> No, I have not tried that yet. I wanted to be sure what the behaviour
> should be first. Actually I believe it worked as I expect it to long
> before, but I am not sure. It is however quite possible I got fooled by
> the different cases we have seen here.
I never used the `font-lock-face' property before hence I can't tell.
The documentation tells me nothing about whether it should override
("supersede") any other face / font-lock-face. The Elisp manual says
... This can be advantageous for special buffers
which are not intended to be user-editable, or for static areas of
text which are always fontified in the same way. ...
and none of these apply in your case. If you want to change the
behavior in your particular use case insert a line like
(unless (text-property-not-all beg (point) 'font-lock-face nil)
before
(when face (put-text-property beg (point) 'face face))
in `font-lock-fontify-syntactically-region'.