Re: [precis] Enforcement as an Idempotent operation

[Peter Saint-Andre <[e-mail]>]

On 3/23/17 1:46 PM, William Fisher wrote:
> I agree with the "implementation note" strategy. In all my testing,
> only the "Nickname" profile can fail to be idempotent for some inputs.
> I have not found any inputs that fail the idempotent test using the
> Username or OpaqueString profiles.  I believe "Nickname" has problems
> because it uses NFKC.  I would add an implementation note/warning to
> the Nickname profile.

Hi Bill, thanks for your feedback. I propose the following text.

1. At the end of Section 7 ("Order of Operations") of 7564bis, add this
note:

   Because of the order of operations specified here, applying the rules
   for any given PRECIS profile is not necessarily an idempotent
   procedure (e.g., under certain circumstances, such as when Unicode
   normalization form KC is used, performing Unicode normalization after
   case mapping can still yield uppercase characters for certain code
   points); therefore, implementations might need to apply the rules
   more than once to an internationalized string.

2. At the end of Section 4 ("Use in Application Protocols") of 7700bis,
add this note:

   Implementation experience has shown that applying the rules for the
   Nickname profile is not an idempotent procedure for all code points.
   Therefore, implementations might need to apply the rules more than
   once to a nickname string.

3. I see no harm in also adding the following note to the end of Section
5 ("Use in Application Protocols") of 7613bis:

   Applying the rules for any given PRECIS profile is not necessarily
   an idempotent procedure for all code points. Therefore,
   implementations might need to apply the rules more than once to an
   internationalized string.

Peter