How does the REPLACE_MAPPING dataset work for special characters?

[pj5233]

In the 2022 version, we see that when we import a user with certain special characters those characters are not changed.  Then if you update any attribute on the user, it seems to fire off a replace of special characters with the replace_mapping dataset.  We would like to change which characters are replaced.  Questions on this:

1.How is this dataset being used.  There values look like this:

1. What actions cause this replace to happen?  (Example we see it on a user attribute edit, but not on a create.)

2. Do we have the option to edit this dataset to change what characters are replaced, and if so what is the format needed (is the replace doing on a contains or is it a full match, etc.)?

3. Is there a configuration setting to disable the replace and then we could create our own and do the logic we need ourselves?

[alanbixby]

I'm curious about this too.

There is some limited relevant documentation in "Database Functions" and the sub-articles of "Managing Datasets".

There's a function FN_EIC_REPLACE, that utilizes that table, and you can either modify that dataset or replace it with your own. For the latter, it's unclear if its mandatory to use ATTRIBUTE_KEYS and ATTRIBUTE_VALUES in the column name scheme, or not, etc.

I'm also not entirely sure how it fetches the keys since I don't see ATTRIBUTE_KEY values for basic accent characters- perhaps they are included already and just don't render properly in the interface, or it uses the individual characters within the key?

[rushikeshvartak]

Please raise FD ticket for same with required details