parsing comments in scripts

[Robert Brenstein]

>#comment
>text --comment
>text "more text -- still more"
>text "more text" --comment
>text "more text" text "-- text"
>text "more text" text "still more" --comment
>(did I leave any out)

plenty :) although it may be that only MetaCard allows to use both 
quoting modes.

from the above set

text "more text -- still more" --comment

same with hash

text #comment
text "more text # still more"
text "more text" #comment
text "more text" text "# text"
text "more text" text "still more" #comment
text "more text # still more" #comment

mixed instances

text "more text # still more" --comment
text "more text -- still more" #comment

I also use the following combinations to mark debugging code

text "text" -- #comment
text "text" #-- comment

You could also consider other combo cases

text "more text --- still more" #comment
text "more text -- still more -- still more" #comment
text "more text -- # -- still more" #comment
text "more text ### still more" #comment

In these cases it matters whether you first check for # or -- so you 
should really check both cases and see which has has lower offset.

In general, I think you need to use the offset function for each line 
and check for either -- or # and count whether there is an even 
number of double-quotes before it (setting for example itemdelim to 
quote and get the item count). If count is odd, you are within 
literal and should continue. If count is even, then you hit a 
beginning of comment. I think this approach will work with all cases.

Robert Brenstein