Forum Discussion
Increment Refresh Clarifications
Yeah, sorry GK600 - I understand. That is how it goes with self service projects. If you can get the system to export in to Excel format or CSV format, you can get things working, but there will be no incremental refresh on those either, the point of your original post.
If you have any problems going forward, start a new thread and we will jump in and help. In the mean time, please mark one or more of these posts as the solution so this thread can be marked as solved - even though not solved in the way you were hoping. 😉
Incremental Refresh requires query folding. Web APIs do not count. You would need an ODATA feed, or a true database like CDS, SQL Server, Azure SQL, Oracle, etc. And as you've discovered, if View Native Query is greyed out, you cannot do it. (There are very rare cases where that is not correct - greyed out and it still folds, but those are edge cases with SQL Server/Oracle, etc. and you can use your database profiling tool to see that even though the PQ UI thinks folding has broken, it is really continuing. This is only for a few command. Retrieving top rows I believe will cause this glitch to manifest)
It requires query folding because the service partitions your data and will generate a different SQL statement for each one - so the current refresh data goes in one partition with one set of data, and the old stuff goes in another partition, then the service will move data between those as they age out of the "current" dataset.
Hi GK600 do you have your answer? If not, post back and we'll try to help. If you do, can you mark the answer as such so we know this thread is done and other's searching will see the answer as well?
Hi edhans ,
Apologies for delay in repsonse.
We dont have sql server etc. available at this time. Ideally, I need to be using API but thats taking huge time for simple transformations. It connects to API everytime I need to look at lets say values in a column.
So, I started to explore importing and loading excel files (20-25 of them) which I will need to maintain manually. I checked at a high level and load time is about 5-10 minutes for 80% of data I could get easily so far. The files will reside on onedrive in that case. But sadly, I will need to maintain reporting manually every month and not real-time data. But I think thats best as of now.
What do you think? Do you suggest an alternative or think that's the right thing to do at this stage.
Thanks once again.
regards,
Gaurav
That certianly doesn't sound ideal GK600 but sometimes you have to go with what works to get a project moving. I would still try to figure out if there is a better way to get the data from the source. If the API doesn't work, could CSV files be dumped, or even get access to the source server behind the API. Don't know if you are accessing corporate data serving up the web API, or a 3rd party app, in which case they will not give you that kind of access.
Thanks for your response edhans ,
Its a corporate data but I doubt if supplier may give access to data source server. I will check though. Not sure how will I use that though but may figure out if needed.
The data cant be pulled into csv yet (I am pushing for that to be done as I believe the size of files will be much shorter). Currently, thats available only in excel.
regards,
Gaurav
