March 2014 - ntg-context

Text background regression in 20140316 beta
by Rik Kabel 16 Mar '14

16 Mar '14

Hans and list, Previous to the 20140316 beta, we were able to prevent text backgrounds from bleeding over the footnote space with \chardef\kindofpagetextareas\plusone as you described in a 2008 list entry. While the problem it resolves is still there, the fix is now broken. The background on the following page is starts too far down the page (although it appears to have the proper depth). The following example demonstrates the problem. Sorry for its length. \setuppapersize[A6][A6] \definetextbackground[blockquotebackground] [ location=paragraph, background=color, backgroundcolor=gray, backgroundoffset=.25em, frame=off, ] \definedelimitedtext[BlockQuote] \setupdelimitedtext [BlockQuote] [ leftmargin=1pc, rightmargin=1pc, before={% \blank[halfline]% \startblockquotebackground% \setupindenting[next]% }, after={% \stopblockquotebackground% \blank[halfline]% }, ] \starttexdefinition startAttribution \setupnarrower[left=1pc] \startnarrower[left] \leavevmode\llap{{--\,}}% \stoptexdefinition \starttexdefinition stopAttribution \stopnarrower \stoptexdefinition %% The following prevents the background from overlaying %% footnotes. Found in a 2008 post, still not in the wiki. %% http://www.mail-archive.com/ntg-context%40ntg.nl/msg15681.html %% %% With the 20140316 beta, things have changed again. Now the %% problem without it is still there, but with it the background %% is misplaced on the second page of a two-page background. %% \chardef\kindofpagetextareas\plusone \starttext \startBlockQuote Honorar und Verbot des Nachdrucks sind im Grunde der Verderb der Litteratur. Writing for money and reservation of copyright are, at bottom, the ruin of literature. \startAttribution Arthur Schopenhauer, chapter {\sc xxiii} \quotation{Ueber Schriftstellerei und Stil} {\em Parerga und Paralipomena} page 420\footnote{Translation by Thomas Bailey Saunders, {\em The Art of Literature: A Series of Essays} page 3} \stopAttribution \stopBlockQuote Again. \startBlockQuote Honorar und Verbot des Nachdrucks sind im Grunde der Verderb der Litteratur. Writing for money and reservation of copyright are, at bottom, the ruin of literature. \startAttribution Arthur Schopenhauer, chapter {\sc xxiii} \quotation{Ueber Schriftstellerei und Stil} {\em Parerga und Paralipomena} page 420\footnote{Translation by Thomas Bailey Saunders, {\em The Art of Literature: A Series of Essays} page 3} \stopAttribution \stopBlockQuote Again. \startBlockQuote Honorar und Verbot des Nachdrucks sind im Grunde der Verderb der Litteratur. Writing for money and reservation of copyright are, at bottom, the ruin of literature. \startAttribution \leavevmode\llap{{--\,}}% Arthur Schopenhauer, chapter {\sc xxiii} \quotation{Ueber Schriftstellerei und Stil} {\em Parerga und Paralipomena} page 420\footnote{Translation by Thomas Bailey Saunders, {\em The Art of Literature: A Series of Essays} page 3} \stopAttribution \stopBlockQuote \stoptext

1 0

Selectfont module and bodyfont size in footnotes
by Alan Bowen 16 Mar '14

16 Mar '14

The following minimal example \definefontfamily[mainface][serif][TeX Gyre Termes Serif] \setupbodyfont[mainface,15pt] \setupnotes[footnote][bodyfont={mainface,12pt}] \starttext text\footnote{Still more} and more text\footnote{And then again} \stoptext gets the size of the first note at 12 pt but sets the second at 15pt. How can I fix this? Alan

2 2

Re: [NTG-context] Regression in register processing in last beta
by Hans Hagen 16 Mar '14

