unpickle_all called twice = boom!

seblabbe commented 13 years ago

Calling twice in a row the function unpickle_all breaks sage:

sage: sage.structure.sage_object.unpickle_all()
x...
Successfully unpickled 586 objects.
Failed to unpickle 0 objects.
sage: sage.structure.sage_object.unpickle_all()
/Users/slabbe/Applications/sage-4.6.1/local/bin/sage-sage: line 300:  
7292 Abort trap              sage-ipython "$@" -i
$

Doing some print i, A inside the loop of the function unpickle_all shows that the process stop on the 170th :

sage: sage.structure.sage_object.unpickle_all()
...
163 _class__sage_combinat_sf_homogeneous_SymmetricFunctionAlgebra_homogeneous__.sobj
164 _class__sage_combinat_sf_jack_JackPolynomials_j__.sobj
165 _class__sage_combinat_sf_jack_JackPolynomials_p__.sobj
166 _class__sage_combinat_sf_jack_JackPolynomials_q__.sobj
167 _class__sage_combinat_sf_kschur_kSchurFunctions_t__.sobj
168 _class__sage_combinat_sf_llt_LLT_cospin__.sobj
169 _class__sage_combinat_sf_llt_LLT_spin__.sobj
170 _class__sage_combinat_sf_macdonald_MacdonaldPolynomials_h__.sobj
/Users/slabbe/Applications/sage-4.6.1/local/bin/sage-sage: line 300:  
6926 Abort trap              sage-ipython "$@" -i
 $

Component: pickling

Reviewer: Sébastien Labbé, Peter Bruin

Issue created by migration from https://trac.sagemath.org/ticket/10705

seblabbe commented 11 years ago

Description changed:

--- 
+++ 
@@ -29,7 +29,3 @@
  $

-If a solution needs to change the `pickle_jar`, be aware, because it may -results in conflicts because I am actually changing it at #10354.

-

AndrewMathas commented 11 years ago

comment:3

See #13636 for another seg fault associated with unpickle_all.

seblabbe commented 7 years ago

comment:9

This is not a problem anymore on 8.0.rc0.

pjbruin commented 7 years ago

Reviewer: Sébastien Labbé, Peter Bruin

pjbruin commented 7 years ago

comment:10

A doctest to check that this is fixed was added in #5838.

sagemath / sage