<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, 19 May 2024 at 11:10, Zoltan Somogyi <<a href="mailto:zoltan.somogyi@runbox.com">zoltan.somogyi@runbox.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex">The attached .err file has two error messages, on lines 8 and 9,<br>
about the same predicate. Both messages refer to "predicate `exp1'/0".<br>
Both color "`exp1'/0" as the subject of the error message. One includes<br>
the word "predicate" in that coloring, the other does not.<br>
Both approaches are justifiable, but in this case, I don't think<br>
it is a good idea to have different error messages differ<br>
in what they color: we should either color "predicate"<br>
(and of course "function" in other cases") together with the symname<br>
and arity, or not, in all diagnostics, or at least all diagnostics<br>
where the func vs pred distinction is itself the error being<br>
reported.<br>
<br>
I slightly prefer the approach of *not* coloring "predicate"<br>
in such cases, simply to reduce the resemblance between<br>
our diagnostics and ransom notes. However, I would like to know<br>
whether anyone objects to this approach, because adopting<br>
the approach requires changing the interface of the utility function<br>
describe_one_pred_name, </blockquote><div><br></div><div>No objections from me.</div><div><br></div><div>Julien.</div></div></div>