From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Lee Green <eric@badtux.org>
To: Jorge Godoy <godoy@conectiva.com>, "Edward C. Bailey" <ed@redhat.com>
Cc: docbook-tools-discuss@sourceware.cygnus.com, Camille Bégnis <camille@mandrakesoft.com>
Subject: Re: Request for Redhat-Centric info (For Ed Bailey)
Date: Wed, 27 Dec 2000 06:36:00 -0000
Message-id: <00070710374201.18849@ehome.inhouse>
References: <Pine.LNX.4.10.10007070859340.2374-100000@traveller.rlf.org> <lfbt09hoq1.fsf@pigdog.meridian.redhat.com> <m3puopyiq9.fsf@dagon.conectiva>
X-SW-Source: 2000/msg00262.html

On Fri, 07 Jul 2000, Jorge Godoy wrote:
> "Edward C. Bailey" <ed@redhat.com> writes:
> 
> > docbook-3.1-4
> > jade-1.2.1-9
> > jadetex-2.7-2
> > psgml-1.0.3-2
> > sgml-common-0.1-8
> > stylesheets-1.54.13rh-1

> We're using the same packages at Conectiva.
> But we are using free tools to generate printed books and guides. 

Any secrets to getting an index that doesn't have a bunch of "-999" entries?
Approximately 3/4ths of the entries in my index have the correct page number.
The rest say "-999". This despite the fact that 'jadetex' is not reporting any
unresolved references. 

Of course that doesn't mean anything. If I run Jadetex twice, vs. 4 times, my
table of contents is full of -999 page numbers even though it's not reporting
any unresolved references.

There's some boojums here, but I can't figure it out.

BTW, no matter how many times I run 'jadetex', I still end up with some -999
page numbers in my index... I once ran 'jadetex' a dozen times to see whether
that would help. Nope. Note that I use single-entry (<primary>) indexes, I
don't do double-level indexing (i.e., with <secondary>).... 

example:

man-in-the-middle attacks, 78,84,84
MD5, -999,-999,85,196,197,205
Message Authentication Checksum, -999,-999

Very frustrating. 

Here is the 'index.sgml' section for the above:

<indexentry>
  <primaryie>man-in-the-middle attacks,
    <ulink url="x1374.html#AEN1382" role="AEN1400">DH:</ulink>,
    <ulink url="x1510.html#AEN1627" role="AEN1636">Re-keying a client</ulink>,
    <ulink url="x1638.html" role="AEN1643">Known problems and limitations</ulink>
  </primaryie>
</indexentry>
 
<indexentry>
  <primaryie>MD5,
    <ulink url="x48.html" role="AEN59">Design Philosophy</ulink>,
    <ulink url="x1296.html#AEN1311" role="AEN1361">Standard Headers</ulink>,
    <ulink url="x1638.html" role="AEN1664">Known problems and limitations</ulink>,
    <ulink url="x3876.html#AEN3991" role="AEN3995">Implementation</ulink>,
    <ulink url="x3876.html#AEN4021" role="AEN4025">Implementation</ulink>,
    <ulink url="g4068.html" role="AEN4232">Glossary of Terms</ulink>
  </primaryie>
</indexentry>
 
<indexentry>
  <primaryie>Message Authentication Checksum,
    <ulink url="x2014.html#AEN2042" role="AEN2106">Agent configuration</ulink>,
    <ulink url="x3413.html#AEN3460" role="AEN3473">Packet Format</ulink>
  </primaryie>
</indexentry>                                                                                   

-- 
Eric Lee Green      There is No Conspiracy
eric@badtux.org     http://www.badtux.org