>> Thanks, this helps: for my version of the LilyPond Notation
>> Reference a value of 10pt works just fine.  Before that I tried
>> 20pt, however, I got a different but equally bad gap as reported
>> originally.  This makes me wonder whether we are we really turning
>> the right screw...
> 
> I tried 30pt and 10pt with texinfo.texi.  Images are attached
> showing the difference this made.

Thanks.

> You can see with the 10pt measurement, chapter 17 and section 17.1
> are listed at the bottom of the page, while with 30pt, they are
> pushed onto the next page.
> 
> I am not really sure which is better.  There is more in chapter 17
> and one argument is that it is better to start a new page for
> chapter 17's contents listing.

Both are OK to me.  However, what I have a problem with are cases
where too much empty vertical space remains, which could be easily
filled with a chapter entry and four lower-level entries, say.

> I know you cannot easily send me the entire input file, but if you
> could email me the *.toc file on its own, I should be able to
> process it and see the gaps you are talking about.

Attached, and thanks for looking into this — and yes, we use A4 paper.

>> If I understand you correctly the bad break is the result of
>> cumulative effects.  Maybe this can be avoided by not cumulating
>> them?  Otherwise I can imagine to make the `plus 60pt` part of the
>> skip used in `\raggedbottom` a user-defined variable.
> 
> The interaction between the penalty and the vertical skip is how our
> implementation of ragged bottom works.

I think I was unclear: I see that the innermost command
`\entryinternal` is used by `\tocentry`, which in turn is used by
`\dochapentry`, which is used by `\numchapentry` – and every level
adds more stuff.  This I mean with 'cumulative'.

> We don't provide such fine-tuning settings for page layout if at all
> possible.

I thought about a command like `@compacttoc` that reduces the skip to
a lower value.


    Werner

Attachment: notation.toc.xz
Description: Binary data

Reply via email to