SQL: generate date/time series

Submitted by st on Fri, 27/04/2012 - 20:31

There are many examples of user defined table functions generating the series of dates(times) between two specified values. Usually, in OLAP you can see a table filled by the date/time series, too. Such table avoid to generate the values "on the fly".

However, in some cases you may be unable to use an UDF or a table. Fortunately, "pure SQL" method still works.

Suppose, the start and finish dates are specified as entry parameters. The transact SQL code below generates the series "moth by month".

Programming

SQL: delete duplicated rows

Submitted by st on Fri, 09/03/2012 - 16:17

How to delete the rows having the duplicated values of one or more columns and considered as duplicates? The example in Transact SQL is below.

Programming

SQL Server

Transact SQL: exporting table or view in CSV

Submitted by st on Mon, 20/02/2012 - 12:38

It's pretty easy on SQL Server but pay attention to the database/server rights of used connection. Usually, the preparing like sp_configure requires a system administrator privileges.

Programming

SQL Server

DateTime in C++

Submitted by st on Sat, 04/12/2010 - 13:48

Many years ago we had a need to manipulate the data of types "date" and "time" (DateTime in C# and Delphi/C++Builder). I had developed the corresponding class for these purposes.

Here is the interface with comments:

Programming

GUI: evolution of descriptions

Submitted by st on Thu, 02/12/2010 - 11:37

Let's see one simple example showing how the methods and the languages of graphical user interface (GUI) description are progressing (or regressing, it depends) during the last 15-20 years.

The primitive form description requires:

Indexing columns of bit type

Submitted by st on Wed, 24/11/2010 - 20:06

Clearly, an index build on only one column of the bit type is useless because the selectivity is poor (50% of a table in average). However, when a table has several columns of the bit type, the composite index may be efficient.

For the test example, we'll create a table and fill it with random data.

Programming

SQL Server

Dynamic filtering with SQL

Submitted by st on Mon, 15/11/2010 - 18:45

The one of the most frequent asked question in database development is "How to filter the data returned by a query according to some user selected criteria?". To complete the solution with dynamic ordering see my other blog post.

For every method template I will add icons indicating corresponding advantages and drawbacks.

SQL Server

SQL quine (self-reproducing SQL)

Submitted by st on Sat, 30/10/2010 - 14:18

A quine is a computer program which takes no input and produces a copy of its own source code as its only output. The challenge for programmers is to write the shortest code in the given language. Here are my examples in Transact SQL and in ANSI SQL.

Note that we do not take in account the output format printing by any console SQL clients. For example, we are unable to suppress the output of column names or separators in the common case.

Programming

Dynamic result ordering in SQL

Submitted by st on Tue, 18/05/2010 - 17:19

There are several methods to implement the dynamic ordering of result returned by SQL query. The examples of stored procedures below use the Microsoft's demo database AdventureWorks.

USE AdventureWorks
GO

Programming

SQL Server

Random sequence in Transact SQL query

Submitted by st on Sat, 04/07/2009 - 13:07

The rand() Transact SQL function produces a sequence of same numbers in the query. For example:

Programming

SQL Server

st's blog