16 Mar '14

On 3/16/2014 7:25 PM, Jan Tosovsky wrote: > Dear All, > > I have to admit I use quite cumbersome code to get desired output, but while > it was working in previous betas, now emergency stop is reported: > 'Extra }, or forgotten \endgroup' > > Here is MWE: > > \setupalign[hz, hanging] > \setupindenting[medium, yes] > \setupnotation[footnote][align={hz, hanging}] > > \starttext > Hello One > \start > \setuptolerance[strict] > \startregister[index][two]{two} > Hello Two > > This is \footnote{ > \parfillskip 14pt plus 1 fill > Footnote \index{content}content. > }\par\stop % this brace is reported as unexpected character here > \stopregister[index][two] > > \page > \placeindex > \stoptext > > Briefly, when footnotes appear in \startregister ... \stopregister block, > processing is somehow confused and ends prematurely. > > ConTeXt-2014.03.16 > LuaTeX 0.79 > Win7-64bit > > Any suggestion how to update my code to get it working again? ah, i have to swap some definition order ... but anyway, your start-stop is somewhat messed up (in fact, the start-stop names for registers are not really properly nested ones so maybe i should use other names anyway new beta coming 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

Re: [NTG-context] Regression in register processing in last beta
by Jan Tosovsky 16 Mar '14

16 Mar '14

On 2014-03-16 Jan Tosovsky wrote: > Dear All, > > I have to admit I use quite cumbersome code to get desired output, but > while > it was working in previous betas, now emergency stop is reported: > 'Extra }, or forgotten \endgroup' > > Here is MWE: > > \setupalign[hz, hanging] > \setupindenting[medium, yes] > \setupnotation[footnote][align={hz, hanging}] > > \starttext > Hello One > \start > \setuptolerance[strict] > \startregister[index][two]{two} > Hello Two > > This is \footnote{ > \parfillskip 14pt plus 1 fill > Footnote \index{content}content. > }\par\stop % this brace is reported as unexpected character here > \stopregister[index][two] > > \page > \placeindex > \stoptext > > Briefly, when footnotes appear in \startregister ... \stopregister > block, > processing is somehow confused and ends prematurely. > > ConTeXt-2014.03.16 > LuaTeX 0.79 > Win7-64bit > > Any suggestion how to update my code to get it working again? > There was mistake in my code: \startregister has to be before \start command to be correctly paired. Maybe parsing is now stricter, which I rather appreciate :-) Sorry for the noise, Jan

1 0

(again) bug?: \about disables \setupnote[split=strict]
by Pablo Rodriguez 16 Mar '14

16 Mar '14

Dear list, I have the following sample: \setuppapersize[A6] \setupnote[footnote][split=strict] \starttext \section[bspaces]{Blank spaces} \section[paragraph]{Paragraphs} \dorecurse{11}{This is a sentence.\par} This is already explained in sections \about[bspaces] y \about[paragraph]. %~ \namedstructurevariable{section}{title} and \namedstructurevariable{section}{title}. \dorecurse{2}{This is a sentence. }.\footnote{\dorecurse{5} {this is a footnote sentence. }} This is another sentence. \stoptext If I use the lines with \about, footnote reference in body and footnote are placed in different pages. If I use \namedstructurevariable, footnote reference in body and footnote are placed in the same page. I have two questions: Could anyone confirm this is a bug in \about? Is there any way to use \namedstructurevariable for sections other than the current? (So I would be able to avoid using \about.) Many thanks for your help, Pablo -- http://www.ousia.tk

2 1

Regression in register processing in last beta
by Jan Tosovsky 16 Mar '14

16 Mar '14

Dear All, I have to admit I use quite cumbersome code to get desired output, but while it was working in previous betas, now emergency stop is reported: 'Extra }, or forgotten \endgroup' Here is MWE: \setupalign[hz, hanging] \setupindenting[medium, yes] \setupnotation[footnote][align={hz, hanging}] \starttext Hello One \start \setuptolerance[strict] \startregister[index][two]{two} Hello Two This is \footnote{ \parfillskip 14pt plus 1 fill Footnote \index{content}content. }\par\stop % this brace is reported as unexpected character here \stopregister[index][two] \page \placeindex \stoptext Briefly, when footnotes appear in \startregister ... \stopregister block, processing is somehow confused and ends prematurely. ConTeXt-2014.03.16 LuaTeX 0.79 Win7-64bit Any suggestion how to update my code to get it working again? Thanks, Jan

