On Fri, Oct 16, 2026 at 12:58 PM David G. Johnston < [email protected]> wrote:
> On Fri, Oct 16, 2026 at 6:32 AM Matemática A3K <[email protected]> > wrote: > >> Given this, I find this change to better align the row with the others, >> so I am attaching a patch. >> > > You forgot to attempt a final build before submitting since you are > missing the closing literal tag on the synopsis. > Yes, indeed, thanks, make check spotted it > You also left an extra } there. > > Once you removed subscript the wording of the existing text, the "...can > be given in two forms:", is no longer is correct - the accessor requires > the brackets where the forms of subscript did not. > After fixing, this "reads right": [{<literal><replaceable>index</replaceable></literal> | <literal><replaceable>start_index</replaceable> to <replaceable>end_index</replaceable></literal>}, ...] but it is not clear that the curly braces and the "or" are not part of the syntax, while the brackets are. How can this be stated? Is this what you referred with "the accessor requires the brackets where the forms of subscript did not"? > David J. > > P.S. I do find the chosen background highlighting in light mode for > literal, light grey on white, to be insufficiently contrasted. Though in > this specific case the bold font face from replaceable would seem to more > than compensate for the deficiencies of literal. Though I do notice we do > seem to have different behaviors on these when they appear inside and > outside of a table, which seems problematic but, also, maybe necessary. In > short, research and changes in this area are possible though likely > nontrivial to derive and get committed. > Thanks for the advice, I won't pursue this ATM as I find it is not clear the costs / benefits ratio.
