Data stored as XML vs a database

[gregw74]

First off, I am incredibly green when it comes to XML, so please bear with me.

We currently have a product listing in the form of an HTML table, about 3000 rows long! To no surprise this table takes quite a while to load and is very painful to edit.

I have no problem getting the data into XML from an Excel spreadsheet but my question is how realistic/feasible would it be to query this much XML data? In practice, a user would specify search criteria and in turn only the relevant XML records are returned, essentially mimicking a database.

I know that a SQL database would be the best solution but that's on the back burner as a long-term solution. Right now we're just trying to find something that's more efficient than what we've got now.

Any advice anyone may have would be great!

Thanks,
Greg

[gregw74]

FYI: I merely want the XML data to be searched and the relevant results displayed. There would be no abilities to edit the XML data through the web-based interface, it would be read-only. I just wondered if processing an XML document containing 3000 records is unreasonable.

[Evilernie]

Without a webserver to process that xml, I would say you're out of luck and should stick with HTML if you cannot code client-side scripting.

If you can, you wouldn't have a problem in the first place.

If you're willing to pick up a book and read, here is how you're going to attack this and avoid having a web server and not buy $ a piece of software.

Microsoft provides a decent XML Editor. It's free. You can use that instead of using notepad.

Read up on XSLT and Javascript. With those knowledge, you can "inject" filtered rows from your XML into your HTML on the fly.

By the way, you don't have to be a Javascript expert to do this. You don't even have to pick up a javascript book. Most XML book that covers XSLT would have sample javascript code that you use.

My 2 cents.

[gregw74]

Evilernie,

I greatly appreciate your reply. What I've currently been testing is a working XML document (authored in Excel) and an ASP page that processes the XML file. The ASP form currently allows the XML file to be searched while only returning the relevant records to the client. It seems to work well.

I just wonder if 3000+ records is too much for an XML solution? Would it tax the server more than having one big raw table? Given that my only other option is a massive HTML table, what I'm currently testing surely seems to be the lessor evil.

By the way, do you know the name of that Microsoft XML application you referred to?

Thanks,

Greg

[Evilernie]

My apologies. You do have a server.

I wouldn't consider 3000 huge but a database will certainly help. Especially, if the table is indexed correctly. You can even just use MS Access and not go to SQL Server immediately.

As for the editor, it's called XML Notepad