Deleted and Unused Classic Custom Fields remain in transactions

I am using desktop version 23.1.10.591 on Windows 11 64 bit.

I have updated to the latest version.

This latest version (and earlier versions) allow the deletion of Classic Custom Fields when data exists in these Custom Fields

Can you explain further. Does it clear the classic custom field data on any existing entry where it was used? If not then it is not really a bug as the data is preserved in the older entries, if it does then this needs to be looked into as it would be unexpected behaviour and has deleted real data when not expected to.

1 Like

In addition to what @eko just said, if the current behavior is what you described @AJD, I cannot recall weather the behavior was any different earlier since I know that certain fields and objects did not enforce referential integrity since the beginning. (i.e. you can delete a referenced field or object without causing any errors to appear anywhere)

I cannot recall if this was the case with Custom Fields so I will have to defer to @lubos on this.

Once the Classic Custom Field is deleted any data that existed does not appear as a column or on printed documents.

But the field data remain in the objects where the custom field is placed. So, you can simply Undo the deletion of the said custom field and get things back to normal.

I cannot place this into bugs for the following reasons:

  1. I’m not sure if this is the intended behavior of Classic Custom Fields
  2. Since the new Custom Fields have matured by now, it is only a matter of time before Classic Custom Fields are dropped, so it wouldn’t make sense to add an idea about something that is soon to be deprecated.

When the Classic Custom Fields are finally dropped I hope that the fields and data will be removed, because at the moment this all remains in the data file after converting to the new custom fields.

Now I think I can understand your issue a bit more, your problem was never about deleting the custom field in the first place but about the ghost data left behind.

As it stands right now, your only option to remove these data is to do a batch update, since these fields should still appear there.

However, regarding the deletion behavior, this is a design decision to be made by @lubos.

Personally, I agree that not being able to deal with ghost data after deletion of custom fields is somewhat of a problem but if the data was deleted along with it and you undo the deletion of the custom field, you will not be able to restore the data–and that would be a bigger problem, imho.

I think there should be another “safer” method to handle these kinds of issues but for now we still have batch update.

Yes, I have used Batch Update to remove the data, but the fields still remain in the Batch update file. The number of columns in Batch Update files has increased recently and having obsolete columns remaining does not help the situation.

My thoughts were that if it was changed so that it was not possible to delete the Classic Custom Fields that included data, then the fields would disappear when the data was all cleared and they were then deleted.

There’s been a recent bug for exactly that that’s been solved in version 23.1.2

What version are you running?

Different issue - there may be a connection though.

I can reproduce this, the data can be removed but the column remains.

There’s no way to delete the additional column.

Previously when you nullified a field value in an object, the entire field is removed from that object but now the value is removed while the field name/key remains.

image

I’m going to place this in bugs.

Thanks

This was an issue with both classic and new custom fields. Anyway, the latest version (23.1.18) is fixing this. Passing empty value in batch update will remove the custom field from the object rather then setting it empty.