I think I get asked this question a lot by individuals that are new to SQL, and they're new to well, DDL statements and construction of tables and manipulating tables. I usually get asked, "Well, when you make structural changes to a table or you create a table, how is it done? What are you thinking about?" Remember, as I was showing you throughout this course, that table structure has to be something you think about very carefully, what purpose is it going to serve? Many times, you create the table and then you alter the table, perhaps to add more columns in it, or perhaps to modify columns that you've already added. Here, you're going to create a new table, for example, and here, you're going to alter or modify a table. You may choose to add columns or modify columns. Yes, it happens all the time. It's something that during the development process, you're not really sure which columns are going to be needed, and which won't be needed, and for what purpose. Unless you get involved with the creation process and manipulation process, sometimes it's difficult to figure out why you would create a table and perhaps, later on, drop the table, remove it altogether, and then create a new one. You don't drop a table and alter it, you drop a table and create it, and then you drop it, and then you create it. Yes, this happens iteratively quite a bit. While you're in the process, you may choose to create a table, and in the development process, alter it quite a few times as well. Let me show you something as part of the creation process to give you an idea as to how easily things can get very complicated very quickly. I don't want to show you all the options of the create table process because it can get very involved, but I do want to show you some hints and some tidbits hopefully to spark an interest. Take a look at something here. As far as the create statement goes, let me show you what I'm talking about. If I do "create table customers," and I do a "custid," I can say that that's going to be a number. In the create statement, I can say that it's going to be generated by default as an identity column. Then I can say, custname is going to be a variable-length character 20. Then I can say that contact first name, and I can say that's going to be a variable-length character 2 with 20. Then I can say contact last name is going to be a variable-length character 2 20. Then I'm going to say contact_full_name is going to be a variable-length character 2 20. The interesting thing here is when I add this full name column, I can say that is going to be generated always as, and I can say that this is going to always be contact_fname plus, and I can say it's going to be combined with the two pipes are concatenations, or basically combining them together, contact_lname, like so. Then I'm going to say that the primary key will be custid. You see, this is a little bit of an involved statement, and I refresh this, becomes a little bit of an involved statement.
