.
sage: 

Apply to sage_scripts:

• trac_11419_set_MAXIMA_PREFIX-p2.patch​

[mariah]

The problem occurs in sage-4.7.1.alpha0. So some change between sage-4.7 and sage-4.7.alpha0 is causing this problem.

[fbissey]

Does rebuilding ecl and possibly maxima helps?

[nbruin]

I think this is due to the instruction load(to_poly_solver) that is in init_code in sage/interfaces/maxima_lib.py. The problem arises earlier in the initialization of maxima. The following is an excerpt of the instructions sent to ecl upon maxima_lib initialization, pasted into the ecl REPL:

sage/4.7.1-moved$ ./sage -ecl
ECL (Embeddable Common-Lisp) 11.1.1
Copyright (C) 1984 Taiichi Yuasa and Masami Hagiya
Copyright (C) 1993 Giuseppe Attardi
Copyright (C) 2000 Juan J. Garcia-Ripoll
ECL is free software, and you are welcome to redistribute it
under certain conditions; see file 'Copyright' for details.
Type :h for Help.  
Top level.
> (require 'maxima)

;;; Loading #P"/usr/local/sage/4.7.1-moved/local/lib/ecl/maxima.fas"
;;; Loading #P"/usr/local/sage/4.7.1-moved/local/lib/ecl/sb-bsd-sockets.fas"
;;; Loading #P"/usr/local/sage/4.7.1-moved/local/lib/ecl/sockets.fas"
("SB-BSD-SOCKETS" "SOCKETS" "MAXIMA")
> (in-package :maxima)

#<"MAXIMA" package>
MAXIMA> (set-locale-subdir)

NIL
MAXIMA> (set-pathnames)     

"/usr/local/sage/4.7.1/local/share/info/./maxima-index.lisp"
MAXIMA> 

That's a hardcoded path! Luckily there is an override. If you do a

export MAXIMA_PREFIX = $SAGE_ROOT/local

before starting sage, the problem goes away. So this should go somewhere in the start-up of sage.

[fbissey]

That would belong to sage-env then, it should be easy to add.

[nbruin]

to be applied in local/bin to let sage-env set MAXIMA_PREFIX

[fbissey]

I will check the patch a bit later but I think it will apply cleanly to sage_scripts in this format.

[fbissey]

set maxima prefix in sage-env

[fbissey]

The original patch didn't initially import cleanly in sage_scripts. So I regenerated it in, the new version is in git format as well.

Mariah can you test it?

[jhpalmieri]

This seems to fix the problem as far as I can tell, but would it be better to check whether MAXIMA_PREFIX is set first? Something like this:

if [ -z "$MAXIMA_PREFIX" ]; then
    MAXIMA_PREFIX="$SAGE_ROOT/local" && export MAXIMA_PREFIX
fi

Is it possible that people will be (successfully) using their own installation of maxima?

We could also use "$SAGE_LOCAL" instead of "$SAGE_ROOT/local"; I'm not sure whether stylistically we should have any preference.

[nbruin]

Replying to jhpalmieri:

Is it possible that people will be (successfully) using their own installation of maxima?

I considered that, but with the maxima.fas on libecl.so setup loaded inside sage's main process, people do not have the choice which installation of sage they run. Allowing sage's maxima work with a maxima library from a different install is risky at best (I don't know what exactly maxima gets from there at runtime) That's why I unconditionally set the variable.

If people want to run their own maxima via a pexpect interface, they can do so by setting the environment variable in a script wrapping the executable.

[fbissey]

I was going to make a detailed answer with the same kind of things. Unless you are running Gentoo or possibly Mandriva you won't have the maxima.fas bits.

And for the pexpect interface, you better have a maxima built with ecl and know how to make sure you get the ecl maxima if you have maxima installed with several lisps.

As for SAGE_LOCAL on principle I would use it instead of $SAGE_ROOT/local but admittely I don't care so much in sage-env. In module_list.py it drives me insane on the other hand.

[mariah]

I applied the patch to sage-4.7.1.alpha2 using hg_scripts.apply(), did 'sage -b', then 'make testlong'. All tests passed. I then moved the sage directory and again did 'make testlong'. Again all tests passed. I also check that the patch fixed the reported problem.

Wanting to check this from a source build, I hand-edited 'sage-env' in spgk/base and in spkg/standard/sage_scripts-4.7.1.alpha2.spkg, then did 'make testlong' which built sage and ran the long tests. All tests passed. I then moved the sage directory and again did 'make testlong'. Again all tests passed.

Positive review!

(If a kind person would explain to me how to apply such script patches so that I can get just a modified source tree - other than doing it by hand - I would appreciate it.)

[mariah]

Replying to mariah:

(If a kind person would explain to me how to apply such script patches so that I can get just a modified source tree - other than doing it by hand - I would appreciate it.)

Answering my own question: 'sage -sdist'