Re: [Texmacs-dev] A first attempt to analyze TeXmacs performance behavior

[Joris van der Hoeven]

> >  line-breaking: to be reimplemented for several reasons
> What are those reasons?

1) A non-trivial speed factor can be gained (we already do this in
   the case of page-breaking) by eliminating certain possible
   break-points which are certainly bad.
2) We should deal with breaking very long strings (like 10000 digit
   numbers) in a better way.
3) We should allow for further fine-tuning: forbid/allow breaks
   on successive lines, forbid/allow breaks after very short words, etc.
4) We should implement user-provided hyphenation patters.
5) We should implement penalties on regions of text and
   penalties which depend on the n-th line.
6) We should allow for paragraphs with variable width
   (at least provide a simple line-breaking algorithm for this case).
7) Better treatment of mathematical languages & penalties.
8) Special rewritings of big expressions like fractions when
   necessary before breaking.

And I may have forgotten some reasons...