I am trying to find a way to easily merge data from one development laptop to another.
Me and another developer are working on a .NET application that uses SQL 2k. Since we are the only two developers, and we both work from home, we spend a lot of time working remotely, so we each have a copy of SQL on our laptops. While this plan works fine for the most part, we are finding that keeping our systems synchronized to be more painful than one would think it should be. Initially, we tried just using DTS packages to copy tables and data from one system to the other, but this would cause one person's data they had to be lost (not a bad thing, but a real pain if you are testing a specific scenario and have to have the data setup a certain way).
We then tried using SQL merge replication, but since the tables are still in active development, modifying, adding, or changing tables is next to impossible (I did find the way to disable the replication temporarily and then re-enable it, but it takes too long just to add a column or something else simple). I would be fine with the replication scenario if someone could suggest a way to easily modify the table schemas with the replication enabled.
I am pretty sure there has to a simple way to do what we are doing, its just that neither of us are familiar with SQL enough to really see how to do it. Is there a way to syncronize each row's data before pushing new schema elements to other servers using DTS, or is there an easy way to enable the replication and still be able to have access to the table's schema?
Exactly. For the normal, day-to-day stuff that we do the merge replication is fine because we are connected over a network (VPN or local LAN), so that works beautifully, I would like to keep it if possible. The pain starts when I have to make an update to the tables or add a new table (which at this point in development is VERY frequent, at least one new table a week and slight modifications to all as time and testing goes along). Unfortunately, we do not have the luxury of working off a plan, so the tables tend to be rather "from the hip", which compounds this problem even more.
I suppose I am more accustomed to LDAP replication (I am learning SQL the hard way), which replicates the schema objects in addition to the data. I am looking for something along those lines, where the master (or in this case Publisher) not only defines the data objects to the subscribers, it defines the schema to be used. I don't see why it would be that much different in SQL.
If I could just get a nod in the right direction, or even a 'it would be possible to do it, but watch for this and that' I would be eternally grateful, this is driving me nuts.