Search and Index engine

[AndrewClark]

I know that it is possible to embed an HTML object in an AMS project, so I could use something like dtSearch to deliver searchable content using an AMS front end. However, I would prefer to use something directly within the AMS screens.

I am thinking of having an edit box where the user could type in a word they want to search for, and a button to start the search. Does AMS have a native content indexing engine, or are there any that can be embedded within it?

[Brett]

The only possibility that I can think of would be to use HTTP.Submit to submit the search to the search page and then parse the results. I know that is vague, but maybe it will give you something to go off of.

[AndrewClark]

Brett,

Thanks for your prompt reply.

Andrew

[Corey]

Well you can do this right in AMS in a variety of ways, depends on how you are storing your info. I'm not familiar with dtsearch but i'll assume it's a client side search script by the same guy who makes dttree.

The bottom line is parsing. Let's say you have only ten files, i.e. html files. That's easy to parse them as text files and find what you need. You can even use named anchors or custom tags to extract just the data portion related to the search, etc.

Best if you are handling much data is some sort of database solution. I reccomend the SQLlite plug-in for AMS coming out in the next week or two but you caould use anything, i.e. Flash, andy Javascript database, and web database, etc.

I saw some cool free Javascript based databases with built in search scripts around the web, perhaps that would be your best bet. Anyhow if you would rather just use .html files and ams to parse try using meta tags to simplify things and reduce the amount of text you need to parse. A master list of tags might also work well.

Bottom line, it's do-able in a variety of ways but first you have to have your data handling strategy abstracted so you know what you are dealing with...

Corey Milner
Creative Director, Indigo Rose Software