\k does some clever-ish things to grab its argument to make it an anchor target. This is very useful for anchoring figures in glossaries.
But it looks like sequences such as \k The \+nd Lord\+nd*\k* are going to be broken for the foreseeable future, with the \+nd* being read as \+nd * .
The work-around at the moment is to use \+k rather than \k, which does not have the magic side-effects I.e. \+k The \+nd Lord\*nd*\+k*, and accept that such sequences will not be anchor points.
A python expert can no doubt write a non-greedy changes.txt line to automate this, looking for \ inside \k...\k* and making the \ks into \+k
A possible solution might be to make the input \+nd ... \+nd * an equivalent of \+nd ... \+nd* I.e. nesting a style inside itself checks a few more tokens ahead to see if there's a * coming.
\k
does some clever-ish things to grab its argument to make it an anchor target. This is very useful for anchoring figures in glossaries. But it looks like sequences such as\k The \+nd Lord\+nd*\k*
are going to be broken for the foreseeable future, with the\+nd*
being read as\+nd *
.The work-around at the moment is to use
\+k
rather than\k
, which does not have the magic side-effects I.e.\+k The \+nd Lord\*nd*\+k*
, and accept that such sequences will not be anchor points. A python expert can no doubt write a non-greedy changes.txt line to automate this, looking for \ inside\k...\k*
and making the\k
s into\+k
A possible solution might be to make the input
\+nd ... \+nd *
an equivalent of\+nd ... \+nd*
I.e. nesting a style inside itself checks a few more tokens ahead to see if there's a * coming.