Hi David, I have no time right now to write the release notes, I will do that in the morning (mine, that is). I'll keep your questions in mind. Cheers, Taco David Arnold wrote:

Hans et al,

I don't understand the word "stub."

What are ctxtools?

What is meant by an update feature in Scite? Where are context features in scite discussed? For example, Hans let me knoow about Ctrl +F12 for compiling my figure library, but that's not on any of the menus of scite. What else is cool but hidden?

Thomas A. Schmitz wrote:

Hi Hans,

just downloaded the new release and regenerated the formats. Will play with it later. For the time being just one question: the stubs in scripts/context/stubs/unix and .../mswin seem to be identical; is that what you meant? I doubt that many unixoid systems will be at ease with .bat executables and c: paths...

hm, i will check that 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 -----------------------------------------------------------------

Hans Hagen wrote:

texmfstart --verbose --unix --stubpath=<yourbinpath> --make all

That is a cool feature. But it does not set executable bits on the stubs yet, could you fix that? Regards, Christopher

Thomas A.Schmitz wrote:

On May 8, 2006, at 1:04 PM, Hans Hagen wrote:

...

Hi,

Today i'll upload a new release. Highlights:

Hans,

another (big) problem with the new release. Unfortunately, I can't really say what breaks, but when I compile a file under the new release, I get hundreds of overfull boxes. I get hyphenation, but TeX doesn't seem to find the same break points as before. The same file compiles wonderfully with version 2006.04.27. Can I send a logfile or something?

Hope this message makes it to the list...

btw, there was a problem with the german patterns for a while; let me check it 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 -----------------------------------------------------------------

Hans, This ruby stuff has me confused to no end. I'm compiling this document (which compiled just fine with the last release - no changes since) with the following command: ruby ~/texmf/scripts/context/ruby/texexec.rb PhD-Thesis.tex For trying to figure out the version I do: ruby ~/texmf/scripts/context/ruby/texexec.rb --check And get TeXExec | current distribution: web2c TeXExec | context source date: unknown TeXExec | format path: TeXExec | start of analysis /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart texexec --help /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart texutil --help /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart ctxtools --help TeXExec | end of analysis TeXExec | TeXExec | TeXExec | TeXExec | Which is mostly related to the UNIX stubs not working no? How do I now actually figure out what I'm running and whether it all got rebuild correctly? Thanks, Joh Hans Hagen wrote:

Johannes Graumann wrote:

...

Trouble with tex/context/base/cont-err.tex:

system : cont-err loaded (/home/balin/texmf/tex/context/base/cont-err.tex Runaway argument? on Otten}] %C %C This module is part of the \CONTEXT\ macro||package \ETC. ! Paragraph ended before & was complete. <to be read again> \par l.13

Can't find anything wrong though ...

just to be sure, i use version 2006.05.09 23:03

i looks like you're ignoring comments

how does the tex file look that you process?

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 -----------------------------------------------------------------

Hans, I linked texmfstart.rb to texmfstart in my local bin directory (which is in the path) and issued 'chmod -R +x ~/texmf/scripts'. If I call 'ruby ~/texmf/scripts/context/ruby/texexec.rb --check' now, I get this error free (but not informative) output: TeXExec | current distribution: web2c TeXExec | context source date: unknown TeXExec | format path: TeXExec | start of analysis TeXExec | end of analysis TeXExec | TeXExec | TeXExec | TeXExec | Any ideas what I might be doing wrong? Joh Hans Hagen wrote:

Johannes Graumann wrote:

...

