- ntg-context - NTG Mailing Lists

Re: bugs (remailed)
by Hans Hagen 31 Mar '03

31 Mar '03

At 10:33 AM 3/31/2003 +0200, you wrote: >This was originally posted to ntg-context. I'm attaching the >C++ pretty typing environment and an example of bugs. > >I'm typesetting an algorithms course in C++ and I came around to few bugs >and personal wishes :) > >1. I've "written" a verb-cpp.tex which knows C++ keywords. However, as you'll > see from the attached example, it needs a few fixes: > - the missing word if it ends with & (the bar& x argument) > - & coloured in green alongside y (it should be grey since it's an > operator) > - &, +=, -=, *=, /=, %=, >>=, <<=, ->*, .* should be coloured like > operators > and must be considered as single tokens > - I'd like the whole comment in a single color (not red start/end and > black text). dark green would be ideal :) > >Since I know nothing about verbatim mechanism (I just took java environment >and stuffed it with C++ keywords) how hard would it be to make those changes? >I put '\input verb-cpp' into verb-ini.tex to make it work in my local >installation. \installprettytype [CPP] [CPP] \definetyping [CPP][option=CPP] >2. As you can see from the example, I want to typeset text next to code. > However, the footnotes within are not working. (context ver. 2002.10.23) > >3. Multi-column paragraphs are not automatically broken across pages. If > the paragraph doesn't fit on the current page, it will be moved to the > next page leaving an ugly empty box at the bottom of the previous page. > How can I achieve an effect similar to the attached example, but that will > automatically break across pages not leaving empty boxes at the bottom? \appendtoks\optimizeverbatimfalse\to\everytabulate \definetabulate [CodeExample] [|p|p|] \definebuffer[Snippet] \startSnippet \startCPP void f(bar& x, bar &y, struct foo *z) { x += z->y; x.*ptr(); z->*ptr(); /* comment */ y = 3; y += x; } \stopCPP \stopSnippet \startCodeExample \NC This is an example\footnote{of non-working} footnote. \NC \typebuffer[Snippet] \NC \NR \stopCodeExample >4. Some time ago I reported \pseudoencodedDJ macros not working in itemize > environment and you posted a fix to the list. Did you integrate the > fix in the latest release? think so, but probably changed a bit; this is what i have now \def\pseudoencodeddj % like in babel, but safer {\leavevmode\hbox\bgroup \setbox\scratchbox\hbox{d}% \scratchdimen\ht\scratchbox \advance\scratchdimen 1ex \scratchdimen.45\scratchdimen \dimen2\expandafter\withoutpt\the\fontdimen1\font\dimen0 \advance\dimen2 .5ex \hbox to \wd\scratchbox {\box\scratchbox\hss \raise\scratchdimen\hbox{\kern\dimen2\vbox{\hrule\!!height0.1ex\!!width0.3em}}}% \egroup} \def\pseudoencodedDJ % like in babel, but safer {\leavevmode \hbox\bgroup \setbox\scratchbox\hbox{D}% \scratchdimen.55\ht\scratchbox \dimen2\expandafter\withoutpt\the\fontdimen1\font\scratchdimen \advance\dimen2 .15ex \advance\dimen2 -.15\fontdimen7\font \hbox to \wd\scratchbox {\box\scratchbox\hss \raise\scratchdimen\hbox{\kern\dimen2\vbox{\hrule\!!height0.1ex\!!width0.3em}}}% \egroup} Hans ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma(a)wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------

1 0

Strange \placeformula
by Michal Kvasnicka 31 Mar '03

31 Mar '03

Good morning! I' have probably made a strange bug in my document layout. In adfadfasdfasdf \placeformula[eq:one] $$a=b,$$\noindentation adfasdfadsf the \placeformula macro doesn't produce a formula number -- the formula has no number and \in{equation}[eq:one] doesn't work as well. Sometimes there is also no space before the formula and after it. Extremely strange is that asdfasdfasdf \placeformula[eq:one] \startformula a=b \stopformula\noindentation adsfasdfasdf works well. The bug is definitely in my macros for document layout since another document works well. Can some of you hint where to find the bug? I'm nearly desperate. Well, I didn't redefined $ nor changed its catcode (at least not voluntaryly). Many thanks for any hint. With many greetings Michal Kvasnicka

