Updating products from Filemaker ?
We have a Filemaker database and would possibly like to update our online store directly from it.
We use it for the internal management of our projects (author rights and the like since we are a publishing company). They would like to reduce the data entering so synchronizing the website from that database would be cool).
I know ODBC can be used to access Thirtybees tables but what about image creation, product removal etc… I fear some mechanisms wont be triggered if I do it that way (instead of importing through the TB .csv import module).
Is there an already made module or anything that could make my life simpler ? Is there an API of some sort ?
Which database are you using? Is it a custom one or a commercially produced one like bookhound or picwick’s database?
We have a very similar situation with a Windows-based POS system that needs to be synced into TB.
I wrote a program that runs as a service on the same machine as the Windows POS system, periodically reads the POS database, and sends the data over to TB using the webservice API. It handles product changes/additions/removals, stock changes, images etc etc.
The webservice API is quite rich in functionality, but it takes some programming skills and effort to use it.
We also use a high degree of automation in our business. We use the webservice for some things and for others we write directly to the database itself over a secure connection.
The web service is ideal for creating the listings when a site is first launched as you can upload photos and you can be sure that the site and each item will be properly set up.
On the other hand if you need to update prices for 100,000 different products (for example) it will take a long time over the webservice and require many separate calls. However if you upload a CSV file with new prices, read it into a temporary table using MySQL commands, and then use SQL joins to update the data, you can update many thousands of items in just a few seconds.
Of course writing directly to the database tables is not a recommended or supported idea. Newer versions of 30bz will probably change the database structure somewhat and any code that writes directly to the database tables many need to be rewritten to be compatible with new versions of the platform. This is the risk, but until there is a bulk update functionality in the webservice (which I hope happens!) direct table updates are the only viable way to update vast amounts of data in a reasonable amount of time.
Awesome infos guys, thanks a lot, I will take a look at the webservice API !
jnsgioia yes, it is custom.