Alternative to eval

[HikksNotAtHome]

This is a very simplified example of an Intranet application. But, is there an
easier (or more efficient) way of getting the decimal equivalent of a fraction?

The actual function gets the select values, this one is a simplified version
where its passed.

function checkIt(selVal){
valueInDec1 = eval(selVal);
//do some calculations here with valueInDec1
}

<select onchange="checkIt(this.value)">
<option value="0">0</option>
<option value="1/16">1/16</option>
<option value="1/8">1/8</option>
...............
<option value="7/8">7/8</option>
<option value="15/16">15/16</option>
</select>

I know that I could change it and have it look up the decimal equivalent for
the fraction, with an Object or an Array. I could also split the value on the /
and convert them to numbers, do the division, and get the decimal.

I discounted the Object/Array approach (and may go back to it, but doubt it).
The reason is that its not just 16ths. Some of the selects are 37ths, some are
7ths, one of them has 128ths in it. Simply put, I don't think that maintaining
10+ objects to contain the decimals is efficient (but open to that approach).
Also, its subject to get some added (different denominators), and some removed.

The split approach just doesn't seem to be as efficient though.

function checkIt(selVal){
myNumber = selVal.split('/')
decimalForm = parseFloat(myNumber[0])/parseFloat(myNumber[1])
//do some calculations here with valueInDec1
}

When the form gets submitted, it has to be in fractional form. I have no
control over the way the selects get generated so I am kind of stuck trying to
do it client side.

Thoughts, suggestions, comments?
--
Randy
All code posted is dependent upon the viewing browser
supporting the methods called, and Javascript being enabled.

[Lee]

HikksNotAtHome said:

This is a very simplified example of an Intranet application. But, is there an
easier (or more efficient) way of getting the decimal equivalent of a fraction?

The actual function gets the select values, this one is a simplified version
where its passed.

function checkIt(selVal){
valueInDec1 = eval(selVal);
//do some calculations here with valueInDec1
}

<select onchange="checkIt(this.value)">
<option value="0">0</option>
<option value="1/16">1/16</option>
<option value="1/8">1/8</option>
..............
<option value="7/8">7/8</option>
<option value="15/16">15/16</option>
</select>

The most efficient solution would seem to be:

<select onchange="checkIt(this.value)">
<option value="0">0</option>
<option value="0.0625">1/16</option>
<option value="0.125">1/8</option>
...............
<option value="0.875">7/8</option>
<option value="0.9375">15/16</option>
</select>

[HikksNotAtHome]

In article <bh*********@drn.newsguy.com>, Lee <RE**************@cox.net>
writes:

The most efficient solution would seem to be:

<select onchange="checkIt(this.value)">
<option value="0">0</option>
<option value="0.0625">1/16</option>
<option value="0.125">1/8</option>
..............
<option value="0.875">7/8</option>
<option value="0.9375">15/16</option>
</select>

I agree with you 100%. I have no control over how the select is generated
though. If I loop through the select and options and programatically change
them to decimals, I have to convert them back to fractions (or, swap them with
the text of the option) so that when the form gets submitted, I get the
fraction instead of the decimal. The eval still seems to be more efficient
though, in the end. ::sigh::
--
Randy
All code posted is dependent upon the viewing browser
supporting the methods called, and Javascript being enabled.

[Steve van Dongen]

On 17 Aug 2003 17:55:11 GMT, hi************@aol.com (HikksNotAtHome)
wrote:

This is a very simplified example of an Intranet application. But, is there an
easier (or more efficient) way of getting the decimal equivalent of a fraction?

I know that I could change it and have it look up the decimal equivalent for
the fraction, with an Object or an Array. I could also split the value on the /
and convert them to numbers, do the division, and get the decimal.

I discounted the Object/Array approach (and may go back to it, but doubt it).
The reason is that its not just 16ths. Some of the selects are 37ths, some are
7ths, one of them has 128ths in it. Simply put, I don't think that maintaining
10+ objects to contain the decimals is efficient (but open to that approach).
Also, its subject to get some added (different denominators), and some removed.