1 0

Re: [NTG-context] [Index] Indenting long primaries after secondaries
by Hans Hagen 16 Mar '14

16 Mar '14

On 3/16/2014 6:31 PM, Jan Tosovsky wrote: > On 2014-03-16 Hans Hagen wrote: >>>>> On 2014-02-26 Jan Tosovsky wrote: >>>>>> >>>>>> when a long primary is preceded with the secondary or tertiary >>>>>> index entry, the second line is not indented as usual, see >>>>>> the following example: >>>>>> >>>>>> \setupregister[balance=no] >>>>>> \starttext >>>>>> \index{primary+secondary+tertiary} >>>>>> \index[primary long]{\dorecurse{5}{primary long }} >>>>>> \index[primary longer]{\dorecurse{5}{primary longer }} % Ok >>>>>> \input{tufte} >>>>>> \completeindex >>>>>> \stoptext >>>>>> >>> >>> My main wish is fixing that indentation (or offering any workaround). >> >> we get this: >> >> p >> primary >> secondary >> tertiary 1 >> primary long primary long primary >> long primary long primary long 1 >> primary longer primary longer primary >> longer primary longer primary longer >> 1 >> > > hmm, I was getting different output in ConTeXt-2014.03.07: > > p > primary > secondary > tertiary 1 > primary long primary long primary > long primary long primary long 1 (!!! this was not indented) > primary longer primary longer primary > longer primary longer primary longer 1 > > (or see the screenshot) > > I've just updated ConTeXt to 2014.03.16 and now it works as expected! > > Thanks a lot! The register mechanism has been adapted a bit and this was probably a positive side effect. Some changes: - using mixed columns - support for: referencemethod=forward (smaller tuc files) - more efficient interactive support (esp when no page numbers) - better keeping entries and subentries together Apart from that the beta has some (small) speed improvements and there will be a slightly different index anchoring once the new version of luatex is available on all platforms. Also, the beta has slightly different low level implementation of some of the interactive features (which can result in more efficient pdf files in some cases). In fact, due to some other changes, pdf files could be smaller anyway. 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 -----------------------------------------------------------------

1 0

Re: [NTG-context] [Index] Indenting long primaries after secondaries
by Hans Hagen 16 Mar '14

16 Mar '14

On 3/16/2014 4:34 PM, Jan Tosovsky wrote: > On 2014-03-16 Hans Hagen wrote: >> On 3/7/2014 6:54 PM, Jan Tosovsky wrote: >>> On 2014-02-26 Jan Tosovsky wrote: >>>> >>>> when a long primary is preceded with the secondary or tertiary >>>> index entry, the second line is not indented as usual, see >>>> the following example: >>>> >>>> \setupregister[balance=no] >>>> \starttext >>>> \index{primary+secondary+tertiary} >>>> \index[primary long]{\dorecurse{5}{primary long }} >>>> \index[primary longer]{\dorecurse{5}{primary longer }} % Ok >>>> \input{tufte} >>>> \completeindex >>>> \stoptext >>>> >> >> how is a \dorecurse supposed to get sorted ... >> > > That \dorecurse is used in this MWE only to make it shorter. Indeed, this > command confuses the letter grouping procedure so this is the reason why I > specified the sorting key via the bracket syntax []. > > I think there is no need to fix this grouping issue (not enough ROI), but if > so, I would expect to resolve the \dorecurse command first and then apply > the grouping/sorting etc. > > My main wish is fixing that indentation (or offering any workaround). we get this: p primary secondary tertiary 1 primary long primary long primary long primary long primary long 1 primary longer primary longer primary longer primary longer primary longer 1 getting this: p primary secondary tertiary 1 primary long primary long primary long primary long primary long 1 primary longer primary longer primary longer primary longer primary longer 1 is pretty hard because it would demand postprocessing entries but i wonder it looks better \defineprocessor[twolines][style=\aftergroup\hangafter\aftergroup\plustwo] \setupregister [balance=no] \starttext \index{primary+secondary+tertiary} \index{twolines->primary long primary long primary long primary long primary long} \index{twolines->primary longer primary longer primary longer primary longer primary longer} \input{tufte} \completeindex \stoptext assuming that we never change the implementation 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

