Update of bug #67571 (group groff):
Status: Confirmed => Need Info
Assigned to: gbranden => barx
_______________________________________________________
Follow-up Comment #33:
I'm back to not knowing whether this is a bug or not.
Dave endorses the "OR-ing of character flags" behavior that GNU _troff_
presently manifests.
This means that as characters come and go as members of a class, the set of
flags that they manifest _as a consequence of membership in the class_ can
change.
(Though I would add that the "go" part of the foregoing is not necessarily
well defined--bug #67703.)
But that in turn means that the Summary of this report cannot be a valid
complaint.
Of **course** `class` works differently depending on where in the input it's
called.
Throwing back to Dave, requesting another reproducer that illustrates the
behavior at issue, and an elucidation of how his expectations are consistent
with dynamic computation of a character's flags depending on class
membership.
Please, please, please use `pchar` to explore formatter state. I despair of
ever figuring this out if you don't.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?67571>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
signature.asc
Description: PGP signature