Hans,

This ruby stuff has me confused to no end. I'm compiling this document (which compiled just fine with the last release - no changes since) with the following command:

ruby ~/texmf/scripts/context/ruby/texexec.rb PhD-Thesis.tex

For trying to figure out the version I do:

ruby ~/texmf/scripts/context/ruby/texexec.rb --check

And get

TeXExec | current distribution: web2c TeXExec | context source date: unknown TeXExec | format path: TeXExec | start of analysis /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart texexec --help /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart texutil --help /home/balin/texmf/scripts/context/ruby/base/kpsefast.rb:659: command not found: texmfstart ctxtools --help TeXExec | end of analysis TeXExec | TeXExec | TeXExec | TeXExec |

Which is mostly related to the UNIX stubs not working no?

cp texmfstart.rb <samepathaspdftex>/texmfstart chmod 755 <samepathaspdftex>/texmfstart

...

How do I now actually figure out what I'm running and whether it all got rebuild correctly?

so, no stub for texmfstart, just a copy (althoughj you may want to call it indirectly

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 -----------------------------------------------------------------

Hans, do I need to re-report that format generation with texmfstart newtexexec --make is still broken on OS X (I'll check on my linux box later)? I still get these monstrous paths like TeXExec | tex engine path: .:/Users/tas/texmf/web2c//:!!/usr/local/ teTeX/share//share/texmf-local/web2c//:!!/usr/local/teTeX/share// share/texmf/web2c//:!!/usr/local/teTeX/share//share/texmf-dist/web2c// pdfetex Best Thomas

Yes, that's strange. I just checked, and everything is golden on a linux system (gentoo, but I gues the same is true for other linux distros). So I really don't see why OS X should behave differently. Is there any debugging info that I could provide? How 'bout the other OS X users (I seem to recall there were quite a few of them): do you get the same error? Best Thomas On Wed, 2006-05-10 at 19:10 +0200, Hans Hagen wrote:

strange, since i suppose that linux and osx behave the same (don't call it newtexexec any more)

Hans

Am Donnerstag, 11. Mai 2006 00:21 schrieb Hans Hagen:

(concerning the path ... what is the path separator on osx?)

The normal unix path separator '/'. On OS9 it was ':'. Uwe

Thomas A. Schmitz wrote:

On May 11, 2006, at 12:21 AM, Hans Hagen wrote:

...

i'm uploading a coupls of zips with fixes for xetex but i'm not sure if they solve your problem

(concerning the path ... what is the path separator on osx?)

Hans

Nope, still get the same bizarre result. Path separator is a ":"

btw, concerning the texmfstart script itself: - on unix it's indeed best to make a symlink from texmfstart -> texmfstart.rb in the whatever-tree it resides in - on windows, one can associate the .rb suffix with ruby and add the context ruby path to the bin path, or use the 'exe' variant which can be located anywhere (texmfstart uses a few libe but when they are not found, it will enter stupid-mode and lack some features that probably go unnoticed) 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 -----------------------------------------------------------------

nico wrote:

On Thu, 11 May 2006 00:11:47 +0200, Thomas A. Schmitz wrote:

...

Yes, that's strange. I just checked, and everything is golden on a linux system (gentoo, but I gues the same is true for other linux distros). So I really don't see why OS X should behave differently. Is there any debugging info that I could provide? How 'bout the other OS X users (I seem to recall there were quite a few of them): do you get the same error?

Strictly speaking it's not OSX, but i've a FreeBSD distro, which is close, and it works fine.

Are you sure you have the permission to write in any of these paths?

BTW, i find dangerous to do (kpse.formatpath method):

# locate writable path if ! formatpath.empty? then formatpath.split_path.each do |fp| fp.gsub!(/\\/,'/') # remove funny patterns fp.sub!(/^!!/,'') fp.sub!(/\/+$/,'') fp.sub!(/unsetengine/,if enginepath then engine else '' end) if ! fp.empty? && (fp != '.') then # strip (possible engine) and test for writeability fpp = fp.sub(/#{engine}\/*$/,'') if FileTest.directory?(fpp) && FileTest.writable?(fpp) then # use this path formatpath = fp.dup break end end end end # needed ! begin File.makedirs(formatpath) ; rescue ; end ;

If none of the paths from formatpath is valid, we keep the whole concat path in this variable... that then is used to create a directory tree. Thomas, maybe that you have ugly directories created starting with an hidden directory named ".:". You should check that.

