2 Replies Latest reply on Nov 17, 2010 12:30 AM by IvanYivoff

    data file production and development versions

    IvanYivoff

      Title

      data file production and development versions

      Post

      Hello there,

      I'm developing a solution where I keep different files for Data and for the Interface Layouts. That way the users can keep working with the data while I keep working on a new version of the Layout, and I just replace the layouts file when needed and everything is nice.

      But, there is problem. Now and then I see that I will need to modify the data file structure. Even if I don't want to touch the real schema; because of the way FM works I may need to add a "special" field. A global, a calculation, a summary field.... Normally I'd test those changes the same way that I would on the layouts: work on a dev version, and move it to production when ready.

      How would you proceed in those situations? I guess I'm not the first contemplating this sort of situation, and I was wondering if there were any sort of "best practices" for this. I can't just "replace" the production version with the dev version, because there could be loss of data, of course.

      Any better way to recreate the dev version in the production but to recreate the new fields one by one?

      Thanks and regards,

      I.-

        • 1. Re: data file production and development versions
          philmodjunk

          If you have a global field that is strictly for the purpose of facilitating a UI feature, you might define a local table of global fields in your UI file. I they need particular default values, you can set up a script to load these fields when the file is opened.

          For changes to the Data side that can't be avoided, you have two options:

          1. Wait until no one is using the system, and modify the working copy directly. This is often a convenient way to make small changes.
          2. Write a script in your data file that imports all the data from all the tables in the current data file and then updates the next serial value settings to your script. With this script, you can click a button and import all the current data into your new file, then swap out the files.
          • 2. Re: data file production and development versions
            IvanYivoff

            I guess it's more or less the "manual way" then. The importer script sounds nice... although it scares me a bit that something could go wrong and leave me with missing data.

            Hopefully I wont have to make many alterations to the initial schema.

            Thanks and regards,

            I.-