Interesting wordOffset behavior

[Peter Haworth]

I'm finding that wordOffset treats comment chars as a word even if they
are immediately followed by another character.

For example:

wordOffset("--","repeat with x=1 to 100 --check something")

...returns 6.  I would have expected it to return zero since there is
no occurrence of "--" surrounded by spaces or quotes.

To add to the plot:

put word 6 of "repeat with x=1 to 100 --check something"

...returns "--check"

And:

wordOffset("--","repeat with x=1 to 100 -- check something")

...also returns 6 (there's a space between the comment chars and "check" in
this example).

It turns out that this is useful in the context of what I'm doing but
whenever I run into what I think is unexpected behavior like this, I'm wary
of using it in case it's actually a bug that might get fixed at some point
and break my code.

Am I being over cautious?

Pete
lcSQL Software <http://www.lcsql.com>