Re: [Groff] Bug in -ms multi-column output with different odd/even page

[Werner LEMBERG]

A mail from the distant past...

From: Egil Kvaleberg <address@hidden>
Subject: [Groff] Bug in -ms multi-column output with different odd/even page
Date: 12 Nov 2002 07:18:33 +0100

> I must admit fail to see the reasoning behind using a separate "save"
> register for keeping the "global" page offset instead of using PO
> directly. 
> 
> With alternate PO's set for odd/even pages, using multi-column
> multi-page output, the net effect is that the > 1st column is not
> positioned correctly.
> 
> Perhaps it would be an idea to add an improved odd/even support in -ms.
> It is just a matter of a couple of extra lines.
> 
> Patch follows: [...]

I've now applied your change to the CVS (see below for reference).
Please test!


    Werner


======================================================================


--- s.tmac.old  Sun Jul 25 10:56:09 2004
+++ s.tmac      Wed Jul 28 17:20:45 2004
@@ -505,7 +505,6 @@
 .nr PN \\n%
 .nr pg*col-num 0
 .nr address@hidden 0
-.nr pg*saved-po \\n[PO]
 .po \\n[PO]u
 .ev h
 address@hidden
@@ -542,12 +541,12 @@
 ..
 .de pg*end-col
 'sp |\\n[pg*col-top]u
-.po (u;\\n[pg*saved-po]+(address@hidden))
+.po (u;\\n[PO]+(address@hidden))
 .\"po +(u;address@hidden)
 .pg*start-col
 ..
 .de pg*end-page
-.po \\n[pg*saved-po]u
+.po \\n[PO]u
 .\" Make sure we don't exit if there are still floats or footnotes left-over.
 .ie address@hidden<address@hidden:address@hidden \{\
 .      \" Switching environments ensures that we don't get an unnecessary