MyArray[3][-2] fails (quelle surprise), how to detect this?

[Fabian]

function getImage(id,x,y ) {
if (mpdef[0 + yy + y][0 + xx + x] == undefined) {
// set to default image
document.getEle mentById(id).sr c = terrdef[13];
}
else {
document.getEle mentById(id).sr c = terrdef[ mpdef[0 + yy + y][0 + xx
+ x] ];
}
return;
}

I use this function to grab an index number which refers to a graphics
file. In theory, firt it checks to see if the mpdef array has a valid
value as teh specified row/colum index, and then either puts in a
default index number or grabs the index number from teh array, depending
on whether the mpdef array has a value at that location.

However, it seems to choke when either teh referenced row or column
is -1 or lower. How should I catch this?
--
--
Fabian
Visit my website often and for long periods!
http://www.lajzar.co.uk

[HikksNotAtHome]

In article <8y**********@h otpop.com>, Lasse Reichstein Nielsen <lr*@hotpop.com >
writes:

You can write
var x = 42 + (y==2?10:37);
You can't write
42 + if (...)...
since "if" generates a statement, and statments aren't expressions.

if (y==2){adder = 10;}else{adder = 37;}
var x = 42 + adder;
--
Randy

[Dr John Stockton]

JRS: In article <bp************ *@ID-174912.news.uni-berlin.de>, seen in
news:comp.lang. javascript, Fabian <la****@hotmail .com> posted at Sat, 15
Nov 2003 12:37:52 :-

Dr John Stockton hu kiteb:

Why add 0 ?

Better to convert them to numbers when first generated. Operating
with numbers ought to be faster than operating with strings; or, at
least, no slower.

Note : the second of those sentences is a comment, not a reason. If the
meaning of a variable is fundamentally numeric, then it should be a
Number.

Aren't there some situations where a number can still get interpreted as
a string? I wasn't aware that there is a way to explicitly declare that
a variable is only ever to be considered as a number.

If the algorithm is correct, one will never want to numerically add a
Number to a String, since the second value will have been converted to a
Number at the first opportunity. If a number is to be converted to a
string and concatenated, it should be obvious what the intent is. These
ate not matters of javascript language design, but of writing code so
that it can be executed and read with the same interpretation.

X ? Y : Z means if X then Y else Z

So this is exactly equivalent to:

if (X) { Y }
else { Z }

No; but I think the converse is true. The first is a superset of the
second, since it can be used where a value is required; A=X?Y:X

--
© John Stockton, Surrey, UK. ?@merlyn.demon. co.uk Turnpike v4.00 IE 4 ©
<URL:http://jibbering.com/faq/> Jim Ley's FAQ for news:comp.lang. javascript
<URL:http://www.merlyn.demo n.co.uk/js-index.htm> JS maths, dates, sources.
<URL:http://www.merlyn.demo n.co.uk/> TP/BP/Delphi/JS/&c., FAQ topics, links.

[Lasse Reichstein Nielsen]

Dr John Stockton <sp**@merlyn.de mon.co.uk> writes:

JRS: In article <bp************ *@ID-174912.news.uni-berlin.de>, seen in

X ? Y : Z means if X then Y else Z

So this is exactly equivalent to: if (X) { Y }
else { Z }

No; but I think the converse is true. The first is a superset of the
second, since it can be used where a value is required; A=X?Y:X

It depends on what Y and Z are. If they can stand for arbitrary
Javascript syntax, then it isn't true. You can't put statements into
the ?:-operator. If Y and Z are just variables, then they are expressions,
and can be used in both places (expressions can be used as statements,
although using an expression with no side effect is pointless).

I.e., there are four different things that can vary:
- use ?: or if-then as conditional branch
E.g., x?...:... or if(x) ... else ...
- have expressions or statements in the branches
E.g., 2+2 (expression) or return 2+2; (statement)
- use the value of the conditional branch or not
E.g., q=(conditional branch); or (conditional branch);
- have side effects in the branches or not
E.g., w=2+2/foo() or 2+2

That give 16 different cases, many of them invalid.

