coopster
msg:3285063 | 11:11 pm on Mar 17, 2007 (gmt 0) |
I don't quite understand what you mean when you say it doesn't work ...? I'm running a simple test in both Fx and IE6 and they both work fine, as long as the object has been defined in the document first.
<input name="my_element" id="my_element">
<script type="text/javascript">
if (my_var = document.getElementById("my_element")) {
alert('yep, my_var exists: ' + typeof my_var)
} else {
alert('nope, my_var does not exist: ' + typeof my_var)
}
</script>
|
Moby_Dim
msg:3285258 | 6:10 am on Mar 18, 2007 (gmt 0) |
Seems strange to me too. I'd guess that IE has even more chanses to return true because ie returns true even in the case when the "name" attribute is "my_element" and id is "his_element" due to coder mistake, for example.
|
penders
msg:3285746 | 8:40 pm on Mar 18, 2007 (gmt 0) |
I agree with coopster... FF1.5, IE6 and Opera8 ('all browsers') all behave the same in the 'test for assignment' as far as I can tell. The reverse test... if (my_var = document.getElementById("element_does_not_exist")) {
alert('[1] yep, it exists!')
} else {
alert('[2] nope, it does not exist (expected result)')
} Produces the [2] alert in all browsers (as expected). | I'd assume this works in Fx because if the DOM call fails my_var is undefined and the if() test fails |
|
document.getElementById("element_does_not_exist") returns null (as opposed to undefined) in all browsers. I can only think that the reason why your test fails in IE is to do with where it is being called (before onload?) and maybe the order in which IE builds the page...?
|
Robin_reala
msg:3286067 | 9:55 am on Mar 19, 2007 (gmt 0) |
Thanks for the answers, all interesting :) For what it's worth it turned out to be a variable initialisation issue. Adding var my_var; before the if() test made this work in IE.
|
|
