utf8 character replacement in terminfo - is it possible?

[Nikita Zlobin]

Hello.
I sent subscribtion request, but while it is in pgrogress...

I had first experiment, customizing terminfo file (xterm-256color,
which is used by many modern gtk terminals as well as kmscon by
default).

My goal was to replace corner characters to use rounded by defaults.
I dumped source with infocmp and changed acsc to this:

acsc=``aaffggiij╯k╮l╭m╰nnooppqqrrssttuuvvwwxxyyzz{{||}}~~,

Later i was told at unixforum.org, that only ascii is allowed. There is
link (thread is russian, but there are screenshots with what i could
achieve, including one with result of acsc modification):
https://unixforum.org/index.php?showtopic=142810

When i replaced these rounded corners with simple pluses '+', i got
'+'-styled corners in mc, as expected. However, when i looked terminfo
for some other terminal types (including Linux), i found it consisting
completey of '\code'-like sequences. As result, i tried following
sequences:
1. Simply draged example character from gucharmap to gtk3-based text
editor (i used scratch from elementaryos project). For top-left and
bottom-left it gives respectively '\u256d\u256d' and '\u2570\u2570'.
2. Another way was using octal representation in C:
'\342\225\255' (also copied from gucharmap).

Both codes gave same result, as direct character copy-pasting.

As last resort i'm thinking to create edited copy of some monospace
font with replaced characters, specially for use in terminals.