Maybe it is all related to this comment?: >From: Deri James <[email protected]> >Subject: [bug #65716] [afmtodit] afmtodit.tables doesn't reflect movement of >characters out of Unicode PUA >Reply-To: [email protected] >Followup-To: gmane.comp.printing.groff.general >Date: Sun, 15 Mar 2026 08:20:58 -0400 (EDT) >Message-ID: <[email protected]> >Newsgroups: gmane.comp.printing.groff.bugs
>Follow-up Comment #12, bug #65716 (group groff): >The basic question is should we be relying solely on the UCPs in the Adobe >Glyph List (which appears to be set in aspic), or should we be patching it so >it matches current Unicode standard (which is used by troff). >It is this difference which has caused confusion for our users On 2026-03-15 Morten Bo Johansen wrote: > On 2026-03-15 Morten Bo Johansen wrote: > >> Well, probably not voodoo but very strange to me. Try this: >> >> echo "\\\f[U-S]\[uF8E8]\f" | groff -Tpdf | pdftotext - - >> >> And if you see the same as me, you'll see a white circle with an >> intersection symbol in it. > > ˝/ % echo '\uf8e8' > > > So that seems to be the correct symbol. Apparently pdftotext then somehow > translates a wrong symbol from groff into a correct one? > > Regards, > Morten > > > >
