On Thu, Dec 10, 2020 at 10:44:34PM +0100, Christian Weisgerber wrote:
> Stefan Sperling:
> 
> > I plan to switch from fparseln(3) to getline(3) eventually, which
> > will keep \n intact. (This patch was extracted out of a larger
> > patch that converts tog to getline(3), which isn't finished yet.)
> 
> I see.  ok

Thanks :) I have posted my getline patch now, it depends on this fix.

> > > It made me curious how got & tog deal with terminal control sequences
> > > in commit messages and blobs.
> > > * tog escapes them (where??), but gets confused about the line length.
> > 
> > Likely happens when tog converts to wide charactors so ncurses can
> > display UTF-8 text? Non-UTF-8 data gets run through strvis in mbs2ws().
> 
> An ESC character is valid UTF-8.  I can't find any call to strvis()
> or the like.

I misremembered the facts. I meant got_mbsavis() which detects invalid
UTF-8 and replaces such bytes with '?' (recall that on OpenBSD UTF-8 is
the only available multi-byte encoding; this might be different on FreeBSD).

What specific problem are you looking at?