[texhax] Specifying baseline-to-baseline skip in TeX

[Uwe Lück]

More general and simple, ignoring TeX, a question to professional graphical 
designers: What relevance is there in specifying baseline-to-baseline 
skips? I thought that the white space is more relevant and should be 
specified instead.

-- Uwe.

At 11:20 14.10.09, Uwe Lück wrote:
>Thanks a lot, Donald.
>
>At 16:17 12.10.09, Donald Arseneau wrote:
>>Uwe Lück <uwe.lueck at web.de> writes:
>>
>> > The /designer/ has specified the skip from the last \chapter baseline 
>> to the
>> > next baseline. A simple \vskip(glue) closing the chapter macros [apart 
>> from
>> > \@afterheading] would result in a baseline-to-baseline distance of 
>> (glue)+(b)
>> > where (b) is the \baselineskip associated with either \normalsize or
>> > \section, I don't know ...
>>
>>The descenders of large-font chapter and section titles
>>screw up surrounding \baselineskips, often causing \lineskip
>>to be used.
>
>I sometimes thought I will rule this out by \nointerlineskip, then 
>realized that this needs estimating ascender heights, worse than 
>\baselineskip, then forgot ...
>
>>You can get a specific ad-hoc baseline-skip between paragraphs
>>using something like:
>>
>>\newdimen\prevdepthcalc
>>
>>\def\blskipper#1{\par
>>\prevdepthcalc=\prevdepth
>>\advance\prevdepthcalc -#1\relax
>>\advance\prevdepthcalc\baselineskip
>>\advance\prevdepthcalc\parskip
>
>\parskip and/or \baselineskip may have shrink/stretch which may be wanted 
>to remove => \newskip
>
>>\ifdim\prevdepthcalc<-99pt
>>   \vskip-99pt
>>   \vskip-\prevdepthcalc
>>   \prevdepthcalc=-99pt
>>\fi
>
>Why this? Did you mean "999" to prevent implicit \nointerlineskip?
>
>>\prevdepth=\prevdepthcalc
>>}
>
>What about the \lineskip problem? We might rather \vskip some 
>\prevdepthcalc results and then \prevdepth=0, or even less (in case of 
>accent stacks in the first following line ...).
>
>>It has to have the \baselineskip for the following
>>paragrah already set, but that is not possible in general
>>(the paragraph could end like {\small\par}). If you
>>want to delay operation until \everypar, then you
>>will have to back out of the paragraph, do \blskipper,
>>and re-start the paragraph.  That goes roughly like:
>>
>>\begingroup
>>   \everypar{}%
>>   \setbox\@tempboxa\lastbox
>>   \@@par
>>   \vskip-\parskip
>>   \blskipper{xxxxpt}%
>>   \ifvoid\@tempboxa \noindent \else \indent \fi
>>\endgroup
>
>Ah, just for a possible indent, I understand.
>
>>But you have to make sure this gets executed *first*, before
>>whatever other code is executed by \everypar.
>
>I thought a little /LaTeX/ package could be made right now. In the 
>meantime I thought of modifying \@setfontsize and \fontsize, testing for 
>vmode, but \everypar may be even better. Moreover, the "next baselineskip" 
>could be specified by the font size command (read baseline skip from 
>expansion) or even the sectioning command (read the font size command from 
>its expansion).
>
>Another problem to solve for the \everypar approach is \@afterheading, 
>especially with \chapter etc.
>
>In my own case, I am assuming \normalsize text is following and 
>suspecting: the designer was not aware of \section, however the few pt 
>exceeding the specification with a larger font are favourable, it is nice 
>if the white space is about the same with a larger following font.
>
>Best,
>
>     Uwe.
>
>
>>If you are worried about the positioning below chapter
>>heads with [twocolumn] you are out of luck.  See bug
>>report latex/3126.
>
>Indeed, the index. But there were suggestions, it seems they were worth 
>trying? (The book class I am working for uses multicol instead.)
>
>_______________________________________________
>TeX FAQ: http://www.tex.ac.uk/faq
>Mailing list archives: http://tug.org/pipermail/texhax/
>More links: http://tug.org/begin.html
>
>Automated subscription management: http://tug.org/mailman/listinfo/texhax
>Human mailing list managers: postmaster at tug.org