I cannot show the true data here for reasons i'm sure you understand so all the below is renamed and randomized.
Each transformed_xml (unlike the example below) holds anywhere from 180 to 600 fields of data.
Please see the following LOAD statement:
LOAD id,
`a`,
`transformed_xml`,
`d`,
`e`,
`f`,
`g`,
h,
`i`,
`j`,
`k`,
`l`;
SQL SELECT *
FROM db.table;
We are currently using the ExtractValue function in the SQL Select portion of the Load which works but takes a very long time to run.. We therefor have to partition the the loads into small portions (usually 1000 lines at a time).
My goal here is to find a QlikView solution to parsing this XML portion as i am sure there is a quicker way to do this?
I have attached a demo version of the transformed_xml file as it would look on a single line of data.
I can't see how the usual XmlSimple qlik load would work in this instance as it would have to be in a resident load and would have to be executed on a loop per line..
It could be done by storing these xml as text-files and loading them again with a xml-fileformat. Similar to this is to load from_field. Here are various examples for it: https://community.qlik.com/search.jspa?q=xml+from_field.
