- dev-context - NTG Mailing Lists

Reset counter with each heading
by Wolfgang Schuster 02 Oct '12

02 Oct '12

Hi Hans, can you add a option for the “way” key for \setupcounter which resets a counter for every section independent of the level which was requested by Peter Schorsch on the mailing list. I looked into the source file (strc-num.lua and strc-sec.mkiv) and only four lines have to be changed to add this feature (I used “head” for the name because there was a predefined variable). strc-sec.mkiv: \setvalue{\??headlevel\v!block}{0} \setvalue{\??headlevel\v!none }{-1} \setvalue{\??headlevel\v!text }{-2} +\setvalue{\??headlevel\v!head }{-3} strc-num.lua: function counters.setown(name,n,value) local cd = counterdata[name] if cd then local d = allocate(name,n) d.own = value d.number = (d.number or d.start or 0) + (d.step or 0) local level = cd.level if not level or level == -1 then -- -1 is signal that we reset manually - elseif level > 0 then + elseif level > 0 or level == -3 then check(name,d,n+1) elseif level == 0 then -- happens elsewhere, check this for block end synchronize(name,d) end end function counters.add(name,n,delta) local cd = counterdata[name] if cd and (cd.state == v_start or cd.state == "") then local data = cd.data local d = allocate(name,n) d.number = (d.number or d.start or 0) + delta*(d.step or 0) -- d.own = nil local level = cd.level if not level or level == -1 then -- -1 is signal that we reset manually if trace_counters then report_counters("adding, name: %s, level: manually, action: no checking",name) end elseif level == -2 then -- -2 is signal that we work per text if trace_counters then report_counters("adding, name: %s, level: text, action: checking",name) end check(name,data,n+1) - elseif level > 0 then + elseif level > 0 or level == -3 then -- within countergroup if trace_counters then report_counters("adding, name: %s, level: %s, action: checking within group",name,level) end check(name,data,n+1) elseif level == 0 then -- happens elsewhere if trace_counters then report_counters("adding, name: %s, level: %s, action: no checking",name,level) end else if trace_counters then report_counters("adding, name: %s, level: unknown, action: no checking",name) end end synchronize(name,d) return d.number -- not needed end return 0 end function counters.check(level) for name, cd in next, counterdata do - if cd.level == level then + if cd.level == level or level > 0 and cd.level == -3 then if trace_counters then report_counters("resetting, name: %s, level: %s",name,level) end reset(name) end end end The question for the counter.check function is if it should also reset the counter for each new sectionblock (replace my change above with “if cd.level == level then”) or only at the start of a new section (that’s how my change behaves now). Wolfgang

2 2

supp-mat.mkiv
by Wolfgang Schuster 16 Sep '12

16 Sep '12

Hi Hans, \let\normalsuperscript \Usuperscript \let\normalsubscript \Usubscript \let\normalstartimath \Ustartmath \let\normalstopimath \Ustopmath \let\normalstartdmath \Ustartdisplaymath -\let\normalstopdmath \Ustopmath +\let\normalstopdmath \Ustopdisplaymath \def\normalmathaligntab{&} % \let\normalmathaligntab\aligntab does to work well in a let to & (a def works ok) \let\normalsuper \Usuperscript % obsolete \let\normalsuber \Usubscript % obsolete \let\startimath \Ustartmath \let\stopimath \Ustopmath \let\startdmath \Ustartdisplaymath -\let\stopdmath \Ustopmath +\let\stopdmath \Ustopdisplaymath Wolfgang

