[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Help-gsl] Vector and matrix views

From: Brian Gough
Subject: Re: [Help-gsl] Vector and matrix views
Date: Thu, 8 Dec 2005 18:07:06 +0000

Dimitris Papavasiliou writes:
 > Would something like this work
 > +====
 > gsl_vector v;  /* a struct, not a pointer */
 > v = gsl_vector_view_from_array(array, n).vector;
 > +====
 > I know this looks ugly too but hidden behind a macro it would be ok.  
 > Would this work correctly?  

This is ok in principle.  What you don't want to do is

  gsl_vector *v;
  v = &gsl_vector_view_from_array(array, n).vector;

which is easily confused with it and causes invalid memory accesses
(since the contents of the vector struct disappears from the stack
immediately after the expression is evaluated).

I use the style "&view.vector" to avoid this problem.  Every vector is
always a pointer and every view is always on the stack, so everything
stays consistent.  A useful macro is &(x).vector.

As James mentions you can construct a vector directly (or define a
small inline function to do it).  This works for non-const vectors.  I
do use it myself, it is alright.  

 > On a side note: why do views have a separate struct with only one 
 > member?  I've read in an old message in the archive that this is so in 
 > order to make const vectors work but I can't imagine how keeping a 
 > struct with one member is any different than keeping the member itself.  
 > If it is't too much trouble could someone please explain?

In C a const object has to be initialised in a single step (because
you can't modify it afterwards).  And initialisers can only contain
constants, not expressions.  So to create a const object whose members
depend on other variables you have to go through an indirection --
create a non-const object first, then make it const.  That is where
the struct comes from.  It's const on the outside but non-const on the

It would be cool to discover other ways to do it but I think that this
might be the only way in C.

Brian Gough

Network Theory Ltd,
Publishing the GSL Manual ---

reply via email to

[Prev in Thread] Current Thread [Next in Thread]