>Dan made a Perl comment-stripper, that would run online, there were still
>some bugs though.
>And Thomas Loo made a script one, very cool called ESC, I'm not sure where
>it is online, his email was:
>[EMAIL PROTECTED]
>This is exactly why we are ending all lines with semi-colons in the script.

I like the idea of a comment stripper. We have an option on how we want to
use it.
1) We could either setup two directories under the CVS (src for commented
and js for uncommented) and be forced to strip out the comments every time
we make a CVS update.

or my preferred way:
2) We could maintain just the src CVS folder fully commented and strip out
the comments only when we create a new official release package.



>> 2) You have to build a doc compiler. This is a project of its own. I
don't
>> think we have the resources for that.
>Oxygen is really cool, works in windows, or Linux. It's touted for C C++,
>but with a little tweaking will work ok with Javascript.
>sample output:
>http://www.richardinfo.com/examples/Richardinfo_Developer_Docs/
>
>http://www.stack.nl/~dimitri/doxygen/index.html.

>Cheers,
>Richard.

I still think this is going to be a project of its own. Creating a comment
stripper is one thing. Creating a doc compiler is several orders of
magnitude harder. What exactly do you mean by a little tweak?

Without knowing a lot about Qxygen, for it to work for us it would have to
be able to do the following or we will have to "tweak it" to do the
following:

It will have to recognize entities such as constructors, deconstructions,
fields, methods/functions ect. For it to recognize these entities, we will
have to create strict language definitions, to which we will have to conform
to in order for the doc compiler to work. Creating a language definition for
DynAPI will be yet an other project of its own, we might as well reformulate
DynAPI in XML or something. Then of course we will want tools that check
weather we conform to the lang specs as we write the code, otherwise this
whole exercise is a waste.

I don't want to burst any bubbles, I really like the idea of a doc compiler.
I just don't see how we are going to create one without doing all of the
above mentioned work. Do we have man power to do it? If we do, let's do it.
I'm all for it.

NanoFace=;^)

ps an after thought:
I suppose one way to get around all this language definitions is to declare
the entities in the comment block i.e.:

/**Constructor
This constructor instantiates widget x
*/

Then use the declaration as it's face value in oxygen or whatever doc
compiler.
But we will still need to confirm to some basic rules. ie: the first non
comment line after an entity declaration is the entity block header. Which
is then used as the entity identifier syntax in the docs.



_______________________________________________
Dynapi-Dev mailing list
[EMAIL PROTECTED]
http://www.mail-archive.com/[email protected]/

Reply via email to