bug#20404: 25.0.50; Sometimes no fontification with jit-lock-defer-time

[Stefan Monnier]

>> Hmm... while I guess it's still possible, it normally shouldn't happen:
>> if the fontification takes place during the scrolling itself, then the
>> next redisplay should not be skipped
> How does this work?

The decision whether redisplay is skipped or not is based on the
input-pending-p at the beginning of the command, i.e. before the call to
move_it.
So, if move_it triggers jit-lock which triggers font-lock, that means
jit-lock found (during move_it) that input-pending-p was nil, which
should imply that input-pending-p was also nil at the beginning of the
command, and hence the next redisplay will not be skipped.

> It's hard to say, but I'm almost sure I sometimes indeed see a small
> part of the scrolling window, near its beginning, fontified.  Or maybe
> I'm just imagining things.

Then I guess it's within the "explainable behavior".


        Stefan