OpenEuphoria

Re: Ways to cause an error

• Posted by Pete Lomax <petelomax at blueyonder.co.uk> Aug 14, 2004
• 481 views

On Fri, 13 Aug 2004 15:48:17 -0400, "icxcnika at hotpop.com"
<icxcnika at hotpop.com> wrote:

>>Hey gang! I've been looking for ways to cause errors upon the execution of
>>a statement. Here are the ones I've found so far:
>>
>>The famous division by zero, i.e.:
>>? 1/0
>>The impossible peek:
>>x=peek4u(#FFFFFFFF)
>>Pass an invalid arguement to a routine.
>>
>>Any others?

Try these (they are all runtime errors, and compile cleanly)

type h(integer c) return not c end type
h halt
object DIE
sequence s
	s="name"
integer i	-- or atom
	i=1
sequence STOP
procedure x()
--	abort("string")	-- argument to abort must be an atom
--	trace("string")	-- argument to trace must be an atom
--	halt=0	-- OK
--	halt=1	-- terminates program ("type check failure, halt is 1")
--	DIE+=1	-- variable DIE has not been assigned a value
--	if s="" then end if	-- sequence lengths are not the same
--	if s="fred" then end if -- true/false condition must be an ATOM
--	if s then end if		-- ""
--	if s=1 then end if		-- ""
--	if 1="" then end if		-- ""
--	if ""="" then end if	-- ""
--	s[0]=1			-- subscript value 0 is out of bounds
--	abort(s[0])		-- ""
--	i[1]=1			-- attempt to subscript an atom
--	abort(i[1])		-- ""
--	s[{1,2}]=1		-- subscript must be an atom
--	abort(s[{1,2}])	-- ""
--	s[3..1]=1		-- slice length is less than 0 (-1)
--	abort(s[3..1])	-- ""
--	s[1..1]="12"	-- lengths do not match on assignment to slice
--	i&=1			-- type check failure, i is {1,1}
--	STOP=1			-- type check failure, STOP is 1
--	i=length(1)		-- length of an atom is not defined
--	if length(0) then end if -- ""
--	abort(length(0))-- ""
--	if getc(-1) then end if	-- bad file number (-1)
--	i=rand(0)		-- argument to rand must be >=1
end procedure
x()

Pete