- dev-context - NTG Mailing Lists

encodings
by Hans Hagen 19 Aug '05

19 Aug '05

Hi, This encoding mess ... somehow i got the feeling that we should start shipping proper tfm files, i.e. texnansi ec qx char (adam's new one) ecx (mojcs's fixed ec) I can probably convince te polish font guys and we need a similar named set (generated of copied) for the abotu 15 free fonts on tex live. There is no point is keeping in tune with old stuff and latex-needs-it-this-way-so-we-cannot-fix-it situation. It's taking too much time. (Comparable with the patterns mess, i now just ship them). In the end we can set up a project that maintains these sets. 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 6

Garden server moving
by Patrick Gundlach 19 Aug '05

19 Aug '05

Hi, I'll move the garden server to a slightly faster and bigger machine. I have found out that a 5 gig disk is just not enough for all the fine ConTeXt thingies out there. I guess that I do the work on monday/tuesday. Just in case someone wants to store things on the machine (museum etc...), this can be done soon. Patrick (I'll write to the main list when I start moving) -- ConTeXt wiki and more: http://contextgarden.net

2 1

kpse
by Hans Hagen 16 Aug '05

16 Aug '05

Hi, The latest distribution has an experimental ruby kpse class in kpsefast.rb (expects updates -). As far as i can measure it's not slower than native kpse and once loaded in a ruby prog it's faster of than reinvoking kpse each time; it can (optionaly) dump the database in a home/temp path which halves loading time. I will probably use this mechanism in the coming texexec/texutil (maybe also in texmfstart). The associated variant of kpsewhich is 'tmftools' (no help yet). Apart from basic kpsewhich functionality it will harbor more tree tools. texmfstart tfmtools --analyze --root=/whatever/tex texmfstart tfmtools --analyze --root=/whatever/tex --trees=texmf,texmf-local can be very instructive on duplicate files on your machine and point you to problems. It has a --delete option which can be used as follows texmfstart tfmtools --analyze --root=/whatever/tex --delete "texmf-local/fonts/.*/lm" this will remove dups in texmf-local (to be sure, one needs to pass --force as well) 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

lua
by Hans Hagen 09 Aug '05

09 Aug '05

Hi Hartmut, think of this: rewrite the map file mess in lua -) pseudo coded: \lua { <open file> <while not eof> <read and interpret line> pdftex.mapline(...) <end> <close file> } 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 2

lua
by Hans Hagen 08 Aug '05

08 Aug '05

Hi, Recently i've started looking into Lua, a language meant for embedding in programs. It has a small footprint, and by reading themanukl i get the impression that it's a pretty well designed and clever language. So (after playing with lua and scite, which has it embedded) i started wondering about including lua in pdftex. Of course i'm dependent on others but i'm curious abotu wat members of this list think of it. It has quite a clever function model. Especially because it has a small footprint (some 100-150 k) it would add a lot to pdftex but at little cost. think of on the one hand: \dimen0=\lua{tex.todimen(tex.thedimen(0)+tex.thedimen(2)} kind of things, or \setbox2=\hbox to \lua{tex.todimen(somevar)} {\lua{tex.text("abcde")}} and alike. Of course the interface is to be determined, but a startingpoint can be access to box dimensions and registers and the ability to pipe text (strings) back into teh tex inpt stream. whatdoyouthink ... 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 6

Re: lua
by Thanh Han The 08 Aug '05

08 Aug '05

On Mon, Aug 08, 2005 at 06:53:38PM +0200, Hans Hagen wrote: > Hi, > > Recently i've started looking into Lua, a language meant for embedding > in programs. It has a small footprint, and by reading themanukl i get > the impression that it's a pretty well designed and clever language. So > (after playing with lua and scite, which has it embedded) i started > wondering about including lua in pdftex. Of course i'm dependent on > others but i'm curious abotu wat members of this list think of it. It > has quite a clever function model. Especially because it has a small > footprint (some 100-150 k) it would add a lot to pdftex but at little cost. > > think of on the one hand: > > \dimen0=\lua{tex.todimen(tex.thedimen(0)+tex.thedimen(2)} > > kind of things, or > > \setbox2=\hbox to \lua{tex.todimen(somevar)} > {\lua{tex.text("abcde")}} > > and alike. Of course the interface is to be determined, but a > startingpoint can be access to box dimensions and registers and the > ability to pipe text (strings) back into teh tex inpt stream. > > whatdoyouthink ... sounds interesting. I did a quick search on google to learn more on this, and I like more the idea. Lua seems to be quite suitable for this purpose: small, fast, simple syntax. Having it embedded certainly makes a lot of fun. Who will give it a try? :) Thanh

2 1

Re: Cron <tex@server> /var/www/aanhet.net/context/bin/cron/context-mirror
by Taco Hoekwater 03 Aug '05

03 Aug '05

Hi Hans, Cron Daemon wrote: > context/current/cont-tmf.zip This zip contains a whole series of extra map files in two new subdirs of map/pdftex/context: fonts/map/pdftex/context/new/*.map fonts/map/pdftex/context/old/psclean.map fonts/map/pdftex/context/old/texnansi-bh-lucida.map as well as: fonts/map/pdftex/context/conbinemap.rb I just want to verify that that is intentional. Taco

2 2

distro zips
by Taco Hoekwater 29 Jul '05

29 Jul '05

Hi Hans (and others), I've been playing with the linuxtex.zip today, and I've noticed some things that may or may not be interesting to you: * it would be better to use tar.bz2 archives (instead of zip). The archive would be quite a bit smaller, and the file permissions would remain unaltered (so that the exes are actually executable after extraction). * the binaries can be stripped to gain some 1.6 megabytes of compressed size (4 mb uncompressed): 51 files, 3091342 bytes uncompressed, 1473221 bytes compressed vs. 51 files, 7230683 bytes uncompressed, 3068044 bytes compressed * cm-super is still included, but should be obsolete now thanks to latin-modern, right? * likewise for the cs/pl/vnr metrics. * perhaps this is even true for the CM pfb files? * the executables that are included are a bit old (pdfetex 1.21a, mpost 0.641). * And of course the 64bit version has no executables at all. Does it make sense for me to (attempt to) clean this up, or would my changes get lost soon anyway? Perhaps me or someone else can even start maintaining these files for you (using a project on FP's SVN server)? Taco

2 2

[Fwd: [tex-fonts] FPL-Neu 0.9]
by Hans Hagen 26 Jul '05

26 Jul '05

1 0

exTeX and Fonts
by Mojca Miklavec 26 Jul '05

26 Jul '05

I was in the Bavarian "TeX Stammtisch" on Saturday and one of the exTeX developers was with us (Michael Niedermair, m.g.n AT gmx.de) and presented the project shortly. Perhaps some of you already saw it at EUROTeX. If they get it working the way they plan to, it can become quite interesting. I can't imagine how that complex work can be done and kept up-to-date, but for ConTeXt I would say the same before I saw it working. The Bible needed approximately 6 minutes to compile and write every single box to an XML file (700 MB). PDF output is in development. The reason I'm writing is: He was asking for some ideas about how to support fonts. As here are some font gurus (Adam, Hans, etc.), perhaps you can give him some tips or present your point of view/how it should be done properly since they don't want to make it the way it is implemented in the standard TeX. The second presentation was about the extentions to palatino fonts: http://home.vr-web.de/was/x/FPL/ (some letters redesignes, but mostly missing characters added, so that no virtual font is needed any more, some kerning data added, ...) They would be glad to get any feedback. Mojca

2 1