Re: Changes to hi-lock before release.

[Chong Yidong]

Richard Stallman <address@hidden> writes:

> I just clarified the documentation of this feature, as well
> as fixing Texinfo usage.  But I think some change is probably
> called for in the feature.
>
>     +(defvar hi-lock-file-patterns-policy 'never
>
> Shouldn't that be a defcustom?
> It should also have a risky-local-variable property
> since its value can be a function.

Let's just keep things as they are for now.

> Also, it seems absurd for an explicit call to hi-lock-find-patterns to
> check this variable.  If the user explicitly gives the command, it
> should be obeyed.  I think that hi-lock-file-patterns-policy should be
> tested only when hi-lock-find-patterns is called in other ways (such
> as by find-file).

What you suggested is indeed the behavior of hi-lock-find-patterns.