Have you ever noticed that you can only highlight 1 object in the Object Explorer in SQL Server Management Studio? This usually is enough for most things you do, but if you want to script multiple objects at once… it becomes a pain. Well there is a way to script multiple objects at once, and it’s super easy.
Open the Object Explorer Details screen by selecting View -> Object Explorer Details.
Highlight the folder with the objects that you would like script. A list of all objects in that folder will appear on the right. This works for tables, stored procedures, etc…
Highlight all the objects that you want, right click, and select Create To.
Finally in SQL Server 2016, Microsoft SQL Server now supports JSON serialization. It provides it through the FOR clause… like XML serialization. This will take a recordset and output it in JSON format with very little effort by the coder.
As you can see. To get the output to come in JSON format, all you need to do is add FOR JSON PATH at the end of your query. Simple right? Here is what the results would look like.
Starting with SQL Server 2014, they added a feature called delayed transaction durability. This is a very interesting concept that could give you some speed improvements in your database. In this post, I will try to explain what this feature is and what some advantages/drawbacks are.
Transaction Log Durability
In 2014, SQL Server gives us the ability to control how the transaction log files are written to the disk. It gives us 2 options. Full transaction durability and Delayed transaction durability.
Full Transaction Durability – This is the default level. This means that as soon as a SQL transaction is executed, the transaction log is written to disk BEFORE control is sent back to the client/user.
Delayed Transaction Durability – This setting means that the SQL transactions will be written to a memory buffer instead of the disk. The control is then sent back to the client/user after writing to the memory buffer.
The delayed transaction durability has a few distinct advantage over full transaction durability.
The first advantage is that it doesn’t need to write to disk before giving control back to the client. In many databases, writing to the disk provides a lot of contention. Log file writes to the disk can sometimes slow queries down quite a bit. Writing to memory is much faster and should speed up your queries in general if your IO to the disk is slow.
Another advantage is that the delayed transaction durability buffers the writes to disk. SQL Server can get optimizations by buffering the write operations. Instead of writing each transaction to the disk, it will fill a buffer and then flush that buffer all at once.
Everything up to this point sounds great, right? Delayed transaction durability comes with one major pitfall. Delayed transaction durability could cause data loss if SQL Server crashes. The data that isn’t written to disk can be lost if SQL crashes.
When Should I Use Delayed Transaction Durability?
Only use delayed transaction durability if you can afford some data loss if SQL Server crashes. If you cannot afford any data loss, do not use delayed transaction durability.
How To Turn On/Off Delayed Transaction Durability
Okay, now that you’re an expert on delayed transaction durability, let’s quickly cover the 3 levels of transaction durability and how to activate them.
There are 3 kinds of transaction durability. Disabled, Allowed, and Forced. DISABLED – This means delayed transaction durability is turned off. All transaction writes will go to disk and there will be no potential for loss. FORCED – This means that delayed transaction durability is turned on for every query. All transaction writes will go to memory and there is a potential for data loss. ALLOWED – This is kinda in the middle. It basically means that you can turn on the delayed transaction durability in the query or stored procedure.
You will configure the durability with one of these statements:
At some point in your life, you’ve probably had to parse a string inside SQL Server. Maybe you had a comma delimited string and you need the data inside a table. So, you go to stackoverflow and find some stored procedure that someone made that you don’t understand. It works, but yuck. Well… those days are over! In SQL 20016 they introduced a new table-valued function called STRING_SPLIT. This makes is very easy to parse a comma delimited string in SQL Server.
The STRING_SPLIT function is really easy to use. It basically just takes 2 parameters. The first parameter is the string that you want to parse. The second parameter is the delimiter for parsing on.
The STRING_SPLIT function will always return a one column result set with a column named value.
Here’s an example of how to use it:
If your input string has spaces by the delimiter, you just need to trim the value that you get from the function. Here’s an example of trimming the spaces and also parsing on a pipe character: