The following specimen was run on live.contextgarden.net with each of the three typesetters: \setupcolors[state=start] \starttext \input knuth \bTABLE \bTR [align=middle]\bTH Range\eTH{}\bTH Value\eTH{}\eTR \bTR \bTD \type{<} 12\eTD{}\bTD 3\eTD{}\eTR \bTR \bTD 12--16\eTD{}\bTD 2\eTD{}\eTR \bTR \bTD \type{>}16\eTD{}\bTD 1\eTD{}\eTR \eTABLE \input knuth \stoptext In the case of pdfTeX and XeTeX, the table is given a width equal to the textwidth; in the caes of LuaTeX, it is typeset correctly with natural widths. When \setupcolors is removed, both pdfTeX and XeTeX also produce the correct result. This sounds like a bug to me (and took me forever to isolate \setupcolors - of all things! - as the culprit).
Michael.Urban@jpl.nasa.gov wrote:
The following specimen was run on live.contextgarden.net with each of the three typesetters:
\setupcolors[state=start] \starttext \input knuth
\bTABLE \bTR [align=middle]\bTH Range\eTH{}\bTH Value\eTH{}\eTR \bTR \bTD \type{<} 12\eTD{}\bTD 3\eTD{}\eTR \bTR \bTD 12--16\eTD{}\bTD 2\eTD{}\eTR \bTR \bTD \type{>}16\eTD{}\bTD 1\eTD{}\eTR \eTABLE
\input knuth \stoptext
In the case of pdfTeX and XeTeX, the table is given a width equal to the textwidth; in the caes of LuaTeX, it is typeset correctly with natural widths.
When \setupcolors is removed, both pdfTeX and XeTeX also produce the correct result.
This sounds like a bug to me (and took me forever to isolate \setupcolors - of all things! - as the culprit).
actually it's the \type (combined with colors) \dontleavehmode \type {xx} works ok probably some pdfliteral is playing havoc here so that the width analyer cannot function well (a limitation of tex is that there's only a small set subset of \un* and \last* node operations); if you see messages like this !!forced exit from shapebox 9!! that is indeed the case. hoewver, in mkiv i use different trickery so there we don't have this problem (we also use a different color mechanism so that helps too) it's not so much a bug in mkii as well as a side effect/limitation okay, we could add \dontleavehmode to \type but i'm not sure if that will not break other things; also, bTABLE is kind of tricky because it needs to be clever with regards to text and non text and such Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Fri, 23 May 2008 22:29:06 +0200
Hans Hagen
Michael.Urban@jpl.nasa.gov wrote:
The following specimen was run on live.contextgarden.net with each of the three typesetters:
\setupcolors[state=start] \starttext \input knuth
\bTABLE \bTR [align=middle]\bTH Range\eTH{}\bTH Value\eTH{}\eTR \bTR \bTD \type{<} 12\eTD{}\bTD 3\eTD{}\eTR \bTR \bTD 12--16\eTD{}\bTD 2\eTD{}\eTR \bTR \bTD \type{>}16\eTD{}\bTD 1\eTD{}\eTR \eTABLE
\input knuth \stoptext
In the case of pdfTeX and XeTeX, the table is given a width equal to the textwidth; in the caes of LuaTeX, it is typeset correctly with natural widths.
When \setupcolors is removed, both pdfTeX and XeTeX also produce the correct result.
This sounds like a bug to me (and took me forever to isolate \setupcolors - of all things! - as the culprit).
actually it's the \type (combined with colors)
\dontleavehmode \type {xx}
works ok
probably some pdfliteral is playing havoc here so that the width analyer cannot function well (a limitation of tex is that there's only a small set subset of \un* and \last* node operations); if you see messages like this
!!forced exit from shapebox 9!!
that is indeed the case. hoewver, in mkiv i use different trickery so there we don't have this problem (we also use a different color mechanism so that helps too)
it's not so much a bug in mkii as well as a side effect/limitation
okay, we could add \dontleavehmode to \type but i'm not sure if that will not break other things; also, bTABLE is kind of tricky because it needs to be clever with regards to text and non text and such
If he use \retype the table works. Wolfgang
participants (3)
-
Hans Hagen
-
Michael.Urban@jpl.nasa.gov
-
Wolfgang Schuster