i can make the fallback '.' in that case, not that it helps much if ! formatpath.empty? then done = false formatpath.split_path.each do |fp| fp.gsub!(/\\/,'/') # remove funny patterns fp.sub!(/^!!/,'') fp.sub!(/\/+$/,'') fp.sub!(/unsetengine/,if enginepath then engine else '' end) if ! fp.empty? && (fp != '.') then # strip (possible engine) and test for writeability fpp = fp.sub(/#{engine}\/*$/,'') if FileTest.directory?(fpp) && FileTest.writable?(fpp) then # use this path formatpath, done = fp.dup, true break end end end formatpath = '.' unless done end 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 -----------------------------------------------------------------

Taco Hoekwater wrote:

This is the very top of cont-err.tex, the 'module header'. If that does not look broken, then probably the file that loads it does so while some really strange catcode settings are in effect. The most likely cause is that you have an invalid zip file or an error has occured during extraction. I went back and looked at the file - nothing obvious. I repeated my install from the pragma server ... no changes.

The document in question compiled just fine before the update.

If you cannot figure it out, try posting a few extra lines from the log, just before 'cont-err loaded'. I append the log region around the 'cont-err loaded' part.

Thanks for any help, Joh chapter : - About this Document publications : warning: cite argument knuth:86 is unknown on 80 publications : warning: cite argument hagen:01 is unknown on 80 publications : warning: cite argument lamport:94 is unknown on 80 publications : warning: cite argument thanh:98 is unknown on 80 publications : warning: cite argument knuth:84 is unknown on 80 publications : warning: cite argument adobe:04 is unknown on 80 Overfull \hbox (4.7209pt too wide) in paragraph at lines 4--4 []\*modern12ptrmtf*: This doc-u-ment was type-set in Com-puter Mod-ern (Xxxxxxx xxx, 0000), us-ing ConT[]Xt \hbox(8.82599+3.0)x438.77812, glue set - 1.0 \kern-0.432 (left margin) \hbox(0.0+0.0)x0.0 \*modern12ptrmtf*: T \*modern12ptrmtf*: h \*modern12ptrmtf*: i etc. publications : warning: cite argument graumann:04 is unknown on 80 publications : warning: cite argument yergeau:04 is unknown on 80 section : - References ) structure : end of sectionblock backmatter [175.162] systems : end file PhD-Thesis at line 77 system : cont-err loaded (/home/balin/texmf/tex/context/base/cont-err.tex Runaway argument? on Otten}] %C %C This module is part of the \CONTEXT\ macro||package \ETC. ! Paragraph ended before & was complete. <to be read again> \par l.13 ? s OK, entering \scrollmode... Overfull \hbox (42.98929pt too wide) in paragraph at lines 1--13 \*modern12ptrmtf*: Sys-tem Files, %D sub-ti-tle=Just A warn-ing, %D au-thor=Han s Ha-gen, %D date=\currentdate, \hbox(8.82599+3.0)x438.77812, glue set - 1.0 \*modern12ptrmtf*: S \*modern12ptrmtf*: y \*modern12ptrmtf*: s \discretionary .\*modern12ptrmtf*: - \*modern12ptrmtf*: t etc. ) Runaway argument? } \writeline \par \endinput ! File ended while scanning use of \xparseXMLelement. <inserted text> \par <to be read again> \relax \dodoreadfile ...\preprocesssuffix \or \fi \relax \the \everyafterreadfile \loadallsystemfiles ...ssage \m!systems 2{#1}}{#2} \else \def \doloadsystemfi... \finalend ...se \the \everybye \the \everygoodbye \global \everybye \emptyto... l.77 </document>

Taco Hoekwater wrote:

Johannes Graumann wrote:

...

...

If you cannot figure it out, try posting a few extra lines from the log, just before 'cont-err loaded'.

I append the log region around the 'cont-err loaded' part.

It looks to me like context found a cont-err.tex when it should not have.

At the end of a job, ConTeXt looks for a file named cont-err.tex, in the directory path that is set by the \setupsystem command in <jobname>.tmp. This file itself is generated by texexec, and the directory value is based on "setuppath").

In your case, it looks like this directory entry is either incorrect, or there is a bug in context, because it finds and then loads the file while XML mode is still active.

I do not know how to proceed from here. Hans?

in core-sys.tex, replace: \ifx\disableXML\undefined \let\disableXML\relax \fi \def\loadsystemfiles {\reportprotectionstate \readsysfile\f!newfilename {\showmessage\m!systems2\f!newfilename}% \donothing \readsysfile\f!oldfilename {\showmessage\m!systems2\f!oldfilename}% \donothing \loadallsystemfiles\f!filfilename \donothing \loadallsystemfiles\f!sysfilename {\loadallsystemfiles{\f!sysfilename.rme}\donothing % new, fall back \doglobal\appendtoks \bgroup \disableXML \loadallsystemfiles\f!errfilename\donothing \egroup \to\everygoodbye}} and let me know if that works ok 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 -----------------------------------------------------------------

Johannes Graumann wrote:

Hans Hagen wrote:

...

in core-sys.tex, replace:

\ifx\disableXML\undefined \let\disableXML\relax \fi

\def\loadsystemfiles {\reportprotectionstate \readsysfile\f!newfilename {\showmessage\m!systems2\f!newfilename}% \donothing \readsysfile\f!oldfilename {\showmessage\m!systems2\f!oldfilename}% \donothing \loadallsystemfiles\f!filfilename \donothing \loadallsystemfiles\f!sysfilename {\loadallsystemfiles{\f!sysfilename.rme}\donothing % new, fall back \doglobal\appendtoks \bgroup \disableXML \loadallsystemfiles\f!errfilename\donothing \egroup \to\everygoodbye}}

I have appended this to core-sys.tex (just before the final '\protect \endinput'; latest version of 11.5.) and used 'texexec.rb --make'. I get the same error - even for a document containing nothing but this:

\defineXMLenvironment [document] {\starttext} {\stoptext} \startXMLdata <document></document> \stopXMLdata

How do we proceed from here?

a minimal test file showing the problem; as taco said, maybe some interference with xml processing 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 -----------------------------------------------------------------

Hans Hagen wrote:

\defineXMLenvironment [document] {\starttext} {\stoptext}

\starttext

\startXMLdata <document></document> \stopXMLdata

\stoptext

You catch my folly as usually. Thanks. While I get the solution above to work, the alternative 'ruby ~/texmf/scripts/context/ruby/texexec.rb --pdf --environments=PhD-Thesis ../XML/main.xml' doesn't. Error is:

User file 'cont-sys.tex' not found, 'cont-sys.rme' has been used instead.

Hints? Joh

Johannes Graumann wrote:

Hans Hagen wrote:

...

\defineXMLenvironment [document] {\starttext} {\stoptext}

\starttext

\startXMLdata <document></document> \stopXMLdata

\stoptext

You catch my folly as usually. Thanks. While I get the solution above to work, the alternative 'ruby ~/texmf/scripts/context/ruby/texexec.rb --pdf --environments=PhD-Thesis ../XML/main.xml' doesn't. Error is:

...

User file 'cont-sys.tex' not found, 'cont-sys.rme' has been used instead.

has nothing to do with ruby, just copy cont-sys.rme to cont-sys.tex in tex/context/user; i ship the rme as example and fallback so that it does not overwrite you rlocal copy Hans

http://wiki.contextgarden.net/Main_Page http://www.pragma-ade.nl/show-man-1.htm nm a écrit :

Hi *,

I'm new w/ ConTeXt.. Where can i found publication, articles, whitepaper template for ConTeXt ?

thanks,

nm. _______________________________________________ ntg-context mailing list ntg-context@ntg.nl http://www.ntg.nl/mailman/listinfo/ntg-context