Custom nameid getting released in version 2 but not version 3

[Cantor, Scott]

> 1)  Is there any way or reason the NameID generation process, particularly
> with legacy configuration, would be implicated?

If it prevents the release of the attribute being encoded, yes. If not, no. The OP implied essentially that filtering applied to some piece of the dependency chain leading to the attribute being encoded was impacting things, and doing so differently. I see no way for that to be possible because of how filtering works (it only runs after full resolution is done, in both versions).

> 2)  Is it relevant that they're using v2-style syntax here, and are the v3
> renderings of v2 filters the same as v2 renderings of v2 filters?

The language is the same, the meaning of what was posted is the same, the meaning of every possible filter policy is not the same.

> 3)  Is that documentation for v3.2 or higher or lower?  Requester and
> basic:AttributeRequesterString seem to both be in the examples, for
> instance, and I don't know whether the two rules behave literally exactly the
> same.

The syntax changes don't include any semantic changes, the implementations were done for 3.0.

-- Scott