UTF-8: cut: the delimiter must be a single character

[doersino]

UTF-8: cut: the delimiter must be a single character

[doersino]

Also a good option, but I think I'll just leave it as a space for now.

No worries about being difficult – I wrote a static site generator as a Bash script, so I'm practically asking for it. 😉

[doersino]

Oops, I used that character under the assumption that it's among those characters represented by a single byte in any ASCII/Unicode-based system – clearly I need to brush up on my ASCII tables! 😉

Interesting, BSD/macOS cut seems to deal with multi-byte delimiters just fine.

Thank you for doing the investigation work for me! I've chosen a different character now – does that fix the issue for you?

[kseistrup]

Thanks, that works (I'm a little puzzled that the ‘#’ isn't parsed as a comment character, though) — thanks for fixing it so soon!

[doersino]

That's true, it could be iffy in other/older versions of Bash. The separator should be very unlikely to occur in filenames, otherwise I would've just gone down the CSV route and used a comma.

Come to think of it, a space seems like a good one.

[kseistrup]

Or perhaps just

| cut -d\# …    # escape the delimiter
| cut -d '#' …  # enclose the delimiter in single quotes

When reading the script and templates as a human, it can be difficult to distinguish a SPACE and a TAB at a glance.

Sorry for being so difficult.