When dealing with large data structures, it seems like an 'Export' option is really only viable if the dataset is within the limitations of the format it is exporting to.. for example, newer version of Excel only support up to 1.1 million rows, and previous versions were only compatible up to 65k or so (memory ceilings I assume).
That said, it seems like a great solution would be the ability to auto-export a table to a format automatically in the script itself.
If I have a pre-built QVD that I want to divide into 3 separate exports (let's just say for argument's sake that it's by 'DIM1' dimension - or a specific site/location), what would be the method to have QlikView script do such a thing? How do you script a loop to only export the values in the 'current' dimension it is on until it has divded up the table into each respective individual dimension tables?
So, a combination of what you BOTH said seems more appropriate. Part of that has to do with my explanation of the issue, and expected output I'm looking for:
1. Erich - Your FOR loop looks like it checks for all possible values in a given field, and moves through them until all of the possible field values have been dealt with. This portion is exactly the kind of behavior I was looking to do...
2. Ralf - You actually had the exact output and concept I was looking for (where Erich was doing more of an immediate export to Word or similar app).
Now, I need to figure out a way to combine the two concepts... I think with this I should be able to collectively put it together. If I come up with an exact step-by-step walkthrough, I will most certainly come back and post it.
Just going to take some time, tweaking, and a little more referencing of the scripting language capabilities.
So the biggest issue here is that we need to be able to implement the variables in the filenames - in otherwords, as it is pushing through the different possible dimensions (say integers 1-3 are the values possible), it should add that as a portion of the filename.
1. Look at QVD, and for all rows where site_id = 1, save to SITE_INVENTORY_1.CSV
2. (loop assumed) Look at QVD, and for all rows where site_id = 2, save to SITE_INVENTORY_2.CSV
... the syntax that you referenced in the other post Erich didn't seem to apply easily to what I was trying to do because of this small detail.