It is invalid to use statments in the branches of the ?: operator.
It is invalid to use the value of an if-statement.
It is meaningless to not have side-effects in an if-statment

Without side effects in branches:
As a statement (not using value):
if (x) {2+2;} else {4+4;} : valid but pointless
if (x) {return 2+2;} else {return 4+4;} : valid and meaningful
x?(2+2):(4+4) : valid but pointless
x?(return 2+2;):(return 4+4;) : invalid
As an expression (value is used):
q = if (x) {2+2;} else {4+4;} : invalid
q = if (x) {return 2+2;} else {return 4+4;} : invalid
q = x?(2+2):(4+4) : valid and meaningful
q = x?(return 2+2;):(return 4+4;) : invalid
With side effects:
As a statement:
if (x) {w=2+2;} else {foo();} : valid and meaningful
if (x) {return w=2+2;} else {return foo();} : valid and meaningful
x?(w=2+2):(foo( )) : valid but you should use if
x?(return w=2+2;):(return foo();) : invalid
As an expression (value is used):
q = if (x) {w=2+2;} else {foo();} : invalid
q = if (x) {return w=2+2;} else {return foo();} : invalid
q = x?(w=2+2):(foo( )) : valid and meaningful
q = x?(return w=2+2;):(return foo();) : invalid

Hope this makes any sense :)

/L
--
Lasse Reichstein Nielsen - lr*@hotpop.com
DHTML Death Colors: <URL:http://www.infimum.dk/HTML/rasterTriangleD OM.html>
'Faith without judgement merely degrades the spirit divine.'

[Thomas 'PointedEars' Lahn]

Lasse Reichstein Nielsen wrote:

Dr John Stockton <sp**@merlyn.de mon.co.uk> writes:

JRS: In article <bp************ *@ID-174912.news.uni-berlin.de>, seen in

X ? Y : Z means if X then Y else Z

So this is exactly equivalent to:

if (X) { Y }
else { Z }
No; but I think the converse is true. The first is a superset of the
second, since it can be used where a value is required; A=X?Y:X

It depends on what Y and Z are. If they can stand for arbitrary
Javascript syntax, then it isn't true.

Right.
You can't put statements into the ?:-operator.
Wrong because it is too general. Try this:

var x = false;
y = x ? alert(x) : alert(42);

A method call is a statement and an assignment
(see below) is a statement, too.
As a statement (not using value):
if (x) {2+2;} else {4+4;} : valid but pointless
if (x) {return 2+2;} else {return 4+4;} : valid and meaningful
x?(2+2):(4+4) : valid but pointless
x?(return 2+2;):(return 4+4;) : invalid
return (x ? 2 + 2 : 4 + 4);
As an expression (value is used):
q = if (x) {2+2;} else {4+4;} : invalid
q = if (x) {return 2+2;} else {return 4+4;} : invalid
q = x?(2+2):(4+4) : valid and meaningful
q = x?(return 2+2;):(return 4+4;) : invalid
return (q = x ? 2 + 2 : 4 + 4);
With side effects:
As a statement:
if (x) {w=2+2;} else {foo();} : valid and meaningful
if (x) {return w=2+2;} else {return foo();} : valid and meaningful
x?(w=2+2):(foo( )) : valid but you should use if
x?(return w=2+2;):(return foo();) : invalid

return (x ? w = 2 + 2 : foo());
PointedEars

[Lasse Reichstein Nielsen]

Thomas 'PointedEars' Lahn <Po*********@we b.de> writes:

You can't put statements into the ?:-operator.

Wrong because it is too general. Try this:

var x = false;
y = x ? alert(x) : alert(42);

A method call is a statement and an assignment
(see below) is a statement, too.

A method call and an assignment are expressions. So are additions and
a single identifier. They are not *by themselves* statements.

*Any* expression [1] *followed by a semicolon* is a statement (more
precisely: an "ExpressionStat ement" in the grammar of ECMAScript -
ECMA 262, section 12.4).

[1] except function expressions and literal objects.

So:
x = 4
is an expression, not a statement (although automatic semicolon insertion
can turn it into one, if it is used in a position where the grammar requires
a statement).
x = 4;
is a statement, not an expression.

x?(return 2+2;):(return 4+4;) : invalid

return (x ? 2 + 2 : 4 + 4);

Yes, but that is not the same. If you wrote the original, invalid,
expression, then this is probably what you meant. But it is just an
example of an invalid construction, not something you would want
to write.

/L
--
Lasse Reichstein Nielsen - lr*@hotpop.com
DHTML Death Colors: <URL:http://www.infimum.dk/HTML/rasterTriangleD OM.html>
'Faith without judgement merely degrades the spirit divine.'