collect problem

[Ben Sapp]

"Richard B. Kreckel" wrote:
> 
> Hi,
> 
> On Fri, 13 Oct 2000, Ben Sapp wrote:
> We could call it: "GiNaC ran out of spec here".  Don't know at the moment
> how to fix it.  Please read on...
> 

Thanks, this is nice to know.  

> > I have attached the archive file and the source to the program that
> > reads the archive file.   I did not give the source to the software that
> > originally produced it because it is a function that is dynamically
> > linked with another program.(GNU Octave)
> 
> The archive file is somewhat strange.  I get:
> 
> higgsino:~/projects$ viewgar foobar.gar
> The symbol = x
> foobar.gar: class 'name' not registered
> 
> Is 'name' one of your classes?  What's going on here?  Can you just print
> the expression and mail it as it seems to be quite small?
> 

Hmmm.   This is not what I see when I run viewgar.  I should mention
that I 
experience similar problems with different expressions.  Most are pretty 
simple expressions. Here is what I see with viewgar and the file I sent
earlier:

[ben at pcben Linux]$ viewgar foobar.gar
The symbol = x
The expression = (-5.0+x)*(3.0+x^(2.0))
[ben at pcben Linux]$ viewgar -d foobar.gar
Atoms:
 0 class
 1 symbol
 2 name
 3 x
 4 The symbol
 5 mul
 6 add
 7 rest
 8 numeric
 9 number
 10 1
 11 coeff
 12 R-1 11529215046068469760 -61
 13 overall_coeff
 14 power
 15 basis
 16 R1 9223372036854775808 -62
 17 exponent
 18 R1 13835058055282163712 -62
 19 The expression

Expressions:
 0 "The symbol" root node 0
 1 "The expression" root node 10

Nodes:
 0
  string "class" 1
  string "name" 3
 1
  string "class" 1
  string "name" 3
 2
  string "class" 8
  string "number" 10
 3
  string "class" 8
  string "number" 12
 4
  string "class" 6
  node "rest" 1
  node "coeff" 2
  node "overall_coeff" 3
 5
  string "class" 1
  string "name" 3
 6
  string "class" 8
  string "number" 16
 7
  string "class" 14
  node "basis" 5
  node "exponent" 6
 8
  string "class" 8
  string "number" 18
 9
  string "class" 6
  node "rest" 7
  node "coeff" 2
  node "overall_coeff" 8
 10
  string "class" 5
  node "rest" 4
  node "coeff" 2
  node "rest" 9
  node "coeff" 2
  node "overall_coeff" 2

[ben at pcben Linux]$   

As far as 'name' being one of my classes it is certainly possible though
I can't 
confirm for certain.  I create classes that are subclasses of
octave_value.  These
classes then have a GiNaC type as the only member variable.   So it
could be a 
remnant of the super class.  I will look into it more if you think this
could be 
the source of the problem. 

Thanks.
-
To UNSUBSCRIBE, email to ginac-list at ginac.de with a subject of "unsubscribe".