2 1

Arranging problem / texexec--arrange
by Willi Egger 31 Mar '03

31 Mar '03

Hi Hans, While trying the possibilities to arrange pages I detected, that using the predefined European standardsizes of the pagesize and the sheets arranging works well. However if I try to use a custom size paper on a standard sized sheet, then I see, that the different pages do not seem to be positioned correctly. When using the option --arrange with texexec, Context halts after loading the basics. You can show this problem with the testfile I attach to this mail. Would you be so kind to have a look at this. I need to prepare a booklet, where I have to use a custom size pagesize. For the demonstration of the problems I attach a test file. Kind regards Willi

1 0

funny letters in (type)buffer
by Patrick Gundlach 31 Mar '03

31 Mar '03

Good evening, \enableregime [il1] \starttext \startbuffer funny letters: Ä ü ß \stopbuffer \typebuffer \stoptext gives me in test-test.tmp buffer: funny letters: ^^c4 ^^fc ^^df and this is also typeset like above. Is it possible to get the real letters back without manually editing the written buffer? Patrick

2 6

puzzled
by Marko Schuetz 30 Mar '03

30 Mar '03

I'm trying to put a right brace to the right of a framed box and thought I could do this: \starttext $\left. \framed[width=7cm,align=right,height=fit]{ something here \hairline \vphantom{something here} \hairline \vphantom{something here} } \right\} \text{entries for something}$ \stoptext but the brace is almost exactly twice as tall as the framed box. All help is appreciated! Thanks Marko

2 1

Texfont on i-Installer for Mac OSX
by Nigel King 29 Mar '03

29 Mar '03

Guys, Gerben Wierda has done an excellent job producing i-Installer for the Mac OSX which works with LaTeX etc very well. It is clear that many people are being introduced to TeX on the Mac now using this installer. Unfortunately there appears to be one problem with ConTeXt which is texfont. The tetex structure that Gerben has chosen is for the main fonts to be in /usr/local/teTeX/share/texmf.tetex/fonts Texfont will not recognise this location and the fonts in it and I suspect it is because of the full stop in texmf.tetex Evidence for this is in the following Running the single command texfont --en=8r --ve=urw --co=courier --so=auto Should work but returns processing aborted : unknown tex root /sw/share/texmf/ Clearly texroot is set incorrectly so forcing that texfont --en=8r --ve=urw --co=courier --so=auto --fontroot=/usr/local/teTeX/share/texmf.tetex/ Should work but returns processing aborted : unknown subpath ../fonts/afm/urw/courier And yet /usr/local/teTeX/share/texmf.tetex/fonts/afm/urw/courier/ucrb8a.afm Exists Are there any other possible reasons for this problem I have had some success now with fonts using adobekb.tex but the situation is not wholly satisfactory and I think I would really like texfont to work. -- Nigel

3 4

Two line head
by Nagy Bence 29 Mar '03

29 Mar '03

Hi all, I want to make a twoline header in a framed box: \setupheads[sectionnumber=no] \define[1]\Sectioncommand{\framed[frame=off,bottomframe=on]{#1}} \definehead[Section][section] \setuphead[Section][textcommand=\Sectioncommand] \Section{first line\crlf second line} I want to get something of the kind first line second line ------------- but my result is first linesecond line ----------------------- The contents seems to be good, the two lines appear. Greetings Bence

3 4

hyphenation in \type{....}?
by Patrick Gundlach 28 Mar '03

28 Mar '03

Dear ConTeXt hackers, I am a bit confused: as stated in the manual, \type{...} does not hyphenate (in contrast to \typ{...}), but in the following example, it does: \enableregime [il1] \starttext ist zwar nur einem ist zwar nur einem \type{averylongword averylongword averylongword averylongword} \stoptext How do I get garanteed-non-hypenated-typing? Patrick

