- dev-context - NTG Mailing Lists

bib: triple dash in article title causes problems (minimal example)
by Sanjoy Mahajan 24 Nov '09

24 Nov '09

To continue my earlier message about "--" in bibtex titles: I replaced the "--" with a triple dash "---" but still had problems unless I used \enableregime[utf]. Even then I had problems if I was also using products and projects, where \enableregime[utf] seemed to have no effect for the bibliography. Attached is a minimal set of files as a .tgz. Unpack and do 'cd test'; then 'make'. The resulting PDF, at least on my system (TL 2009 Debian + context 2009.10.27) has the accent characters instead of the em-dash. A small-size png image of it is attached. -Sanjoy `Until lions have their historians, tales of the hunt shall always glorify the hunters.' --African Proverb

2 4

Mathdesign / MnSymbol
by Wolfgang Schuster 22 Nov '09

22 Nov '09

Hi, can the mathdesign [1] and mnsymbol [2] fonts be included in the minimals. [1] http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=mathdesign [2] http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=mnsymbol Wolfgang

2 3

bib: article title with -- comes out mangled
by Sanjoy Mahajan 22 Nov '09

22 Nov '09

This test file has a mangled article title where the "--" was located in the bibtex entry (a few accented characters show up). Commenting in the \enableregime removes the accented characters, but now there's a mix of different dashes almost joined together. % \enableregime[utf] \usemodule[bib] \setupbibtex[database=project] \starttext Look up the Shannon--Nyquist sampling theorem \cite[Jerri:1977]. \blank[3*big] \placepublications[criterium=all] \stoptext Here is the bibtex file @Article{Jerri:1977, author = {A. J. Jerri}, title = {The {S}hannon sampling theorem —- {I}ts various extensions and applications: A tutorial review}, journal = {Proceedings of the IEEE}, year = 1977, volume = 65, number = 11, pages = {1565--1596}} (A workaround is to make the " -- " into "---", which is what it should have been anyway.) This is with context 2009.10.27 16:35 MKII in the Debian TL2009 packages. I also see the problem with 2009.11.20 on top of vanilla TL2009 or vanilla TL2008. -Sanjoy