Re: [NTG-context] [Index] Indenting long primaries after secondaries
by Hans Hagen 16 Mar '14

16 Mar '14

On 3/16/2014 4:34 PM, Jan Tosovsky wrote: > On 2014-03-16 Hans Hagen wrote: >> On 3/7/2014 6:54 PM, Jan Tosovsky wrote: >>> On 2014-02-26 Jan Tosovsky wrote: >>>> >>>> when a long primary is preceded with the secondary or tertiary >>>> index entry, the second line is not indented as usual, see >>>> the following example: >>>> >>>> \setupregister[balance=no] >>>> \starttext >>>> \index{primary+secondary+tertiary} >>>> \index[primary long]{\dorecurse{5}{primary long }} >>>> \index[primary longer]{\dorecurse{5}{primary longer }} % Ok >>>> \input{tufte} >>>> \completeindex >>>> \stoptext >>>> >> >> how is a \dorecurse supposed to get sorted ... >> > > That \dorecurse is used in this MWE only to make it shorter. Indeed, this > command confuses the letter grouping procedure so this is the reason why I > specified the sorting key via the bracket syntax []. > > I think there is no need to fix this grouping issue (not enough ROI), but if > so, I would expect to resolve the \dorecurse command first and then apply > the grouping/sorting etc. > > My main wish is fixing that indentation (or offering any workaround). stylistic variant: \defineprocessor[twolines][command=\MyStyle] \unexpanded\def\MyStyle#1% {\setbox\scratchbox\hbox{#1}% \scratchdimen\dimexpr\hsize-\leftskip-\hangindent\relax \ifdim\wd\scratchbox>\scratchdimen \aftergroup\hangafter \aftergroup\plustwo \aftergroup\break \else\ifdim\wd\scratchbox>2\scratchdimen \aftergroup\hangafter \aftergroup\plustwo \aftergroup\break \fi\fi \unhbox\scratchbox} \setupregister [balance=no] \setupregister [index:1] [textstyle=\MyStyle] \starttext \index{primary+secondary+tertiary} \index{twolines->primary long primary long primary long primary long primary long} \index{twolines->primary longer primary longer primary longer primary longer primary longer} \input{tufte} \completeindex \stoptext ----------------------------------------------------------------- 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 -----------------------------------------------------------------

1 0

Re: [NTG-context] [Index] Indenting long primaries after secondaries
by Hans Hagen 16 Mar '14

16 Mar '14

On 3/7/2014 6:54 PM, Jan Tosovsky wrote: > On 2014-02-26 Jan Tosovsky wrote: >> >> when a long primary is preceded with the secondary or tertiary >> index entry, the second line is not indented as usual, see >> the following example: >> >> \setupregister[balance=no] >> \starttext >> \index{primary+secondary+tertiary} >> \index[primary long]{\dorecurse{5}{primary long }} >> \index[primary longer]{\dorecurse{5}{primary longer }} % but this is Ok >> \input{tufte} >> \completeindex >> \stoptext >> > > I've tested this in new beta and the issue is still there... how is a \dorecurse supposed to get sorted ... ----------------------------------------------------------------- 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