Chapter 4. Data Sources
We've completed the "seeing is believing" phase with big success. We've shown HighCloud Airlines the potential value that QlikView can bring to their business and how they will be able to give their raw data the meaning that their business requires to make everyday decisions. Now, the natural question that arises after seeing what QlikView can do on the frontend is: what type of database does QlikView require to work?
The straight answer to this question is simply that QlikView does not necessarily require a specific database or Data Warehouse (DWH) to pull data from. It can benefit from using a DWH, but that is not required. However, the data must reside somewhere in order to be able to pull it into QlikView, visualize it, discover patterns in it, and build all kinds of charts with it. That "somewhere" can be almost any standard database, flat file (for example, .xlsx or .csv), web page, web service, and so on, or even any combination of these.
When building the data model for the application created in the previous chapter, we used tables stored in a QlikView Data Format (QVD). However, as pointed out, this data can be stored and managed in a wide range of different systems. Therefore, it requires different methods for extraction. That's where data sources come in. In this chapter, you will learn:
- How to load data from different sources
- How to extract data using the built-in wizard
- What QVD and QVX files are
- How to load data from disparate sources
Although there are many different Database Management Systems (DBMS) out there, we can, for our purposes, group them into four different categories:
- Those that provide connectivity via ODBC/OLE DB drivers (we'll talk about what these are in a moment)
- Those that use proprietary systems with no standard connectivity
- Those that are located on the Web and connected to via APIs
- Those that are not necessarily DBMSs but, rather, have tables stored in plain files, such as Excel, CSV, TXT, XML, and the like
We'll discuss some key points in each of these categories so that we have a general understanding of the implications we must consider.