1 0

linebreak regression with TL2009 versus TL2008
by Sanjoy Mahajan 21 Nov '09

21 Nov '09

The following test file ============================================================= \usetypescript[palatino][ec] \setupbodyfont[palatino,10.5pt] \setuplayout[width=4.75in] \showframe \starttext knowledge. The diverse examples help separate the tool---the general principle---from the particular applications so that you can grasp and transfer the tool to problems of particular interest to you. \stoptext ============================================================= works fine with the a recent context (2009.11.21) on top of TL 2008. But the linebreak is terrible if I use the same context on top of TL2009. I cannot find the cause. >From the logfile: Overfull \hbox (25.72205pt too wide) in paragraph at lines 10--14 []\*palatino10.5ptmmnnrm*: knowledge. The di-verse ex-am-ples help sep-a-rate t he tool---the gen-eral principle--- \hbox(7.55998+2.96098)x343.28249, glue set - 1.0 .\hbox(0.0+0.0)x0.0 .\*palatino10.5ptmmnnrm*: k .\*palatino10.5ptmmnnrm*: n .\*palatino10.5ptmmnnrm*: o .\kern-0.2625 .etc. Below is the full log file. At first I thought the problem was in TeX Gyre versus the former Palatino, but both runs use TeX Gyre. The only other difference that I can see is the pdftex versions: TL2009: This is pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009) (format=cont-en 2009.11.20) 21 NOV 2009 15:18 Tl2008: This is pdfTeXk, Version 3.1415926-1.40.9 (Web2C 7.5.7) (format=cont-en 2009.6.12) 21 NOV 2009 15:16 Other than that, the runs with TL 2008 and TL 2009 produce exactly the same log file except for the overfull hbox message, and '2008' becoming '2009' in all the filenames, and write18 being enabled in one but not the other (I think with 2008). Any ideas whether this is a font problem, pdftex problem, or a context problem (probably unlikely given that both runs use the same context source files)? -Sanjoy This is pdfTeX, Version 3.1415926-1.40.10 (TeX Live 2009) (format=cont-en 2009.11.21) 21 NOV 2009 16:00 entering extended mode file:line:error style messages enabled. %&-line parsing enabled. (/home/sanjoy/texmf-ctx/tl2009-20091121/web2c/natural.tcx) **test.tex \emergencyend (./test.tex ConTeXt ver: 2009.11.21 11:45 MKII fmt: 2009.11.21 int: english/english system : cont-new loaded (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/cont-new.tex systems : beware: some patches loaded from cont-new.tex (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/cont-new.mkii)) system : cont-fil loaded (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/cont-fil.tex loading : ConTeXt File Synonyms ) system : cont-sys.rme loaded (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/user/cont-sys.rme (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-tmf.tex (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-tmf.mkii)) (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-siz.tex (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-siz.mkii)) (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-one.tex (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/type-one.mkii))) system : test.top loaded (./test.top ) bodyfont : 12pt rm is loaded language : patterns nl->texnansi:texnansi->1->2:2 nl->ec:ec->2->2:2 us-> ec:ec->3->2:3 de->texnansi:texnansi->4->3:3 de->ec:ec->5->3:3 da->ec:ec->6->2:2 sv->ec:ec->7->2:2 gb->ec:ec->8->2:2 fr->texnansi:texnansi->9->2:2 fr->ec:ec->1 0->2:2 es->ec:ec->11->2:2 ca->ec:ec->12->2:2 it->texnansi:texnansi->13->2:2 it- >ec:ec->14->2:2 la->ec:ec->15->2:2 pt->texnansi:texnansi->16->2:2 pt->ec:ec->17 ->2:2 ro->ec:ec->18->2:2 pl->pl0:pl0->19->2:2 pl->ec:ec->20->2:2 pl->qx:qx->21- >2:2 cs->il2:il2->22->2:2 cs->ec:ec->23->2:2 sk->il2:il2->24->2:2 sk->ec:ec->25 ->2:2 hr->ec:ec->26->2:2 sl->ec:ec->27->2:2 tr->ec:ec->28->2:2 agr->agr:agr->30 ->2:2 fi->ec:ec->31->2:2 hu->ec:ec->32->2:2 ru->t2a:t2a->34->2:2 uk->t2a:t2a->3 5->3:3 loaded specials : dvips loaded \openout3 = `test.tui'. \openout0 = `test-mpgraph.mp'. \openout0 = `mpgraph.mp'. systems : system commands are disabled language : language en is active specials : loading definition file tpd (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/spec-tpd.mkii specials : loading definition file fdf (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/spec-fdf.mkii)) (/home /sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/page-run.mkii loading : ConTeXt Page Macros / Runtime Macros ) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) (./test.tuo) publications : file test.bbl not found, waiting for bibtex (./test.tuo) systems : begin file test at line 8 Overfull \hbox (25.72205pt too wide) in paragraph at lines 10--14 []\*palatino10.5ptmmnnrm*: knowledge. The di-verse ex-am-ples help sep-a-rate t he tool---the gen-eral principle--- \hbox(7.55998+2.96098)x343.28249, glue set - 1.0 .\hbox(0.0+0.0)x0.0 .\*palatino10.5ptmmnnrm*: k .\*palatino10.5ptmmnnrm*: n .\*palatino10.5ptmmnnrm*: o .\kern-0.2625 .etc. fonts : resetting map file list {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-empty .map} fonts : using map file: original-base {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-base. map} fonts : using map file: original-ams-base {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-ams-b ase.map} fonts : using map file: original-public-lm {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-publi c-lm.map} fonts : using map file: lm-ec {/usr/local/texlive/2009/texmf-dist/fonts/map/dvips/lm/lm-ec.map} fonts : using map file: lm-math {/usr/local/texlive/2009/texmf-dist/fonts/map/dvips/lm/lm-math.map} fonts : using map file: lm-rm {/usr/local/texlive/2009/texmf-dist/fonts/map/dvips/lm/lm-rm.map} fonts : using map file: original-ams-euler {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-ams-e uler.map} fonts : using map file: qpl-ec {/usr/local/texlive/2009/texmf-dist/fonts/map/dvips/tex-gyre/qpl-ec.map} fonts : using map file: original-youngryu-px {/home/sanjoy/texmf-ctx/tl2009-20091121/fonts/map/pdftex/context/original-young ryu-px.map} (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/sort-def.m kii) (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/sort-lan.mkii) [1 .1] systems : end file test at line 15 system : cont-err loaded (/home/sanjoy/texmf-ctx/tl2009-20091121/tex/context/base/cont-err.tex systems : no file 'cont-sys.tex', using 'cont-sys.rme' instead ) ) Here is how much of TeX's memory you used: 2697 strings out of 456429 49119 string characters out of 2525280 4775366 words of memory out of 7192449 43186 multiletter control sequences out of 15000+200000 178380 words of font info for 99 fonts, out of 3000000 for 9000 396 hyphenation exceptions out of 8191 51i,17n,77p,259b,542s stack positions out of 5000i,500n,10000p,200000b,50000s {/usr/local/texlive/2009/texmf-dist/fonts/enc/dvips/tex-gyre/q-ec.enc}</usr/ local/texlive/2009/texmf-dist/fonts/type1/public/tex-gyre/qplr.pfb> pdfTeX warning: pdftex: no GlyphToUnicode entry has been inserted yet! Output written on test.pdf (1 page, 25868 bytes). PDF statistics: 13 PDF objects out of 1000 (max. 8388607) 8 compressed objects within 1 object stream 0 named destinations out of 1000 (max. 500000) 1 words of extra memory for PDF output out of 10000 (max. 10000000)

