![](https://secure.gravatar.com/avatar/46562c7d27388490d62e8d7a5bc960dd.jpg?s=120&d=mm&r=g)
On 4/27/2021 9:00 PM, Pablo Rodriguez wrote:
On 4/27/21 8:57 AM, Hans Hagen wrote:
[...] as mentioned in an earlier mail, after decades of utf in tex we should use the normal symbols instead ... you can disable collapsing with
\nohyphencollapsing
but then you need to enable the tlig font feature. In days when often texts is imported from elsewhere and editors can show these dashes we need to adapt
Not sure I understand your explanation.
I have been using UTF-8 as charset for my documents since 2002 (otherwise polytonic Greek was unreadable for me).
It was also more readable to use real character for em- and en-dashes than three or two hyphens.
It took me a while since I accidentally discovered a document with a wrong line break between a real em-dash and a point followed by a footnote number.
So my question is what \hccode stands for?
From luametatex.pdf in the distribution, I see that this is a LuaTeX primitive, but luatex.pdf doesn’t mention it. sure, it's a luametatex primitive, but that is true for some more, like
hyphencharcode good old tex only has \exhyphenchar, in luametatex we can tag any char as such in traditional tex th ehyphenation, ligature building and kerning wre integrated, so handling - -- --- ---- ---------- is kind of complex in luatex these stages are split but there's still messy logic wrt successive hyphens in luametatex the hyphenation machinery has all kind of extra controls and the hccode mechanism was introduced as side effect of a feedback mechanism that also can handle compound words better (althoiugh one can of course dispute the 'better' part) the detailed control of how characters are treated in various stages 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 -----------------------------------------------------------------