The split approach just doesn't seem to be as efficient though.

function checkIt(selVal){
myNumber = selVal.split('/')
decimalForm = parseFloat(myNumber[0])/parseFloat(myNumber[1])
//do some calculations here with valueInDec1
}

When the form gets submitted, it has to be in fractional form.

The Object/Array approach involves a fair bit of overhead, either in
download time if all the contents are hardcoded, or time if they are
generated at some point.

As far as parsing the string goes, I've always considered compiled
code to be faster than script-based parsing; split may not do as bad
as you think.

Usually I recommend against using eval because there is almost always
a clearly better way of accomplishing the goal, but in this case I
suspect eval will be equal to or better than any other method, plus,
it appears that the input strings are static, so I think eval is the
way to go.

Regards,
Steve

[Richard Cornford]

"HikksNotAtHome" <hi************@aol.com> wrote in message
news:20***************************@mb-m21.aol.com...
<snip>

function checkIt(selVal){
valueInDec1 = eval(selVal);
//do some calculations here with valueInDec1
}
If you just want to avoid the word "eval" appearing in your source code
you could use:-

valueInDec1 = new Function ('return ('+selVal+');')();

- but that will perform worse that eval. And the Function constructor
will probably use eval internally so it is really no more than avoiding
eval by name.

<snip> function checkIt(selVal){
myNumber = selVal.split('/')
decimalForm = parseFloat(myNumber[0])/parseFloat(myNumber[1])
//do some calculations here with valueInDec1
}

Unary + is about 4 times faster at string to number conversion than
parseFloat, so:-

decimalForm = (+myNumber[0])/(+myNumber[1]);

-should be fractionally quicker. But then the division operator requires
numeric operands so it will type-convert anyway so just:-

decimalForm = (myNumber[0]/myNumber[1]);

- will do.

Generally I can't think of a non-eval approach that is better than the
split-and-divide function above.

Richard.

[HikksNotAtHome]

In article <bh*******************@news.demon.co.uk>, "Richard Cornford"
<Ri*****@litotes.demon.co.uk> writes:

Unary + is about 4 times faster at string to number conversion than
parseFloat, so:-

decimalForm = (+myNumber[0])/(+myNumber[1]);

-should be fractionally quicker. But then the division operator requires
numeric operands so it will type-convert anyway so just:-

decimalForm = (myNumber[0]/myNumber[1]);
The difference I see, in testing, of these two:

decimalForm = (myNumber[0]/myNumber[1]);
decimalForm = parseFloat(myNumber[0])/parseFloat(myNumber[1])

Is in the 50-60 ms range, on 400-500 ms, The more passes, the stranger the
results get.
- will do.

Generally I can't think of a non-eval approach that is better than the
split-and-divide function above.

Changing the test that I posted to
decimalForm = (myNumber[0]/myNumber[1]);
instead of the parseFloat's, I still get 440-500 milliseconds on 10,000 passes.

I do see a difference with/without the parseFloat though (about 50-60
milliseconds on 10,000 passes).

With eval, I get in the range of 110 - 220. Which actually makes eval faster
than the split/divide method in any form.

Upping it to 50,000 passes, I get these results:

Test1:
eval took 1100 milliseconds
split took 8070 milliseconds
split with parseFloat took 8020 milliseconds
split with parseInt took 8020 milliseconds

Test2:
eval took 1050 milliseconds
split took 7960 milliseconds
split with parseFloat took 8020 milliseconds
split with parseInt took 8180 milliseconds

Test3:
eval took 1090 milliseconds
split took 7910 milliseconds
split with parseFloat took 8080 milliseconds
split with parseInt took 8020 milliseconds

Test4:
eval took 550 milliseconds
split took 7860 milliseconds
split with parseFloat took 7960 milliseconds
split with parseInt took 7970 milliseconds

Test5:
eval took 1040 milliseconds
split took 8130 milliseconds
split with parseFloat took 8020 milliseconds
split with parseInt took 8130 milliseconds

The test page I used can be found at:
<URL: http://members.aol.com/hikksnotathom...est/index.html />
Its set at 50,000 passes, view-source: may be a better way to get to it than
letting it potentially lock up a browser.
I added a 4th test, using parseInt instead of parseFloat, just for
kicks/giggles.

