Inline Line Editing Structure

gbrown · 2024-03-13T18:17:50+00:00

There was an error rendering this rich post.

Hello,

I am trying to use inline editing to make changes to a dataset by labeling a line item 'YES' or 'NO'.

The data I receive is replaced each morning (Some line items are updated, some are removed, some stay the same. ) Currently any changes I make today to the data will be erased the next morning. I have tried doing a recursive on the data to just append each day. How would I go about using inline editing and at what point of the dataflow should it be added to?

Example:

The item above has been marked as a yes. If that item shows up on tomorrows report I would still like the exception column to be labeled yes.

I have not found any solutions yet.

Thanks

Find more posts tagged with

Data Flows

Accepted answers

ArborRose

If I understand your issue correctly, you want to make edits in a webform that is being replaced each morning. My method is to make a separate form and join it to the webform via ETL. My changes and fields are in the side form…joined to the main dataset. No matter how the main set changes, my values for Yes/No stay the same as long as I join with the proper key field identifier.

By the way - I would recommend staying away from using question marks in a data heading. Characters like asterisk and question marks have specific meanings in code and could mess with logic. You could rename Exception? to IsException or something similar.

All comments

ArborRose

anewton

@ArborRose would you be able to expand some more on your logic with adding a separate form to the webform via ETL? I am having the same issue with my edited data being overwritten each time the ETL runs. I have tried adding the webform as an input data to join into my current data, however the webform is only showing as the 'Inline Placeholder App' which does not contain any actual data. I am not sure how to pull in the edited data from the inline editor. Thank you!!

ArborRose

The original posted question mentioned the dataset changing every morning. And that refresh was over-writing the responses entered.

If you use more than one dataset, you can refresh the main set without impeding upon your answers in another. Use a "key" field, like an ID that is common to both. In the data, let's assume the item no as a unique "key". Every line having a distinct item no.

Put the responses in a separate table and join them with a left outer join as shown below. Your main set can be refreshed without affecting the response set.

Note: a left outer join says - show me everything in the left table and only the things in the right table where I have a match on the matching field (item).

anewton

Thanks @ArborRose. Yes, so I have an ETL that transforms my two datasets, and outputs a final dataset. I'm currently using the inline editor on that final dataset - Total Transactions, shown below. So we make edits with the inline editor to the below total transactions data set. Then, when new data comes in, this entire dataflow is rerun and overwrites the inline editor changes previously made. I tried pulling in the Change Log, however, that did not have a column to join on (only rowId in which I could not find a way to find that in the Total Transactions dataset). And I cannot make the Total Transactions an append update method because when the ETL runs, it is appending the entire datasets, which creates duplicates.