- dev-context - NTG Mailing Lists

xtag-mmp.tex patch
by Duncan Hothersall 24 Mar '06

24 Mar '06

At line 492 in xtag-mmp.tex: \def\pMULTISCRIPTmmlPRE#1#2#3#4% {\ifdone \ifodd\scratchcounter ^{\naturalxmlr{#1}{#2}{#3}{#4}}{}% \else _{\naturalxmlr{#1}{#2}{#3}{#4}}% \fi \else \doif{mprescripts/}{#1}{\donetrue{}}% weak, the / here \fi \ifdone \advance\scratchcounter 1 \fi} I think the super/sub scripts are the wrong way around; suggest swapping fourth and sixth lines: \def\pMULTISCRIPTmmlPRE#1#2#3#4% {\ifdone \ifodd\scratchcounter _{\naturalxmlr{#1}{#2}{#3}{#4}}% \else ^{\naturalxmlr{#1}{#2}{#3}{#4}}{}% \fi \else \doif{mprescripts/}{#1}{\donetrue{}}% weak, the / here \fi \ifdone \advance\scratchcounter 1 \fi} According to the MathML spec http://www.w3.org/TR/2003/REC-MathML2-20031021/chapter3.html#presm.mmultisc… the syntax is <mmultiscripts> base (subscript superscript)* [ <mprescripts/> (presubscript presuperscript)* ] </mmultiscripts> Thanks from the strange world of presentational MathML, Duncan

1 0

Re: [dev-context] Cron <context@server> /var/www/aanhet.net/context/bin/cron/context-mirror
by Taco Hoekwater 22 Mar '06

22 Mar '06

Cron Daemon wrote: > receiving file list ... done > ./ > context/current/ > context/current/cont-tmf.zip Hi Hans, I've created a wiki page for this, but there is almost no news, right? http://wiki.contextgarden.net/Context_2006.03.22 Cheers, Taco

1 0

Re: [dev-context] [NTG-context] Curious blank page
by Taco Hoekwater 20 Mar '06

20 Mar '06

Hans Hagen wrote: >> > > can you check the version i uploaded a moment ago? > Hans, can you check the summary I wrote here: http://wiki.contextgarden.net/Release_Notes I am not sure I've understood everything correctly. Cheers, Taco

2 1

Re: [dev-context] kpse
by Hans Hagen 15 Mar '06

15 Mar '06

Taco Hoekwater wrote: > > Hi Hans, > > Somehow, it does not find my texmf.cnf. > > I get: > > [taco@glenlivet ruby]$ ruby tmftools.rb texnansi.enc > ./base/kpsefast.rb:650:in `basename': cannot convert nil into String > (TypeError) > from ./base/kpsefast.rb:650:in `_is_cnf_?' > from ./base/kpsefast.rb:654:in `find_file' > from ./tmftools.rb:111:in `main' > from ./tmftools.rb:110:in `each' > from ./tmftools.rb:110:in `main' > from ./tmftools.rb:168:in `send' > from ./tmftools.rb:168 > > I what environment settings do I need? I now have: > > TEXMFCNF=/opt/tex/texmf.local/web2c:/opt/tex/teTeX/texmf/web2c > TEXMFLOCAL=/opt/tex/texmf.local > TEXMFMAIN=/opt/tex/teTeX/texmf > > and > > [taco@glenlivet ruby]$ kpsewhich texmf.cnf > /opt/tex/texmf.local/web2c/texmf.cnf a better definition is: def _is_cnf_?(filename) filename == File.basename(((a)cnffiles.first rescue 'texmf.cnf')) end (is in files i sent you in aseparate mail) 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 -----------------------------------------------------------------

2 1

kpse
by Hans Hagen 14 Mar '06

14 Mar '06

Hi, Attached is a zip with ruby stuff. Not perfect, also not best organized but it shows some ideas. compare kpsewhich texnansi.enc tmftools.rb texnansi-enc the first alternative is a bit faster, the second can be faster when we use a serialized hash. now, say (windows) set KPSEMETHOD=drb set KPSEPORT=7000 and run tmftools --serve and next tmftools.rb texnansi-enc since we now run kpse as a service, we have a pretty fast response. One may wonder why we need such things ... but imagine (as part of a workflow): texmfstart bin:xsltproc --output whatever.xml kpse:something.xsl whoknows.xml texmfstart rlxtools --manipulate kpse:test.rlx whatever.rlg we can now save calls to kpsewhich. texmfstart already has many built in optimizations to limit the number of calls and kpsefast also has some caching. Btw, the server serves multiple trees at the same time. (I have to reshuffle the kpse classes/modules a bit.) Other new features involve passing environment variables. In threaded scripts the problem is that the environment hash is shared. This means that one cannot easily do something ENV['TEXMF']= = "..." system("texexec ...") in treads because each thread may have different assignments to TEXMF. For this purpose, texmfstart already had: --tree=/mytex/version-1 with setuptex.tmf is part of the minimal distributions and and is just a series of assignments. This file is put in the root of a tree. (btw, this makes it easy to put several trees on the wiki and to select a version; it's juts a matter if adding the --tree switch) # usage: texmfstart --tree=f:/minimal/tex ... # # this assumes that calling script sets TEXPATH without a trailing # slash; %VARNAME% expands to the environment variable, $VARNAME # is left untouched; we also assume that TEXOS is set. TEXMFMAIN = %TEXPATH%/texmf TEXMFLOCAL = %TEXPATH%/texmf-local TEXMFFONTS = %TEXPATH%/texmf-fonts TEXMFEXTRA = %TEXPATH%/texmf-extra TEXMFPROJECT = %TEXPATH%/texmf-project VARTEXMF = %TMP%/texmf-var HOMETEXMF = TEXMFOS = %TEXPATH%/%TEXOS% # OSFONTDIR = %SYSTEMROOT%/fonts TEXMFCNF = %TEXPATH%/texmf{-local,}/web2c TEXMF = {$TEXMFOS,$TEXMFPROJECT,$TEXMFFONTS,$TEXMFLOCAL,$TEXMFEXTRA,!!$TEXMFMAIN} TEXMFDBS = $TEXMF TEXFORMATS = %TEXMFOS%/web2c/{$engine,} MPMEMS = %TEXFORMATS% TEXPOOL = %TEXFORMATS% MPPOOL = %TEXPOOL% PATH > %TEXMFOS%/bin PATH > %TEXMFLOCAL%/scripts/perl/context PATH > %TEXMFLOCAL%/scripts/ruby/context EXAMPLEROOT = %VARTEXMF% TEXINPUTS = MPINPUTS = MFINPUTS = In addition to this file, one can also feed an extra file with the --environment option. This means that one can leave most dirty details to texmfstart and keep scripts simple. 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 -----------------------------------------------------------------