I had a new Function test in it but it was killing it, 11,000+ milliseconds (as
you guessed it would do), so I took it out.

But I wasn't avoiding eval for the sake of avoiding it. When I first tried it,
I actually expected it to be slower, and got to looking at it, and thinking
about what it was actually doing, and then started testing it. Saw where eval
was faster, so I wanted other opinions on it. There may even be a flaw in my
test itself?
--
Randy
All code posted is dependent upon the viewing browser
supporting the methods called, and Javascript being enabled.

[Richard Cornford]

"HikksNotAtHome" <hi************@aol.com> wrote in message
news:20***************************@mb-m06.aol.com...
<snip>

alerts 330 and 500 for me, which indicates eval is faster. But
thats on a 2.2 Ghz Machine with 256mbs Ram. Will test it again
tomorrow at work, on a slower machine, just to see how much
difference it makes. But it can only make a bigger difference.

<snip>

It looks to me like the results vary considerably depending on the
browser. For IE eval is quickest, on Opera 7 and Mozilla 1.3 split
followed by division comes out on top (500 Mhz PIII Win 98).

Taking:-
myNumber = tempObj[c].split('/')
N = myNumber[0]/myNumber[1];
- as the basis for comparison (100% on each browser)
IE = 100%
Op = 100%
Mz = 100%

myNumber = tempObj[c].split('/')
N = (+myNumber[0])/(+myNumber[1]);
IE = 98.74%
Op = 98.36%
Mz = 100%

myNumber = tempObj[c].split('/')
N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);
IE = 108.26%
Op = 118.03%
Mz = 109.52%

N = eval(tempObj[i]);
IE = 17.50%
Op = 154.10%
Mz = 122.22%

N = new Function('return ('+tempObj[i]+');')();
IE = 163.58%
Op = 286.88%
Mz = 152.38%

For comparison, this is my test page:-

<html>
<head>
<title></title>
<script type="text/javascript">
var frm = null;
var fncts = ['emptyL','evalVersion','FunctionObject',
'splitNparseFloat','splitNunaryPlus','splitNdivide '];
var running = false;

function setButtons(bl){
frm['loopLimit'].disabled = bl;
var sw = frm['bt'];
if(typeof sw.length == 'undefined'){
sw = [sw];
}
for(var c = 0;c < sw.length;c++){
sw[c].disabled = bl;
}
}

var tempObj = new Object();
function doTests(){
if(!running){
var t = ['','/',''];
frm = document.forms['f'].elements;
setButtons(true);
t[2] = +frm['loopLimit'].value;

