Follow-up Comment #13, bug #67380 (group groff):

At 2025-10-25T21:41:57-0400, Dave wrote:
> Follow-up Comment #12, bug #67380 (group groff):
>
> [comment #11 comment #11:]
>> I think it strengthens the case for a GNU _troff_ extension to
>> the `fl` request that makes it argumentful,
>
> Ah, yes, if you're considering only expanded syntax that introduces
> new behavior, I have no objection: that breaks no (valid) existing
> usage.  I read the last paragraph of comment #8 as a proposal to
> change the behavior of an unadorned .fl.

Right--there I said (correcting my botched syntax for the no-break
control character):

> Also, we could change "'fl" to go ahead and write the document
> preamble.

Writing the preamble would be _necessary_ if the `fl` request were given
any arguments that marked various bits of output driver state as
"dirty"; doing so is synonymous with telling the driver to write
applicable _grout_ commands.  But it's not valid to write those commands
before the preamble is emitted.

So a hypothetical document that started this way...


'fl hpos
\X'pdf: /wackycoolimageinsertion feature /Exec'\c
foobar


...would need to write the document preamble upon reading the first
input line.

(Maybe in a real document, the flush request would come _after_ the
device extension escape sequence.  We're so deep into *roff estoerica
that I don't have any practical scenarios to offer.  Deri probably
could.  He knows all about device extension commands that alter
drawing properties of the output; image interpolation in particular
demands close management of the drawing position if chaos is to be
avoided.)

But in contrast, a document that looks like this...


'fl
foobar


...wouldn't *need* to write a document preamble at all.

And indeed it does not.


$ printf "'fl\n" | ~/groff-1.23.0/bin/groff -Z | grep . || echo NO OUTPUT
NO OUTPUT
$ printf "'fl\n" | ~/groff-HEAD/bin/groff -Z | grep . || echo NO OUTPUT
NO OUTPUT




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?67380>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

Attachment: signature.asc
Description: PGP signature

Reply via email to