Follow-up Comment #1, bug #66328 (group groff):

I checked the output of 4.2BSD and 4.3BSD-Reno _ms_, and _groff_ 1.22.4 was
indeed more consistent with them than 1.23.0 was, in this respect.

(For the purpose of the specimen below, 4.2BSD and 4.3BSD-Reno implementations
of _ms_ appear to produce identical PostScript.  I'm using _groff_ Git as the
_formatter_ for them; those BSD releases did not support PostScript output at
all.) 

I say "more" consistent, not "completely" consistent.

I observe further points.

A.  BSD _ms_ spaced leader dots _much_ more densely than _groff_ does.  This
is of course as unspecified in _ms_ documentation (anyone's--until _groff_
1.23.0 introduced the `TC-LEADER` string) as the line length change, but I am
disinclined to change _groff_ for closer BSD emulation in this area.  I think
_groff ms_'s leader dot spacing looks better, and makes the gap before the
page number seem less like a yawning chasm.  (The span of that chasm is
configurable in _groff_ 1.23.0 [`TC-MARGIN`], but not elsewhere short of
altering or replacing the package's macros.) 

B.  _groff ms_ numbers the TOC page (in lowercase roman) in both 1.22.4 and
1.23.0, BSD _ms_ does not.

C.  _groff_ 1.23.0
[https://git.savannah.gnu.org/cgit/groff.git/tree/NEWS?h=1.23.0#n484 increases
the default line length as documented in the "NEWS" file].  All _ms_
implementations known to me--all the way back to Unix V7 _ms_--support
registers to change this line length and page offset, and _groff_ has
supported custom paper sizes for decades.  That's important for users outside
the U.S. and Canada.

D.  BSD _ms_ exhibits cut marks on the pages; _groff ms_ does not.

E.  BSD _ms_ uses a larger top margin before emitting the TOC.

F.  _groff_ 1.22.4 _ms_ exhibits a bug with this input; a spurious row of
leader dots on a line by itself follows the last TOC entry.  _groff_ 1.23.0
fixed this.

Specimen:


$ cat ATTIC/longtoc.ms 
.ds C1 Chapter the First, \
in which we Ramble On at Excessive Length for the Purpose of \
Contriving a Table of Contents Entry So Garroulous That It Will Surely \
Overrun the Line in Any Sensible Page Configuration\"
.ds C2 Chapter the Second, \
in which we Do Much the Same, \
but Apply an Added Twist\"
.ds C3 Of a Supplement for the Table of Contents, \
Which in a Traditional Work Might Be Something Like \
an Em-Dash-Separated List of Subsections or Sundry Other Topics \
Covered Therein
.NH 1
\*(C1
.XS
\*(C1
.XE
.LP
This is an
.I ms
document.
.NH 1
\*(C2
.XS
\*(C2
.XA no
.in 8n
\*(C3
.in
.XE
.LP
It is part of a regression test.
.TC


(file #56512, file #56513, file #56514, file #56515)

    _______________________________________________________

Additional Item Attachment:

File name: longtoc.4.2BSD.png             Size: 96KiB
    <https://file.savannah.gnu.org/file/longtoc.4.2BSD.png?file_id=56512>

File name: longtoc.4.3BSD-Reno.png        Size: 96KiB
    <https://file.savannah.gnu.org/file/longtoc.4.3BSD-Reno.png?file_id=56513>

File name: longtoc.groff-1.22.4.png       Size: 94KiB
   
<https://file.savannah.gnu.org/file/longtoc.groff-1.22.4.png?file_id=56514>

File name: longtoc.groff-1.23.0.png       Size: 91KiB
   
<https://file.savannah.gnu.org/file/longtoc.groff-1.23.0.png?file_id=56515>


    AGPL NOTICE

These attachments are served by Savane. You can download the corresponding
source code of Savane at
https://git.savannah.nongnu.org/cgit/administration/savane.git/snapshot/savane-54b47d16bd4a9240ffe81a7d43c02b33c9b6b5a0.tar.gz


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?66328>

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

Attachment: signature.asc
Description: PGP signature

Reply via email to