Comment on page
Edit an output
This article provides a guide on how to edit an output in Upsolver.
In order to ensure data lineage, once you run an output it becomes immutable (as it is a production task). However, you can still edit the output, but it does not stop it from running.
Changes to the output are saved automatically and the edited output is still the same logical output (effectively equivalent to a new version of an existing output).
The output appears in Draft mode while it is being edited and it appears as a separate tile in the Outputs page.
A suffix
_<Number> is added to the ID of the output to indicate that it is a new version of the output.Once you have edited the output, you can duplicate, discard, or run the edited output.
When you run the edited output, you can choose:
- To alter the existing table and write to a different location in the existing output data table, making it possible to identify which files were created by which version of the output.
- Create a new table. This is equivalent to duplicating the output. The existing table and output are not affected in any way by this operation.
Note: You can only edit the contents of a new lookup table that has not yet run.
Once a lookup table has run, you cannot edit the lookup table. Instead duplicate the lookup table and edit the copy.
1. From the Outputs page, select the output you wish to edit.
2. Click Edit.
3. Edit the output as required.
4. Click one of the following options:
Discard Draft
Duplicate
Run
Abandons the edited output.
Duplicates the edited output.
Runs the edited output.
To run the output:
- 1.Select whether to Alter Existing Table or Create New Table.
- 2.Click Next.
If you selected Create New Table:
- The Duplicate Output window will appear.
- This is equivalent to duplicating the output.
If you selected Alter Existing Table:
- The Run window will appear.
- Select the Compute Cluster and select the Processing Time Range.
- The original output finishes running at the time when the new edited output starts running.
- The data from the new version of the output is located in a separate folder.
- If you select a date and time in the past, any data stored after that date by the original output is deleted, and the new output processes the data from that point in time onwards.
- The original and new locations are transparent to the user.
Last modified 3yr ago