help parsing window.location.href

[wkenny]

Can anybody tell me how to parse window.location.href to get the page name.
e.g. thesite/thefolder/thesubfolder/thepage.htm
and I want to return "thepage"
I will not know in advance how deep the page name is.
Thanks

[jscheuer1]

<script type="text/javascript">
thePage=unescape(location.href)
thePage=thePage.substr(thePage.lastIndexOf('/')+1)
thePage=thePage.substr(0,thePage.indexOf('.')-1)
alert(thePage)
</script>Not foolproof but, good for normal page names without query strings* containing the '/' character.

*query strings follow the page name with a '?' and data, ex:

thesite/thefolder/thesubfolder/thepage.htm?var1=password

[jscheuer1]

<script type="text/javascript">
thePage=unescape(location.href)
thePage=thePage.indexOf('?')==-1? thePage.substr(thePage.lastIndexOf('/')+1) : thePage.substring(thePage.lastIndexOf('/')+1,thePage.indexOf('?'))
thePage=thePage.substr(0,thePage.lastIndexOf('.')-1)
alert(thePage)
</script>This one will work even with dots ('.') in the filename and with query strings containing ('/'). There could be other considerations I've missed for other potential location.href values.

[wkenny]

Just the trick. Thanks a lot, John

Billy

[jscheuer1]

Actually, looking at it again, my second version will still fail on query strings containing ('/'). Use:

<script type="text/javascript">
thePage=unescape(location.href)
if(thePage.indexOf('?')!==-1)
thePage=thePage.substring(0,thePage.indexOf('?'))
thePage=thePage.substr(thePage.lastIndexOf('/')+1)
thePage=thePage.substr(0,thePage.lastIndexOf('.'))
alert(thePage)
</script>

[Twey]

Or just use %2f instead of / in GET query strings, like you're meant to.

[jscheuer1]

After it is unescaped, what good would that do?

[Twey]

None... but who mentioned unescaping it?

/EDIT: Oh wait, you did. I see why; perhaps unescape it after getting the appropriate portion? Check for ., then if it's -1, use %2e instead.

[jscheuer1]

I think unescaping it in the first place is the safest route, no telling what you are dealing with before that.

[Twey]

You can't escape the path, only the page. Split the page down, then unescape it. You needn't worry about the path.

[jscheuer1]

You are right in part but, going back to where you said:

Or just use %2f instead of / in GET query strings, like you're meant toI would, of course, always do that :rolleyes: . However, I have no control over what others have or may do. Still best to unescape first and deal with a known quantity than to hope everybody else is playing by the rules and basing your code on that.

[Twey]

Of course they'll follow it... it's a convention to follow the standard ;)
Due to it being a standard, I'd say it's safe to rely on it being followed. It might not work, but no-one will blame you. :)

[jscheuer1]

Why is there never a cold haddock around when you need one? :confused: ;)

[Twey]

Pfft, the congor eel beats it every time. ;)
Massive long things, great range. And they keep wriggling after you gut them. There's nothing worse to be hit in the face with than a long, cold, wet, headless, wriggling eel. :p