[bug #66122] [mom] but for `BoxStop`, package is `-wall`-clean

[G. Branden Robinson]

URL:
  <https://savannah.gnu.org/bugs/?66122>

                 Summary: [mom] but for `BoxStop`, package is `-wall`-clean
                   Group: GNU roff
               Submitter: gbranden
               Submitted: Thu 22 Aug 2024 06:42:09 PM UTC
                Category: Macro package mom
                Severity: 3 - Normal
              Item Group: Warning/Suspicious behaviour
                  Status: Need Info
                 Privacy: Public
             Assigned to: PTPi
             Open/Closed: Open
         Discussion Lock: Any
         Planned Release: None


    _______________________________________________________

Follow-up Comments:


-------------------------------------------------------
Date: Thu 22 Aug 2024 06:42:09 PM UTC By: G. Branden Robinson <gbranden>
Hi Peter,

One of the "transparent device whatchamacallit" errors, the one that had
nothing to do with mom(7), fell very swiftly today.

I decided to turn on `-b` to help me track down the ones in _mom_ herself.

It occurred to me while I was there that while we know that `-ww` blitzes us
blue, `-wall`, the more gentle form prescribed for "traditional macro
packages", might not.

And this is the case.  In fact there appears to be only one thing standing
between _mom_ and "-wall" cleanliness.


troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24125: macro 'PAUSE'
troff: backtrace: file '../contrib/mom/examples/slide-demo.mom':270
troff:../contrib/mom/examples/slide-demo.mom:270: warning: setting computed
line length 0u to device horizontal motion quantum
troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24125: macro 'PAUSE'
troff: backtrace: file '../contrib/mom/examples/slide-demo.mom':273
troff:../contrib/mom/examples/slide-demo.mom:273: warning: setting computed
line length 0u to device horizontal motion quantum
troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24125: macro 'PAUSE'
troff: backtrace: file '../contrib/mom/examples/slide-demo.mom':276
troff:../contrib/mom/examples/slide-demo.mom:276: warning: setting computed
line length 0u to device horizontal motion quantum
troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24125: macro 'PAUSE'
troff: backtrace: file '../contrib/mom/examples/slide-demo.mom':281
troff:../contrib/mom/examples/slide-demo.mom:281: warning: setting computed
line length 0u to device horizontal motion quantum
troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24172: macro 'BOX'
troff: backtrace: file '../contrib/mom/examples/slide-demo.mom':286
troff:../contrib/mom/examples/slide-demo.mom:286: warning: setting computed
line length 0u to device horizontal motion quantum
troff: backtrace: '../contrib/mom/om.tmac':24613: macro 'BoxStop'
troff: backtrace: '../contrib/mom/om.tmac':24172: macro 'BOX'
troff: backtrace: file '../contrib/mom/examples/mom-pdf.mom':616
troff:../contrib/mom/examples/mom-pdf.mom:616: warning: setting computed line
length 0u to device horizontal motion quantum


Thanks to the backtrace we can see that this is the same problem over and
over.

Here's the code.


24612 .   if \\n[#IBR_ACTIVE] \{\                                            

24613 .     ll \\n[pdfbx-pre-ll]u                                            

24614 .     IB \\n[#IBL_ACTIVE]u \\n[#IBR_ACTIVE]u                           

24615 .     rr pdfbx-pre-ll                                                  

24616 .     rr #IBR_ACTIVE                                                   

24617 .     rr #IBL_ACTIVE                                                   

24618 .   \}


Apparently `pdfbx-pre-ll` is zero (possibly undefined).  Could this be a logic
error?  Would it be easy for you to fix?

Since you're deleting that register immediately, I conjecture that having it
undefined isn't a problem.

I therefore tentatively propose the following patch.


diff --git a/contrib/mom/om.tmac b/contrib/mom/om.tmac
index eb1c10181..8d1ae6298 100644
--- a/contrib/mom/om.tmac
+++ b/contrib/mom/om.tmac
@@ -24610,7 +24610,7 @@ .de BoxStop
 .     rr #IR_ACTIVE
 .   \}
 .   if \\n[#IBR_ACTIVE] \{\
-.     ll \\n[pdfbx-pre-ll]u
+.     ll (u;\\n[.H]>?\\n[pdfbx-pre-ll])
 .     IB \\n[#IBL_ACTIVE]u \\n[#IBR_ACTIVE]u
 .     rr pdfbx-pre-ll
 .     rr #IBR_ACTIVE


That quietens the build for me, and the shaded boxes demonstrated in
"slide-demo.pdf" look fine.

(Since the `.H` register _is_ the output device's horizontal motion quantum,
this simply does explicitly what a _roff_ formatter, even an old one, does in
such cases anyway.)


$ make -C build -j
make: Entering directory '/home/branden/src/GIT/groff/build'
make  all-recursive
make[1]: Entering directory '/home/branden/src/GIT/groff/build'
make[2]: Entering directory '/home/branden/src/GIT/groff/build'
  GROFF    doc/automake.pdf
  GROFF    contrib/mom/examples/letter.pdf
  GROFF    contrib/mom/examples/mom-pdf.pdf
  GROFF    contrib/mom/examples/mon_premier_doc.pdf
  GROFF    contrib/mom/examples/sample_docs.pdf
  GROFF    contrib/mom/examples/slide-demo.pdf
  GROFF    contrib/mom/examples/copyright-default.pdf
  GROFF    contrib/mom/examples/copyright-chapter.pdf
  GROFF    contrib/mom/examples/typesetting.pdf
troff: backtrace: '../contrib/mom/om.tmac':24111: string 'PDF_AUTHORS'
troff: backtrace: '../contrib/mom/om.tmac':4914: macro 'AUTHOR'
troff: backtrace: file '../contrib/mom/examples/mon_premier_doc.mom':13
troff:../contrib/mom/examples/mon_premier_doc.mom:13: error: cannot translate
character code 233 to special character ''e' in device-independent output
troff: backtrace: '../contrib/mom/om.tmac':24111: string 'PDF_AUTHORS'
troff: backtrace: '../contrib/mom/om.tmac':4914: macro 'AUTHOR'
troff: backtrace: file '../contrib/mom/examples/mom-pdf.mom':20
troff:../contrib/mom/examples/mom-pdf.mom:20: error: cannot write a node to
device-independent output
make[2]: Leaving directory '/home/branden/src/GIT/groff/build'
make[1]: Leaving directory '/home/branden/src/GIT/groff/build'
make: Leaving directory '/home/branden/src/GIT/groff/build'


As shown above, the only diagnostics remaining are those I promised to look
into.

Abusing this bug for unrelated purposes, I noticed that the "image" and "pdf
image" examples in "slide-demo.mom" have a layout problem; they don't appear
where they're supposed to but they _do_ show up...on a slide by themselves
later.

Would you like me to commit the change I proposed above?







    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?66122>

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

signature.asc
Description: PGP signature