[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: C++ Question
From: |
Paul Kienzle |
Subject: |
Re: C++ Question |
Date: |
Mon, 31 Mar 2003 20:39:06 -0500 |
User-agent: |
Mozilla/5.0 (Windows; U; Win 9x 4.90; en-US; rv:1.3a) Gecko/20021212 |
Quentin H. Spencer wrote:
Hello all,
I have been writing some functions in C that use the octave ComplexSVD
class, and I'm having trouble understanding some strange behavior.
Initially I wrote the function with the following syntax:
ComplexSVD svd;
svd = ComplexSVD( matrix , SVD::std);
ComplexMatrix U = svd.left_singular_matrix();
This is a copy constructor
Later I changed the syntax to the following (which mirrors the syntax
used in svd.cc).
ComplexSVD svd( matrix , SVD::std);
This is a constructor.
ComplexMatrix U = svd.left_singular_matrix();
When using the first syntax, the function would usually return
properly the first time, but subsequent runnings of my C++ function
(and sometimes even octave's builtin svd function) often resulted in
the following:
error: ComplexSVD: U not computed because type == SVD::sigma_only
This is usally followed by a segmentation fault. There could be some
technicality about C++ that I don't understand here, but I don't see
the different between these two implementations. Can anyone enlighten me?
The copy constructor code in liboctave/CmplxSVD.h, which is the function
ComplexSVD& operator= (const ComplexSVD& a)
does not copy the type_computed variable, so it keeps the value it had
before the copy. Since your svd is uninitalized it is assumes its default
value (which may be random --- I don't know what C++ does with
uninitialized enums).
Copy it, test it and send a patch to bug-octave.
Thanks,
Paul Kienzle
address@hidden
-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.
Octave's home on the web: http://www.octave.org
How to fund new projects: http://www.octave.org/funding.html
Subscription information: http://www.octave.org/archive.html
-------------------------------------------------------------
- C++ Question, Quentin H. Spencer, 2003/03/31
- Re: C++ Question,
Paul Kienzle <=