1 0

Luatex 0.45.0 announcement
by Taco Hoekwater 18 Nov '09

18 Nov '09

Hi, I have just uploaded the archives for a new luatex release, 0.45.0. New features: * Loading big OTF fonts via lua is now much faster than before, at the expense of a bit of precision loss in the calculated bounding boxes. * Two new token lists are provided: \pdfxformresources and \pdfxformattr, as an alternative to \pdfxform keywords. * \input and \openin now accept braced filenames. * The \endlinechar can now be set to any value between 0 and 127. * Luatex can now optionally use kpathsea to find lua require() files. * The new primitives \aligntab and \alignmark are aliases for the use of & and # in alignments. * \pdfrefximage [width <dimen>] [height <dimen>] [depth <dimen>] <imageref> overrules settings from \pdfximage for this \pdfrefximage call only * The print precision for small numbers in lua code (tostring) has been improved. * Akira Kakuto suppled a patch that restricts writing to files with names such as latex.bat etc. (on windows only) * (un-feature): from now on luatex now only handles 4 direction specifiers: TLT (latin), TRT (arabic), RTT (cjk), and LTL (mongolian). The other directions now generate an error. * (un-feature): xpdf's automatic lower- and uppercasing retry attempts for nonfound image pdf files has been removed. * The callbacks hpack_filter, vpack_filter and pre_output_filter pass on an extra string argument for the direction, when known (nil == use default). * The lua functions node.hpack(), node.vpack() and node.dimensions() now accept an optional direction argument (not given == use default). * \pdfrefxform [width <dimen>] [height <dimen>] [depth <dimen>] <formref> scales a form object; similar principle as with \pdfximage: depth alone doesn't scale, it shifts vertically. * A change to the relative ordering inside font name discovery for use by the backend: now it tries <f>.psname first, as that is much less likely to contain spaces. * For ttc fonts, now the used subfont name and its index id are printed to the terminal, and if the backend cannot find a ttc subfont, the run is aborted. Bug fixes: * Prevent crashes for lua-loaded fonts without a (tfm) name. * Remove spurious newlines in the output pdf that could be the result of using \pdfximage for an included pdf figure too early. * Full font embedding for wide otf and ttf fonts now really works (older versions only pretended that it did work). * node.dimensions() sometimes continued past the 'stop' node. * Images inside virtual fonts were broken. * The processing of \language=-1 was wrong, resulting in hyphenation attempts even though there was no proper language set up. * Fix for finding files in --output-directory. * Fix for handling of --file-line-error. * tex.box[] assignments are now checked to make sure the assigned node has the right id (hlist or vlist). * show_node_list (as in \showbox) produced spurious local_par whatsit prints in the output after direction nodes. * The handling of pdf literals has been sped up. The archives can be downloaded from supelec as usual: http://foundry.supelec.fr/gf/project/luatex/ You could also check out the sources via anonymous svn: svn co http://foundry.supelec.fr/svn/luatex/tags/beta-0.45.0 Bugs and feature requests can be added to the issue tracker at http://tracker.luatex.org Have fun, Taco

