>>>> No, there are cases where this fails. We went thru this before. >>> >>> Only if you have data with > 69 spaces in a row that needs to be >>> preserved. >> >> More generally, anytime the fold occurs where space characters >> follow. > > No, b/c you wouldn't fold there.
A human wouldn't, but an automated folding solution sure could. >> But why are you arguing for this? - the double-backslash approach >> works great for when indents are desired. > > Let's agree that it works :) > > I'm arguing that *if* we are to define two solutions, we should use > this one. (And as I explained, my preference is this one over the > double-backslash solution, but I accept the WG consensus for the > double-backslash solution) Okay, now I understand your position, but I don't agree with it and, as mentioned before, how to support indents was thoroughly debated before. I can't recall if you were sailing then, but I'm guessing you were which is why you're raising this now. >> My interest in this thread was/is only to cover the common case >> when there are no indents, and the continuation line always begins >> on column 1, in which case the 2nd backslash is unneeded and >> somewhat counterintuitive. > > But your continuation line will not start at column 1 (since artwork > is indented with 3 spaces by default). So this solution doesn't > really work. Not in the XML. If rfcstrip/xym wish to extract from plain-text, then they should detect the artwork-level indent/outdent and apply that adjustment to the entire content prior to running the unfolding logic. Kent // contributor
_______________________________________________ netmod mailing list [email protected] https://www.ietf.org/mailman/listinfo/netmod