for (var k=0;k<t[2];k++){
t[0] = k;
tempObj[k]= t.join('');
}
frm["Dur0"].value = '';frm["Avr0"].value = '';
for(var c = 1;c < fncts.length;c++){
frm["Dur"+c].value = '';
frm["Avr"+c].value = '';
frm["CAvr"+c].value = '';
frm["PAvr"+c].value = '';
frm["Res"+c].value = '';
}
running = true;
act(0);
}
}
function act(p){
/* setTimeout is used to minimise the occurrences
of 'a script on this page is running slow' dialogs. */
if(p >= fncts.length){
setTimeout('report()',100);
}else{
setTimeout((fncts[p]+'('+p+');'),200);
}
}
function report(){
var lim = +frm['loopLimit'].value;
var emDur = +frm["Dur0"].value
var unaC = (frm["Dur"+(fncts.length-1)].value - emDur) / lim;
frm["CAvr"+(fncts.length-1)].value = unaC;
frm["PAvr"+(fncts.length-1)].value = '100';
for(var c = 1;c < (fncts.length-1);c++){
if(frm["Cb"+c].checked){
var evaC = (frm["Dur"+c].value - emDur) / lim;
frm["CAvr"+c].value = evaC;
frm["PAvr"+c].value = ((evaC/unaC)*100);
}else{
frm["CAvr"+c].value = 'X';
frm["PAvr"+c].value = 'X';
}
}
setButtons(false);
running = false;
}
function emptyL(p){
var lim = +frm['loopLimit'].value;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
;
}
totTime = (new Date().getTime() - stTime)
frm["Dur0"].value = totTime;
frm["Avr0"].value = (totTime/lim);
act(p+1);
}
function evalVersion(p){
if(frm['Cb'+p].checked){
var lim = +frm['loopLimit'].value;
var N;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
N = eval(tempObj[c]);
}
totTime = (new Date().getTime() - stTime)
frm["Dur"+p].value = totTime;
frm["Avr"+p].value = (totTime/lim);
frm["Res"+p].value = N;
}else{
frm["Dur"+p].value = 'X';
frm["Avr"+p].value = 'X';
frm["Res"+p].value = 'X';
}
act(p+1);
}
function FunctionObject(p){
if(frm['Cb'+p].checked){
var lim = +frm['loopLimit'].value;
var N;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
N = new Function('return ('+tempObj[c]+');')();
}
totTime = (new Date().getTime() - stTime)
frm["Dur"+p].value = totTime;
frm["Avr"+p].value = (totTime/lim);
frm["Res"+p].value = N;
}else{
frm["Dur"+p].value = 'X';
frm["Avr"+p].value = 'X';
frm["Res"+p].value = 'X';
}
act(p+1);
}
function splitNparseFloat(p){
if(frm['Cb'+p].checked){
var lim = +frm['loopLimit'].value;
var N,temp;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
temp = tempObj[c].split('/')
N = parseFloat(temp[0])/parseFloat(temp[1]);
}
totTime = (new Date().getTime() - stTime)
frm["Dur"+p].value = totTime;
frm["Avr"+p].value = (totTime/lim);
frm["Res"+p].value = N;
}else{
frm["Dur"+p].value = 'X';
frm["Avr"+p].value = 'X';
frm["Res"+p].value = 'X';
}
act(p+1);
}
function splitNunaryPlus(p){
if(frm['Cb'+p].checked){
var lim = +frm['loopLimit'].value;
var N,temp;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
temp = tempObj[c].split('/')
N = (+temp[0])/(+temp[1]);
}
totTime = (new Date().getTime() - stTime)
frm["Dur"+p].value = totTime;
frm["Avr"+p].value = (totTime/lim);
frm["Res"+p].value = N;
}else{
frm["Dur"+p].value = 'X';
frm["Avr"+p].value = 'X';
frm["Res"+p].value = 'X';
}
act(p+1);
}
function splitNdivide(p){
if(frm['Cb'+p].checked){
var lim = +frm['loopLimit'].value;
var N,temp;
var totTime,stTime = new Date().getTime();
for(var c = 0;c < lim;c++){
temp = tempObj[c].split('/')
N = temp[0]/temp[1];
}
totTime = (new Date().getTime() - stTime)
frm["Dur"+p].value = totTime;
frm["Avr"+p].value = (totTime/lim);
frm["Res"+p].value = N;
}else{
frm["Dur"+p].value = 'X';
frm["Avr"+p].value = 'X';
frm["Res"+p].value = 'X';
}
act(p+1);
}
</script>
</head>
<body>
<p>
<form name="f" action="#">
Loop Length = <input type="text" value="50000"
name="loopLimit"> Some browsers will put up an &quot;A script on
this page is making the browser run slowly&quot; dialog. If this
happens the results for the test will be invalid and a shorter loop
will be needed. However, JavaScript Date objects do not tend to be
accurate to less than 10 milliseconds so duration results that are
not different by at least 20 milliseconds (and preferably 100+) are
not necessarily meaningful and a longer loop may be needed to acquire
