[NTG-context] upload

[Hans Hagen]

Hi,

Normally we don't report on successive loe level improvements in the 
engine or the core of context (we do so in articles and wrapups), and 
much goes unnoticed anyway (which is good). But ... the last couple of 
weeks we have been playing with some adapted functionality in lmtx and 
some of the experimental low level code has now made it in core. 
Normally users will not notice this but who knows.

There are already features in lmtx that are not (and are unlikely to be) 
in mkiv (most significant in the metapost related part). When possible 
new macros and additions to mechanism will be in both mkiv and lmtx, but 
lmtx users will be better off in the end. We're mostly talking of the 
more tricky mechanisms here, those that currently have limitations due 
to all kind of possible interferences. To summarize:

mkii : frozen (seldom patches)
lmtx : cutting edge
mkiv : when possible following lmtx, otherwise frozen

Because we develop lmtx and luametatex alongside it is no problem if we 
have experimental (and even temporary imperfect) code in the engine: 
context knows about it and especially with experiments, we try to make 
sure it doesn't hit users (who we don't expect to use those, often 
undocumented, new tricks anyway). It's sometimes the only way to go 
forward (stepwise testing).

So, here is an example of where lmtx can do better than mkiv:

\placefigure[left]{}{}      \input ward  \par
\placefigure[left]{}{} {\bf \input ward} \par

Another area is in using footnotes. As a first step to more robust 
handling of deeply nested notes, lmtx now defaults to (the more than a 
decade optional in mkiv) \automigrateinserts, but with less overhead 
(for performance reasons, it was never enabled in mkiv, apart from 
occasional side effects). It is features like these that will stepwise 
be improved, at least that is what we hope for.

So, the current upload has some new stuff, hopefully without too many 
side effects, but if so, we'll deal with it as usual. As a result, the 
latest lmtx could be a little slower, but I wonder if someone will 
notice it in practice.

Hans

-----------------------------------------------------------------
                                           Hans Hagen | PRAGMA ADE
               Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
        tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl
-----------------------------------------------------------------