From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from scc-mailout.scc.kit.edu (scc-mailout.scc.kit.edu [129.13.185.202])
	by krisdoz.my.domain (8.14.3/8.14.3) with ESMTP id p9C0IEWo020809
	for <discuss@mdocml.bsd.lv>; Tue, 11 Oct 2011 20:18:15 -0400 (EDT)
Received: from hekate.usta.de (asta-nat.asta.uni-karlsruhe.de [172.22.63.82])
	by scc-mailout-02 with esmtp (Exim 4.72 #1)
	id 1RDmWS-0006dz-Gr; Wed, 12 Oct 2011 02:18:12 +0200
Received: from donnerwolke.usta.de ([172.24.96.3])
	by hekate.usta.de with esmtp (Exim 4.72)
	(envelope-from <schwarze@usta.de>)
	id 1RDmWS-0004bK-Fe; Wed, 12 Oct 2011 02:18:12 +0200
Received: from iris.usta.de ([172.24.96.5] helo=usta.de)
	by donnerwolke.usta.de with esmtp (Exim 4.72)
	(envelope-from <schwarze@usta.de>)
	id 1RDmWS-000558-Eg; Wed, 12 Oct 2011 02:18:12 +0200
Received: from schwarze by usta.de with local (Exim 4.72)
	(envelope-from <schwarze@usta.de>)
	id 1RDmWS-0008L0-Dl; Wed, 12 Oct 2011 02:18:12 +0200
Date: Wed, 12 Oct 2011 02:18:12 +0200
From: Ingo Schwarze <schwarze@usta.de>
To: Thomas Klausner <wiz@netbsd.org>
Cc: discuss@mdocml.bsd.lv
Subject: Re: mandoc and nroff differs for Em macro construct in columnated
 list
Message-ID: <20111012001812.GA10686@iris.usta.de>
References: <20111011222000.GK25396@danbala.tuwien.ac.at>
X-Mailinglist: mdocml-discuss
Reply-To: discuss@mdocml.bsd.lv
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20111011222000.GK25396@danbala.tuwien.ac.at>
User-Agent: Mutt/1.5.21 (2010-09-15)

Hi Thomas,

Thomas Klausner wrote on Wed, Oct 12, 2011 at 12:20:00AM +0200:

> In a columnated list, use of Em macro differs with a quoted string of
> words separated by tab characters. While nroff do emphasis all words,
> mandoc do only make the first one.

Yes, this is a known bug in mandoc.
It is listed in the TODO list here:

- In .Bl -column,
  .It Em Authentication<tab>Key Length
  ought to render "Key Length" with emphasis, too,
  see OpenBSD iked.conf(5).

Actually, it's one of the nastier bugs we have and not at all
easy to fix.  It's a bad nesting issue:

  begin .It
    begin .Ta
      begin .Em
    end .Ta
    begin .Ta
    end .Ta
      end .Em
  end .It

Two blocks (.Ta and .Em) overlap, but none of them fully contains
the other.  On top of that, .Ta requires different handling
depending on whether it's written "Ta" or <tab>.  And on top
of that, the quoted string also spans blocks.

I have implemented bad nesting support for .Xo some time ago,
but this case it yet more complicated since .Bl -column is by
far the most complicated mdoc block even without such additional
complications.

I hope we will come round to fix that one day, but it's not
going to be quick or easy.

Yours,
  Ingo
--
 To unsubscribe send an email to discuss+unsubscribe@mdocml.bsd.lv