Bernd Fondermann wrote:
I don't think every single setter/getter method in Vysper needs javadoc.
They are typically generated automatically by any decent IDE, it's a no
brainer.
That's where we disagree.
If it's a no brainer, nobody needs it.
Well, it's just a placeholder When I use Eclipse (my IDE of choice, most
certainly the best, but, eh, I'm used to it like my old shoes), I
defoine my private fields, and when done for all of them, I just ask
Eclipse to generate all the getters/setters *with* javadoc. Then if I
need to add something in it, I don't have to add all the @ : they are
already there. As I said, an absolute no brainer, and it kills all
discussion with old fart like me who requires Javadoc all over the code ;)
More seriously, I think it's way more important to add some javadoc on
the fields themselves : nothing is more painfull than a field with no
javadoc and a bad selected name :/ However, I can live with a field name
'next' or 'previous' without any Javadoc, if they are used for a cursor
implementation, for instance.
It's all about balance, I think ...
If there is generated javadoc, you won't have an indication where
auto-generated doc might not be sufficient.
But you are certainly right on the fact that Javadoc is more important
for MINA, as it's more important for commons-XXX than anything else.
In Vysper's commit log history you will notice one or more doc'athons
where a lot of doc is added. I will keep doing that. It's fun.
I don't think that only one way leads to Rome :)
Yeah, that's right, we won't get into fist fights about javadoc.
I'm not that type of guy! ;-)
I like the idea of a a fight in redcurrent jam, hmm, well, ok, not
applicable to coders, I think :)
--
--
cordialement, regards,
Emmanuel Lécharny
www.iktek.com
directory.apache.org