2 1

Spurious parameter in m-visual.mkiv
by Marco 16 Jul '12

16 Jul '12

Hi! There's a spurious parameter in \ruledbaseline, which breaks the visual module. I guess it's a leftover from the clean-up. I assume the parameter can simply be omitted (untested): \unexpanded\def\ruledbaseline {\vrule \!!width \zeropoint \bgroup \dontinterfere \doglobal\increment\ruledbaselines \scratchdimen\nofruledbaselines\baselineskip \setbox\scratchbox\normalvbox to 2\scratchdimen {\leaders \normalhbox {\strut \vrule \!!height \testrulewidth \!!depth \testrulewidth \!!width 120\points} \normalvfill}% \smashbox\scratchbox \advance\scratchdimen \strutheightfactor\baselineskip \setbox\scratchbox\normalhbox {\normalhskip -48\points \normalhbox to 24\points {\normalhss - {\ttxx#1}% - \ruledbaselines + {\ttxx\ruledbaselines}% \normalhskip6\points}% \raise\scratchdimen\box\scratchbox}% \smashbox\scratchbox \box\scratchbox \egroup} Marco

2 1

beta
by Hans Hagen 06 Jul '12

06 Jul '12

Hi Wolfgang, The current beta has pack-rul.mkiv adapted ... it also uses the commandhandler code now so you might want to have a look at it. There are now only a few mechanisms left that don't use the new parent chain code. AFAIKS the speed penalty is neglectable. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

2 1

\framed[width=broad]
by Wolfgang Schuster 19 Jun '12

19 Jun '12

Hi, framed has the two keyword “fit” and “broad” for the height parameter but both have the same effect. Can this be changed and “height=broad” result in “height=\vsize” like “width=broad“ results in “width=\hsize”. There are cases like this \setupheadertexts[{\framed[width=\hsize,height=\vsize]{…}}] where I would prefer to write “width=broad” and “height=broad” instead of passing dimensions to the keys. Wolfgang

3 4

Bug in \usepath
by Marco 18 Jun '12

18 Jun '12

There's something fishy with \usepath. A simple call fails with ! LuaTeX error ...-new/tex/texmf-context/tex/context/base/data-res.lua:688: attempt to get length of local 'subpaths' (a nil value) stack traceback: ...-new/tex/texmf-context/tex/context/base/data-res.lua:688: in function 'registerextrapath' ...-new/tex/texmf-context/tex/context/base/file-job.lua:71: in function 'usepath' <main ctx instance>:1: in main chunk. \usepath [chapters] \starttext Foo \stoptext Marco

2 1

Adding setuvalue etc to syst-gen.mkii
by Aditya Mahajan 18 Jun '12

18 Jun '12

Hi Hans. Can you add the following block of macros from syst-aux.mkiv to syst-gen.mkii? \def\setuvalue #1{\normalprotected\expandafter \def\csname#1\endcsname} \def\setuevalue #1{\normalprotected\expandafter\edef\csname#1\endcsname} \def\setugvalue #1{\normalprotected\expandafter\gdef\csname#1\endcsname} \def\setuxvalue #1{\normalprotected\expandafter\xdef\csname#1\endcsname} I wonder how long should I support MkII for my modules? Aditya

2 1

Change preset to setups in page-lay.mkiv
by Aditya Mahajan 14 Jun '12

14 Jun '12

Hi, \setuplayouts has a "preset" option, which is used for specifying setups. This is the only instance where "preset" is used to specify "setups". All other uses of "preset" is in layers. I think that it will be more consistent to change that option to setups. Aditya

2 1

ctx file can’t be used
by Wolfgang Schuster 11 Jun '12

11 Jun '12

Hi Hans, ctx files can’t be used in the last beta. This context --ctx=s-mod.ctx tabl-ntb.mkiv generates the following error message: ctxname: s-mod.ctx...ex/texmf-context/scripts/context/lua/mtx-context.lua:261: bad argument #1 to 'match' (string expected, got nil) Wolfgang

2 2

bTABLE in header and document give unexpected behavior
by Sido Jensma 11 Jun '12

11 Jun '12

This is the first time I am posting here. I am not really sure what a typical bug report should look like. I had posted a question on stack-exchange and it was suggested to post a bug report here. I’m not sure if this is a bug, but it was unexpected behavior to me. The original question was posted here: http://tex.stackexchange.com/questions/59134/strange-problem-with-frames-co… And a ‘not-so-minimal’ example of what is happening is attached (see the last page). I have tested this with extreme tables and don’t get the same behavior.

4 3