useful results.<br><br>
<input type="button" value="Test" name="bt" onclick="doTests();">
Repeat tests to reduce/expose the influence of background tasks.
<br><br>
Empty Loop Duration (milliseconds) = <input type="text" value="X"
name="Dur0"><br>
Empty Loop Average (milliseconds) = <input type="text" value="X"
name="Avr0" size="22"><br><br>
Test <code>N = eval(tempObj[i]);</code> ? <input type="checkbox"
name="Cb1" checked><br>
<code>N = eval(tempObj[i]);</code> Duration (milliseconds) =
<input type="text" value="X" name="Dur1"><br>
<code>N = eval(tempObj[i]);</code> Average (milliseconds) =
<input type="text" value="X" name="Avr1" size="22"><br>
(result = <input type="text" value="X" name="Res1" size="22">)<br><br>
Test <code>N = new Function('return ('+tempObj[i]+');')();</code> ?
<input type="checkbox" name="Cb2"><br>
<code>N = new Function('return ('+tempObj[i]+');')();</code>
Duration (milliseconds) = <input type="text" value="X"
name="Dur2"><br>
<code>N = new Function('return ('+tempObj[i]+');')();</code>
Average (milliseconds) = <input type="text" value="X"
name="Avr2" size="22"><br>
(result = <input type="text" value="X" name="Res2" size="22">)<br><br>
Test <code>N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);</code>
? <input type="checkbox" name="Cb3" checked><br>
<code>N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);</code>
Duration (milliseconds) = <input type="text" value="X"
name="Dur3"><br>
<code>N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);</code>
Average (milliseconds) = <input type="text" value="X"
name="Avr3" size="22"><br>
(result = <input type="text" value="X" name="Res3" size="22">)<br><br>
Test <code>N = (+myNumber[0])/(+myNumber[1]);</code> ? <input
type="checkbox" name="Cb4" checked><br>
<code>N = (+myNumber[0])/(+myNumber[1]);</code>
Duration (milliseconds) = <input type="text" value="X"
name="Dur4"><br>
<code>N = (+myNumber[0])/(+myNumber[1]);</code>
Average (milliseconds) = <input type="text" value="X"
name="Avr4" size="22"><br>
(result = <input type="text" value="X" name="Res4" size="22">)<br><br>
Test <code>N = myNumber[0]/myNumber[1];</code> ?
<input type="checkbox" name="Cb5" checked><br>
<code>N = myNumber[0]/myNumber[1];</code>
Duration (milliseconds) = <input type="text" value="X"
name="Dur5"><br>
<code>N = myNumber[0]/myNumber[1];</code>
Average (milliseconds) = <input type="text" value="X"
name="Avr5" size="22"><br>
(result = <input type="text" value="X" name="Res5" size="22">)<br><br>
<input type="button" value="Test" name="bt" onclick="doTests();">
Repeat tests to reduce/expose the influence of background tasks.
<br><br>
Results: (duration of test - duration of empty loop) / loop length<br>
<code>N = eval(tempObj[i]);</code> Average (milliseconds) =
<input type="text" value="X" name="CAvr1" size="22"><br>
<code>N = new Function('return ('+tempObj[i]+');')();</code>
Average (milliseconds) = <input type="text" value="X"
name="CAvr2" size="22"><br>
<code>N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);</code>
Average (milliseconds) = <input type="text" value="X"
name="CAvr3" size="22"><br>
<code>N = (+myNumber[0])/(+myNumber[1]);</code> Average (milliseconds)
= <input type="text" value="X" name="CAvr4" size="22"><br>
<code>N = myNumber[0]/myNumber[1];</code> Average (milliseconds) =
<input type="text" value="X" name="CAvr5" size="22"><br><br>
Differences (<code>((Math.round(Math.random()))||-1)</code> = 100%)
<br><code>N = eval(tempObj[i]);</code> = <input type="text" value="X"
name="PAvr1" size="22">%<br>
<code>N = new Function('return ('+tempObj[i]+');')();</code> =
<input type="text" value="X" name="PAvr2" size="22">%<br>
<code>N = parseFloat(myNumber[0])/parseFloat(myNumber[1]);</code> =
<input type="text" value="X" name="PAvr3" size="22">%<br>
<code>N = (+myNumber[0])/(+myNumber[1]);</code> = <input type="text"
value="X" name="PAvr4" size="22">%<br>
<code>N = myNumber[0]/myNumber[1];</code> = <input type="text"
value="X" name="PAvr5" size="22">%<br>
</form>
</p>
</body>
</html>

