Hello,

If I have a test file

----------------------------------------------------------------------

* test 1

<2023-01-05 Thu 19:00-19:20>

----------------------------------------------------------------------


and then convert to an .ics file using C-c C-e c f

I get the attached test.ics file, but copied here too:

BEGIN:VCALENDAR
VERSION:2.0
X-WR-CALNAME:test
PRODID//Stephen J. Eglen//Emacs with Org mode//EN
X-WR-TIMEZONE:GMT
X-WR-CALDESC:
CALSCALE:GREGORIAN
BEGIN:VEVENT ***
DTSTAMP:20230105T221431Z ***
UID:TS1-81DE7E9C-0385-49CE-8528-C2FD51372CB4 ***
DTSTART:20230105T190000Z ***
DTEND:20230105T192000Z ***
SUMMARY:test 1 ***
DESCRIPTION:<2023-01-05 Thu 19:00>–<2023-01-05 Thu 19:20> ***
CATEGORIES:test ***
END:VEVENT
END:VCALENDAR

The lines from BEGIN:VEVENT to CATEGORIES:test inclusive have CR (^M) as well
as LF (^J) as newline [in the text above, I've converted the ^M to ***
to see them eaer.  Is that intended?  Uploading the file to
https://icalendar.org/validator.html gives me the error

> Lines not delimited by CRLF sequence near line # 1
> Reference: RFC 5545 3.1. Content Lines

Nicolas (cc'ed) has confirmed that it seems to be a problem.  Before I
work slowly on a suggested patch, does anyone have an idea on the best
way forward?  Is it possible to encode the file as "DOS" ,using
something like set-buffer-file-coding-system?  SOme lines do however
have \r and \n see e.g. org-icalendar-fold-string which inserts both \r
and \n




Best wishes, Stephen