For larger programs, using a true editor like emacs, vi, nedit, ... in combination with the read command is recommended.
The mode buttons bar will help you enter runstream control instructions
(instruction group {}
, end instruction ;
, newline,
if
test, for
loop, return
end of function).
The ->cmd
button leaves the programming editor and
copies the program in the commandline, type Enter to put
it in the history.
If your program begins by an affectation instruction (example
myprog(a,b) := a+b; ), xcas will research in the
history a similar affectation and replace it with the current
one and will recompute the following history levels.
The exec
button can be used to run in history the
current line (like for scripts). The parse
button
can be used to verify the syntax correctness of the program.
In the Xcas) programming style, the syntax is similar to C. The main differences are :
f(x,y):={ local z; instructions; }
.
Be warned that the respective priorities of := and ,
will require using parenthesis if you affect local variables
during declaration :
f(x,y):={ local (z:=0),v,w; ... }
.
If you choosed Maple, Mupad or TI89/92 syntax in the red cas configuration
screen, please refer to these software programming syntax. Note
that this choice will affect all the software, for example
= I in Maple/Mupad mode and
= i in xcas mode.
If your program does not behave like you think, you can execute
it step by step (under Linux and Mac OS X) using the command
debug(function_name(arguments))
Note that this functionnality is not available under Windows.
You can see variables evolution, set breakpoints, ...
For example load the bezout
file of the directory
examples/arit and run debug(pgcd(15,25))
.