Richard.

[Dr John Stockton]

JRS: In article <20***************************@mb-m21.aol.com>, seen in
news:comp.lang.javascript, HikksNotAtHome <hi************@aol.com>
posted at Sun, 17 Aug 2003 17:55:11 :-

This is a very simplified example of an Intranet application. But, is there an
easier (or more efficient) way of getting the decimal equivalent of a fraction?

The actual function gets the select values, this one is a simplified version
where its passed.

function checkIt(selVal){
valueInDec1 = eval(selVal);
//do some calculations here with valueInDec1
}

<select onchange="checkIt(this.value)">
<option value="0">0</option>
<option value="1/16">1/16</option>
<option value="1/8">1/8</option>
..............
<option value="7/8">7/8</option>
<option value="15/16">15/16</option>
</select>

Don't be afraid of eval. In this case you need to evaluate a string
chosen at run time, for which eval is appropriate. Pre-conversion of
value to a numeric literal form gives long numbers or inaccuracy, if the
denominator is not a power of two.

Speed does NOT matter, since the evaluation only occurs after a manual
operation. Here, efficiency is code shortness.

--
© John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v4.00 MIME. ©
Web <URL:http://www.merlyn.demon.co.uk/> - FAQqish topics, acronyms & links;
some Astro stuff via astro.htm, gravity0.htm; quotes.htm; pascal.htm; &c, &c.
No Encoding. Quotes before replies. Snip well. Write clearly. Don't Mail News.

[HikksNotAtHome]

In article <20***************************@mb-m02.aol.com>,
hi************@aol.com (HikksNotAtHome) writes:

But I wasn't avoiding eval for the sake of avoiding it. When I first tried
it, I actually expected it to be slower, and got to looking at it, and thinkingabout what it was actually doing, and then started testing it. Saw where eval
was faster, so I wanted other opinions on it. There may even be a flaw in my
test itself?

Many thanks to everyone who replied.

For now, I am going to go ahead and use the split/divide approach, not because
of speed (I dont see how it can make *any* perceptible difference to the user,
not when you start talking in the decimal range of ms to execute it once,
either way), for no other reason than my own sanity with regards to eval. I
have spent the better part of 6 months fighting the use of eval. It took me 2
months to show them that I could get at any form element, no matter how
bastardized they made it, without eval, so instead of going back and saying
"Hey, in IE, eval actually does it faster", I will just use the split/divide
method. Then, if its ever converted (the intranet), to a non-IE environment, I
will have the fastest of the 2 already in place.

Again, thanks to all. Now, I can get to hacking/whacking at Jims crypto page.
--
Randy
All code posted is dependent upon the viewing browser
supporting the methods called, and Javascript being enabled.

[Grant Wagner]

HikksNotAtHome wrote:

In article <20***************************@mb-m02.aol.com>,
hi************@aol.com (HikksNotAtHome) writes:

But I wasn't avoiding eval for the sake of avoiding it. When I first tried
it, I actually expected it to be slower, and got to looking at it, and

thinking

about what it was actually doing, and then started testing it. Saw where eval
was faster, so I wanted other opinions on it. There may even be a flaw in my
test itself?

Many thanks to everyone who replied.

For now, I am going to go ahead and use the split/divide approach, not because
of speed (I dont see how it can make *any* perceptible difference to the user,
not when you start talking in the decimal range of ms to execute it once,
either way), for no other reason than my own sanity with regards to eval. I
have spent the better part of 6 months fighting the use of eval. It took me 2
months to show them that I could get at any form element, no matter how
bastardized they made it, without eval, so instead of going back and saying
"Hey, in IE, eval actually does it faster", I will just use the split/divide
method. Then, if its ever converted (the intranet), to a non-IE environment, I
will have the fastest of the 2 already in place.

Again, thanks to all. Now, I can get to hacking/whacking at Jims crypto page.
--
Randy

May I also suggest that instead of:

<select onchange="func(this.value);">

you use:

<select onchange="func(this.options[this.selectedIndex].value);">
:)

--
| Grant Wagner <gw*****@agricoreunited.com>

* Client-side Javascript and Netscape 4 DOM Reference available at:
*
http://devedge.netscape.com/library/...ce/frames.html

* Internet Explorer DOM Reference available at:
*
http://msdn.microsoft.com/workshop/a...ence_entry.asp

* Netscape 6/7 DOM Reference available at:
* http://www.mozilla.org/docs/dom/domref/
* Tips for upgrading JavaScript for Netscape 7 / Mozilla
* http://www.mozilla.org/docs/web-deve...upgrade_2.html