Philip Hazel wrote:
That's because no_anything is the same as anything=false and the index
entry is for anything. Perhaps I should add this index entry:
no_<anything> see <anything>
Hey, I never realized that! I've never actually read the spec from
cover to cover, so I'd never seen section 6.7 before.
Hmm. I wonder if it's possible to do that in DocBook? Maybe not. The
current docs have a dozen or so "see xxx" index entries, and none of
these are present in the new world. I will put this on the list of
infelicities to investigate. Also, it would be nice in the HTML index to
be able to hyperlink to the various letters, I feel.
You could mark the "no_" and "not_" that appear in 6.7 as index terms,
so at least they'd refer back to that section. It's not quite as clear
as the index entry you propose, but at least it puts a no_ entry in the
index, which is better than nothing (or no_thing, if you prefer :) ).
Now that I've read 6.7, I think I have a bug to report in it. It states
that "yes", "no", "true" and "false" are the only valid values for a
boolean option. This is repeated in the documentation of the more
option in section 15, and I'm guessing it'll be repeated in other
boolean/expanded options. I thought that "1" and "0" were also valid
values?
I wrote a script to put them in for the spec (see HowItsDone.txt), and
didn't bother for the filter. I guess I could write another script...
I'd appreciate it if you did. Of course, I'm already on record as
thinking you should combine the two docs, which would eliminate the need
for that script... :)
- Marc
--
## List details at http://www.exim.org/mailman/listinfo/exim-dev Exim details
at http://www.exim.org/ ##