2 1

Plant flu breakout?
by Taco Hoekwater 13 Nov '09

13 Nov '09

Hi, It looks like the garden is down again. Best wishes, Taco

1 0

refereces in mkiv
by Mojca Miklavec 10 Nov '09

10 Nov '09

Hello Hans, what's the status of referencing mechanism in MKIV? (I have some memories that you were implementing some fancy features, but the simple example doesn't work at the moment): \placeformula[test] \startformula a+b=c \stopformula \in[test] Thanks, Mojca

3 8

Luatex 0.44.0 announcement
by Taco Hoekwater 29 Oct '09

29 Oct '09

Hi, I have just uploaded the archives for a new luatex release, 0.44.0. There are not a lot of new features yet, but this releases fixes a fair amount of bugs found in 0.43.0 (and earlier versions). New features: * libpng is now 1.2.40. * hlist nodes now have a subtype to distinguish hlists generated by the paragraph breaking, explicit \hbox commands, and other (as-yet-undefined) sources. * \pdfobj now has an extra optional keyword: 'uncompressed', which forces the object to be written to the PDF in plain text (see the manual for the exact details). * The lua file from the luamd5 library (which provides the md5.hexsuma() function) is now embedded in the executable. Bug fixes: * \latelua now works in DVI mode again. * status.output_file_name could crash when used too soon. * In 0.43, rules were not output at all in DVI mode. * \leavevmode\penalty-10000\penalty-10000 crashed luatex. * the API of margin kern nodes to lua was using the wrong fields in the nodes. * A few fixes to the manual. * PDF Xref stream dicts were slightly non-conformant. * There was too much space between two consecutive math displays (this bug was inherited from Aleph) * Synctex was broken (thanks, Akira) * Font mark classes were handled incorrectly by the fontloader code. * There was a crash possible with certain types of named PDF destinations. * Math delimiters and large operators in new MATH fonts no longer apply the italic correction (this fixes placements of super- and subscripts on such items for those fonts). * The converter from lua strings to token lists (for the node interface) sometimes generated invalid tokens. * CFF (OpenType) fonts now listen to 'embedding="full"'. TTF is still todo. * Font tables with 'embedding="no"' would cause luatex to crash. * \pdfobj file {} failed to find the file if there was no lua callback in place. * An optimization in MS Mincho caused luatex to crash. The archives can be downloaded from supelec as usual: http://foundry.supelec.fr/gf/project/luatex/ You could also check out the sources via anonymous svn: svn co http://foundry.supelec.fr/svn/luatex/tags/beta-0.44.0 Bugs and feature requests can be added to the issue tracker at http://tracker.luatex.org Have fun, Taco

6 6

bibliography
by Taco Hoekwater 14 Oct '09

14 Oct '09

Hi Patrick, Can you kick the bib module off contextgarden? Hans has moved the code to the core, so the standalone module is now obsolete. Best wishes, Taco

3 5

Re: [dev-context] Second problem with Pizza and Problems
by Taco Hoekwater 04 Oct '09

04 Oct '09

Hi David, The tikz problem was on the mailing list already. This problem is because of euler math: the virtual fonts for that use computer modern for the missing glyphs in the euler set, and those are not provided by the minimals. A workaround is to add this statement after the \assignment \pdfmapline{=cmsy10 LMMathSymbols10-Italic "enclmmathsy ReEncodeFont" <lm-mathsy.enc <lmsy10.pfb } (it has to come after the font initialization, hence the ugly place) Perhaps the minimals should ship the traditional math fonts (cc dev-context), otherwise we could re-create the euler virtual fonts, or this mapline (and a dozen or so similar ones) could go into the euler typescript. Best wishes, Taco David Arnold wrote: > Hans, Taco, > > If I remove the \usemodule[tikz] and the tikz code, as in the attached > junk.tex, I still have a problem. > > systems : end file junk at line 71 > system : cont-err loaded > (/Users/darnold/context/tex/texmf-context/tex/context/base/cont-err.tex > systems : no file 'cont-sys.tex', using 'cont-sys.rme' instead > ) > ){/Users/darnold/context/tex/texmf/fonts/enc/dvips/tex-gyre/q-texnansi-sc.enc > }{/Users/darnold/context/tex/texmf/fonts/enc/dvips/tex-gyre/q-texnansi.enc} > !pdfTeX error: pdftex (file cmsy10.pfb): cannot open Type 1 font file > for readi > ng > ==> Fatal error occurred, no output PDF file produced! > TeXExec | runtime: 0.527318 > > This uses the same style files I sent with previous post. > > > > > ------------------------------------------------------------------------ >

2 1