atom feed28 messages in org.oasis-open.lists.docbook-appsRe: DOCBOOK-APPS: jadetex 3.9
FromSent OnAttachments
Jan ScheffczykJul 2, 2001 12:07 am 
Ian CastleJul 2, 2001 6:44 am 
Richard SharpeJul 2, 2001 8:54 am 
Richard SharpeJul 2, 2001 7:13 pm 
Sebastian RahtzJul 3, 2001 4:15 am 
Richard SharpeJul 3, 2001 4:54 am 
Bob McIlvrideJul 4, 2001 8:55 am 
Norman WalshJul 5, 2001 10:53 am 
Norman WalshJul 5, 2001 10:53 am 
Bob McIlvrideJul 6, 2001 8:36 am 
Ian CastleJul 9, 2001 6:26 am 
Ian CastleJul 9, 2001 6:31 am 
Richard SharpeJul 9, 2001 6:49 am 
Ian CastleJul 9, 2001 10:14 am 
Sebastian RahtzJul 11, 2001 7:00 pm 
Joe CooperJul 11, 2001 11:03 pm 
Kevin DunnJul 12, 2001 2:14 pm 
Ian CastleJul 12, 2001 11:50 pm 
Joe CooperJul 13, 2001 12:22 am 
Richard SharpeJul 13, 2001 12:38 am 
Juan R. MigoyaJul 13, 2001 1:16 am 
Kevin DunnJul 13, 2001 1:46 am 
Joe CooperJul 13, 2001 2:01 am 
Ian CastleJul 13, 2001 3:11 am 
Juan R. MigoyaJul 13, 2001 3:50 am 
Sebastian RahtzJul 13, 2001 3:50 am 
Richard SharpeJul 13, 2001 5:49 am 
Juan R. MigoyaJul 14, 2001 3:44 am 
Subject:Re: DOCBOOK-APPS: jadetex 3.9
From:Kevin Dunn (kdu@hsc.edu)
Date:Jul 12, 2001 2:14:21 pm
List:org.oasis-open.lists.docbook-apps

---------- Original Message ---------------------------------- From: Joe Cooper <jo@swelltech.com> Date: Thu, 12 Jul 2001 01:10:05 -0500

Hi Sebastian,

Looks like there are still some issues. I've fetched and installed 3.9, and I'm still getting undefined references messages, and wierdness in the TOC. It's a new wierdness, so I guess you're poking the right spot, but it's still not quite right.

Beginning with TOC references to roman numbered pages beyond i (ii, iii, etc.) the entry looks like 'qqq ??' or 'iqqq ??' (for iii).

Actually now that I'm looking at it closer... Every page number that is two of the same number becomes 'qqq'. So 11, 22, 33, 44, etc. are all qqq ??. This only occurs in the TOC--the pages themselves have correct page numbers.

I see the same thing. To find out what's going on, I modified the following in jadetex.dtx:

\def\@Setref#1#2{% \ifx#1\relax \immediate\write\@mainaux{\string\pagelabel{#2}{qqq}}% \protect\G@refundefinedtrue \nfss@text{\reset@font\bfseries ??}% \@latex@warning{Reference `#1' `#2' on page \thepage \space undefined}% \else \if#1{qqq} \protect\G@refundefinedtrue \nfss@text{\reset@font\bfseries ??}% \@latex@warning{Reference a `#1' `#2' on page \thepage \space undefined}% \else \ifNestedLink{#1}% \else \hyper@@link[Page]{}{page.#1}{#1}% % \typeout{Made a Link at \the\inputlineno, to page.#1}% \fi \fi \fi }

That is, I had jadetex print `#1' on the warning line and I distinguished the first warning from the second by adding "a" after "Reference" for the second warning.

Upon processing my document, the log file contains:

LaTeX Warning: Reference a `118' `CHREDOX' on page iii undefined on input line LaTeX Warning: Reference a `55' `TBLPOT1' on page v undefined on input line 189 LaTeX Warning: Reference a `77' `TBLRXN1' on page v undefined on input line 200 LaTeX Warning: Reference a `11' `FIGFIRE4' on page v undefined on input line 27 LaTeX Warning: Reference a `11' `FIGFIRE5' on page v undefined on input line 27 LaTeX Warning: Reference a `111' `FIGLIME1' on page vi undefined on input line LaTeX Warning: Reference a `66' `EQNSTRING1' on page vi undefined on input line

etc.

That is, #1 has a numerical value, yet it still passes the test: if#1{qqq} when and only when #1 has the first two digits the same.

Interestingly, the page number "211" processes correctly, so it seems that the first two digits are the important ones.