1 0

Re: [dev-context] stijlbijdrage
by Hans Hagen 14 Mar '06

14 Mar '06

Hoi, > > Ik heb de cdcover-code van Dennis van Dok (uit de laatste NTG MAPS) > bewerkt voor ConTeXt en er nog wat toeters en bellen aan toegevoegd. > Het lijkt erop dat de eerste kinderziekten er nu uit zijn (fingers > crossed). Of dat echt zo is blijkt waarschijnlijk pas als anderen het > gebruiken. > Mijn vraag: is er interesse om dit op te nemen in het thirdparty > aanbod op de Pragma site? > Zo ja dan kan ik je het bestand toezenden. Jij en Taco kunnen er dan > eerst naar kijken om te zien of je het wat vindt. > Dennis van Dok was accoord met verdere verspreiding van zijn geesteskind. third party modules are now coordinated via the wiki; i still need to set up a way to fetch the styles etc from the wiki so that we can have an additional zip; the website is generated from xml files, so i need to add entries for tp modules there (maybe we need a kind of xml description file for that as well: name, author, description, etc) so, best is to create an entry in the wiki on the third party page and then announce it; after that people from the dev team will pick it up and check the file for (for instance) low level dependencies and such the 'latest versions' are stored in the wiki and from there submitted to ctan (in a next stage we can consider to also have a third party svn branch) 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 -----------------------------------------------------------------

4 3

yarv
by Hans Hagen 03 Mar '06

03 Mar '06

Hi, Has someone tried this? http://www.atdot.net/yarv/ Ruby compiling 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 -----------------------------------------------------------------

3 5

\dodoregister bug
by Duncan Hothersall 22 Feb '06

22 Feb '06

With latest ConTeXt: -- \starttext This is a paragraph of text with an index term at the very end of it. This is a paragraph of text with an index term at the very end of it. \index{test} This is another paragraph. This is another paragraph. This is another paragraph. \stoptext -- These two paras run together. If I remove the recent fix (which I requested - sorry Hans!) to \dodoregister in core-reg: \def\dodoregister[#1]#2#3% {\doprocesspageregister[#1]{#2}{#3}} % intead of %% \def\dodoregister[#1]#2#3% %% {\doflushatpar{\doprocesspageregister[#1]{#2}{#3}}} This makes things work correctly again, but of course it also removes the nice fix which stopped \index calls at the start of paras causing additional white-space. I suggest that this is reverted though, because this is probably a more serious and common problem than the one we were trying to fix. Duncan

2 1

bfm math bug?
by Adam Lindsay 18 Feb '06

18 Feb '06

compare the following formulae with your favourite working bold math setup (confirmed with CM, Iwona, Lucida here): \startformula % this works \sin{\frac{{\bfm \pi}}{2}}=\sqrt{\bfm2} \stopformula \startformula % this doesn't \sin{\frac{{\bfm \pi}}{2}}=\sqrt{2} \stopformula % \startformula % this crashes % \sin{\frac{\bfm \pi}{2}}=\sqrt{2} % \stopformula I can't claim to know about all the family magic that goes on behind the scenes with bold math, but I'm fairly confident I can configure things properly. Smells like a bug. In other news, these will make bold numbers, if you need them: \definemathcharacter [48] [nothing] [mr] ["30] \definemathcharacter [49] [nothing] [mr] ["31] \definemathcharacter [50] [nothing] [mr] ["32] \definemathcharacter [51] [nothing] [mr] ["33] \definemathcharacter [52] [nothing] [mr] ["34] \definemathcharacter [53] [nothing] [mr] ["35] \definemathcharacter [54] [nothing] [mr] ["36] \definemathcharacter [55] [nothing] [mr] ["37] \definemathcharacter [56] [nothing] [mr] ["38] \definemathcharacter [57] [nothing] [mr] ["39] Again, checked on my limited test cases and knowledge of mathematics. I'm sure there will be unanticipated side effects. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl(a)comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

2 5

[Fwd: Re: [NTG-context] gnuplot mode]
by Taco Hoekwater 16 Feb '06

16 Feb '06

Hi Mojca, Can you tell/send me what I need to get (and where) to compile a GNUplot from source with your terminal in it (on linux)? Cheers, taco -------- Original Message -------- And a final word: I finally installed "something to compile gnuplot with" yesterday. I have the first pre-alpha ;) "set terminal context" working and I will need both Hans and Taco's hand as well as gnuplot/metafun users to help me with suggestions/requests/brainstorming about how to implement certain parts, so that the output will remain highly configurable regarding fonts, colors & styles, but more about that in another mail.

2 1