Introduction to the outset
Today, when you load data using QV, you run into some state in the column, and the information is separated by a symbol, which is not conducive to data analysis because the content in the string is itself a dimension. Search the Internet to find a solution to the method, record.
For example, in the first picture, s200,m250,r35 are all Invoice types, which need to be taken out as the dimension DIMENSION of the analysis.
You can use the following code to achieve the separation effect, take out all the Invoice type, so you can do statistical analysis of the data. Of course, the other columns will be re-data, the aggregation of the time to pay attention to.
Code
Source:Load *Inline[ID, Invoice 1, S200/M250/R35/2, M1687/A35/D879/F689/3, b33625/s55/];D estination:LoadID, NewinvoiceWhere Len(Newinvoice)> 0;LoadID, Invoice, SubField (Invoice,'/') asNewinvoice;Load *Inline[ID, Invoice 1, S200/M250/R35/2, M1687/A35/D879/F689/3, b33625/s55/];
For more bi articles, see the Bi-series essay list (SSIS, SSRS, SSAS, MDX, SQL Server) If you feel this article has helped you, please help recommend it to make it easy for others to see these articles quickly in the Biwork blog recommendations Bar.
QlikView ETL-method for separating strings SubField