[Bugzilla] [Bug 13838] Samuel GOUGEON has commented the bug

[bugzilla.admin at scilab.org]

http://bugzilla.scilab.org/show_bug.cgi?id=13838
--------- Bug Summary -----------
a = sparse(rand(5,2)); a(8) = a(8) - %i   => crashes Scilab. a(8) = a(8) - 1   runs OK

---------- Changed by ------------
sgougeon at free.fr

---------- What changed ----------

--- Comment #2 from Samuel GOUGEON <sgougeon at free.fr> ---
This line of instructions does not crash Scilab 4.1.2 but gives systematically a strange result like this one:

-->a = rand(5,2), a = sparse(a), a(8) = a(8) - %i
 a  =

    0.5015342    0.9184708
    0.4368588    0.0437334
    0.2693125    0.4818509
    0.6325745    0.2639556
    0.4051954    0.4148104
 a  =

(    5,    2) sparse matrix

(    1,    1)        0.5015342
(    1,    2)        0.9184708
(    2,    1)        0.4368588
(    2,    2)        0.0437334
(    3,    1)        0.2693125
(    3,    2)        0.4818509
(    4,    1)        0.6325745
(    4,    2)        0.2639556
(    5,    1)        0.4051954
(    5,    2)        0.4148104
 a  =

(    5,    2) sparse matrix

(    1,    1)       0.5015342
(    1,    2)       0.9184708
(    2,    1)       0.4368588
(    2,    2)       0.0437334
(    3,    1)       0.2693125
(    4,    1)       0.6325745
(    4,*****)       2.122-314 + 2.122-314i
(    5,    1)       0.4051954
(    5,    2)       0.4818509
(    6,    1)       0.0437334

Here :
1) The (3,2) element disapears
2) The (4,2) element (index "******" and related values) is corrupted.
3) an oversized i-index == 6 appears (last line)

-- 
You are receiving this mail because:
You are watching all bug changes.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/bugzilla/attachments/20150424/c8007e28/attachment.htm>