Table structure for inventory pricing
I am going to attempt to price my books. My plan is to download the source code of two pages where my books are for sale: Amazon.co.uk and also bookfinder.com. I may end up using just one of these pricing pages.
My plan is to scrape the prices of all of the offers available on these pages. Usually the cheapest offer is displayed first. I would also like to scrape the conditions of each of these books. Sometimes books are listed as Brand New etc.
I envisage that I would have a table with many prices and their corresponding conditions. I would then use a calculation to decide what price to set my book at.
What I would like advice on, at this stage, is how to structure this, table wise. I have an inventory table. This could be connected to the price table (yet to be developed) with sku as the matching field. Each record in the inventory table could have several child records. Each child record would have, ISBN, price, condition and vendor. (Its important to list the vendor as some of them have unfeasibly high prices. Therefor its important to discart certain vendors.) Is this a good structure?
After running the pricing script I would like to have the option of keeping my old prices or referring back to old prices. E.g. if I notice that sales were particularly good on a certain date I'd like to replace my today's prices with those of the date of the good sales.
A drop down list of dates for pricing dates would be handy. Or I may decide that this is not a good idea after all seeing as there is a school of thought that prices are only competitive for a very short time period.
One problem I have with this process is that it takes my database, which is on FileMaker Server 12, about four seconds to download a source code, parse it and run calculations on each record. So it could be a bit unfeasible to run it this way, as I can't use Filemaker Pro while a script is running. I hear that Amazon have an API for pricing. Has anyone any experience with it?