Comments on: Find as you type sorting on large record sets

By: Cornelius

Cornelius — Fri, 18 Apr 2008 15:51:50 +0000

Hi,

thanks for the code, it really helped a lot. I found a little flaw in it: If the user types in a number which matches an index in the filtertext variable the first XML node is returned.

To avoid this change
[code]
pos = attributelist.indexOf(term, i);
if(pos > -1){
index = int( attributelist.substring(attributelist.indexOf(”", pos)) );
output[count] = list[index];
count++;
i = pos+term.length;
}
[/code]

[code]
pos = attributelist.indexOf(term, i);
if(pos > -1){
if (attributelist.charAt(pos-1) != “<”){
index = int( attributelist.substring(attributelist.indexOf(”", pos)) );
output[count] = list[index];
count++;
}
i = (pos + term.length < attributelist.length) ? pos + term.length : -1;
}
[/code]

I hope I got the code highlighting right

By: Sean Christmann

Sean Christmann — Mon, 03 Mar 2008 02:49:41 +0000

Hey guys, I just moved over to a new publishing system so the code above should be showing correctly now.

By: John Wright

John Wright — Fri, 02 Nov 2007 02:32:24 +0000

Actually, that is wrong above. It should be:

i = attributelist.indexOf(”>”, pos);

don’t add the pos, that will mess things up.

By: John Wright

John Wright — Wed, 31 Oct 2007 02:32:07 +0000

Arne, I found a solution for multiple fields. In the searchStringInAttribute method replace the setting of the new search position with this line:

i = pos + attributelist.indexOf(”>”, pos);

instead of what it is currently. This way when 1 match is found in the row it will go onto the next row.

By: arand

arand — Thu, 18 Oct 2007 02:31:34 +0000

I added code to pull additional text from other attributes however sometimes there are multiple occurrences of search term and the output will send multiple instances of that node. Any way to prevent that?

By: Arne

Arne — Wed, 16 May 2007 02:31:16 +0000

Now I understand: “ins” and ”\ins” as tags are some artefacts of your blogging tool; both should just be ”+” in AS code.

Sorry

By: Arne

Arne — Wed, 16 May 2007 02:30:58 +0000

I don´t get it; what exactly happens here:
———
for each(node in rawXMLList){
fullNameIndex = node.@fullName.toLowerCase()””; count++; }
——
What is this about and where does this come from?
I am very curious, because I need more XML parsing speed very urgent.

Kind regards
______
Arne

By: Henry

Henry — Wed, 11 Apr 2007 02:30:07 +0000

i’m so glad i found this page on google, i’m trying to implement exactly what you’re talking about (iTunes-like search on a directory with a huge XML dataset) and am running into issues with search speeds… was trying to do it in AS2 and flash but after reading your extremely helpful and enlightening post i’m going to start educating myself in AS3 and flex. thanks again for the great information and inspiration!

By: Nick

Nick — Wed, 28 Mar 2007 02:29:43 +0000

I actually found myself saying “crafty” while reading this post.
Thanks!

By: Joel

Joel — Mon, 26 Mar 2007 02:29:12 +0000

I just watched your demo from Apollo Camp, and this was one of the things I was curious about. Thanks for the details. Great stuff.