[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
(1) keyboard/error (2) access variables (3) keywords
From: |
S.Hefti |
Subject: |
(1) keyboard/error (2) access variables (3) keywords |
Date: |
Fri, 16 Apr 1999 18:13:06 -0400 |
Hey,
I have accumulated three questions, so here they are in one
e-mail:
1) I like the routine keyboard() for debugging. I think it
would be useful if I could cause error() to go into
the keyboard() mode, rather than just to go back to
the top level and to print the traceback.
Is this possible in some way ?
2) Is there a way to access variables from a different function
even though they are not defined as global ? Obviously, this
is somewhat against the definition of global. What I would like
is something like perl's syntax to access data:
main::x would give you the variable x from the main control
level. caller::x would give you variable x from the calling
function. Since octave maintains all these variables anyway,
this should not be too hard to implement, I think.
Let me know what you think.
3) I like the keyword feature IDL has. For those unfamiliar with
IDL: You can pass options to a function with a syntax like:
res = do_somethig(x,y,weight='exponential',lamda=llamda);
where llamda is a local variable. The nice thing about this
syntax is that you can also call the routine like
res = do_somethig(x,y,lamda=llamda,weight='exponential');
or
res = do_somethig(x,y,lamda=3.1);
or
res = do_somethig(x,y,simple=1);
if "simple" is an other keyword supported by do_something();
I think that this approach has a lot of advantages over the
if( nargin > 5 ) ...
approach. Note that if you would like to turn on option 10
of the nargin approach, you have to choose and type in options
1..9 as well. Using keywords, you just specify which option
to set. I also have a lot less trouble recalling keywords
than I have to recall the number of the option I would like
to use.
Is there a way to use keywords with octave ? Would it be
possible to implement it ?
Again, I am interested in your opinions.
Thanks a lot,
Simon.
--
---------------------------------------------------------------------
Simon Hefti, Ph D address@hidden
---------------------------------------------------------------------
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- (1) keyboard/error (2) access variables (3) keywords,
S.Hefti <=