[luatex] Bug involving rubber lengths

Robin Fairbairns Robin.Fairbairns at cl.cam.ac.uk
Tue Oct 19 17:42:23 CEST 2010


Taco Hoekwater <taco at elvenkind.com> wrote:

> On 10/19/2010 04:09 PM, Robin Fairbairns wrote:
> > Patrick Gundlach<patrick at gundla.ch>  wrote:
> >
> >>>> In luatex there are four explicit keywords:
> >>>>
> >>>> filll
> >>>> fill
> >>>> fil
> >>>
> >>> That makes luatex incompatible with tex/pdftex/xetex because "l" is a separate keyword that can be separated by spaces:
> >>>
> >>> a\hskip 0pt plus 1fIl L la
> >>> \bye
> >>
> >> I wonder how many users depend on/use that feature.
> >
> > trip test?
> 
> Yes. Then again, that fails in myriad small ways, nor does it test
> any of the new code, so I have not paid any attention to it so far.
> The trip test failures are something to evaluate in the step from
> luatex 0.9 to 1.0.

well, quite.  there's quite a lot of things don't get tested by trip.

(and you omitted my suggestion that tex's fi[l*] behaviour is hardly
something to yearn after)

> In general though, to anybody who wants all of the idiosyncrasies
> of TeX82 to be retained a 100%:
> 
>   luatex is not for you.

there are no doubt such people, who're still hankering after luatex, but
it's hard to see how one might satisfy them.


More information about the luatex mailing list