2 1

Fwd: Re[4]: [NTG-context] MetaPost/Fun question
by Hans Hagen 28 Mar '03

28 Mar '03

Hi, Here is some comment on GB's intersection problem/solution from the polish mp expert Boguslaw Jackowski (those who will be in bremen next week can meet this personality). Hans > >EHL> I implemented it by finding some intersectiontime and the recursively > >EHL> finding the intersectionstimes of the two subpaths (each with some > >EHL> neighbourhood of the first intersection point cut off). For the time > >EHL> being I cannot find my code, but here it is in pseude code. > > > >EHL> When done this way, the intersection times will be sorted according to > >EHL> the first path. > >Frankly speaking, such a procedure should belong to plain from the very >beginning... Obviously, we added it in our plain_ex that we use for >generating fonts (even more -- we find also common outlines). > >Attached you'll find the relevant excerpt plus examples. > >Incidentally, recursion seem still banned from the MF/MP realm. I complained >about it some time ago, but it seems that nobody but me is hurted by this >annoyin situation. Attached you'll find an example: 31-element series >(in a reversed order) break down w2c mpost... > >And finally, let me quote the post from Larry Siebenmann (from 30 Jun 2000): > > > I believe that value (-1,-1) for "intersectiontimes" is a > > 100% ironclad guarantee that its two path arguments are > > disjoint. > > > > Is the converse true? That is, if "intersectiontimes" > > yields positive times, do the two paths necessarily > > intersect? NO! In writing the last post, I confirmed to my > > satisfaction that MP and MF report positive intersection > > times for paths that only come near. Start by looking > > at the following -- in which one of the paths is constant. > > > > path c; > > pair v,w; > > c:=fullcircle scaled 2; > > v:=(1+19epsilon,0) rotated 33.1234567; > > w:=(1-15epsilon,0) rotated 33.1234567; > > show (v intersectiontimes c); > > show (w intersectiontimes c); > > v:=(1+20epsilon,0) rotated 33.1234567; > > w:=(1-16epsilon,0) rotated 33.1234567; > > show (v intersectiontimes c); > > show (w intersectiontimes c); > > > > end > > > > The log: > > > > This is MetaPost, Version 0.641 > > **&Plain filetest.mp > > (filetest.mp > > >> (0.47754,0.73682) > > >> (0.96869,0.7373) > > >> (-1,-1) > > >> (-1,-1) ) > > > > Thus "intersectionpoint" may return a point when the paths > > are disjoint. It is guaranteed only to return a point of > > close approach. Close approach algorithms are complicated > > when efficient. But they are trivial in principle and enjoy > > incredible generality:- indeed, since we have explicit > > bounds modulus of continuity, it suffices to measure image > > distances on a finite but fine net of points in source. > > > > Question: Is there an *if-and-only-if* test for b'ezier > > curve intersection? > > > > The answer is yes for linear segments, though in some, but > > not all, cases one needs exact rational arithmetic. So > > think of working in C. > >I think you have now more comments than enough ;-) > >Cheers -- Jacko > >-- >BOP s. c. >ul. Bora-Komorowskiego 24, 80-377 Gdansk, Poland >tel. (+48 58) 553 46 59, fax (+48 58) 511 03 81 >bop(a)bop.com.pl, http://www.bop.com.pl > >================================================================ >Deze e-mail is door E-mail VirusScanner van Planet Internet gecontroleerd >op virussen. >Op http://www.planet.nl/evs staat een verwijzing naar de actuele lijst >waar op wordt gecontroleerd. ------------------------------------------------------------------------- Hans Hagen | PRAGMA ADE | pragma(a)wxs.nl Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com ------------------------------------------------------------------------- information: http://www.pragma-ade.com/roadmap.pdf documentation: http://www.pragma-ade.com/showcase.pdf -------------------------------------------------------------------------

1 0

color band
by Idris S Hamid 27 Mar '03

27 Mar '03

Dear syndicate, For a book cover I want a green-colored rectangular band, centered, at the very top of the page, with text in some other color (to be determined) overlapping the band. The band is small, just, say, 2cm high and 12 cm wide. Since it's at the top it is surrounded by white on only three sides. I'm not even sure where to get started here (I don't think the LaTeX Graphics Companion will be much help:->) Here is part of my present layout. I'll have to put this band in the topspace and increase it, I presume: \definepapersize[journal][width=5.5in, height=8.5in] \setuppapersize[journal][letter] \setuplayout[width=middle, | height=19cm, | footer=0cm, | header=0cm, | topspace=28pt, | location={middle,doublesided}, | marking=on] I hope the solution is Plain-compatible since I may have to use Gamma for some Arabic insertions. Otherwise I'll have to eps those insertions. Thnx in advance Idris

7 23