Documentation?

[Pauly]

Hi,
Is there a reference document somewhere that I'm missing (or more likely can't find! :-))?
The reason I ask is that there seems not to be a 'Language reference manual' or whatever you want to call it with Nettalk, so the syntax and parameters of keywords is questionable.

I've bougt the book, which gives some examples of specific scenarios, however I'm hungry for some more effective 'hunt and peck' type documentation.

Is there anything available?

Thanks.
Pauly 

[Bruce]

the best documentation is, of course, the source code.
But given that I don't know what you have found, it's hard to comment on what you haven't found.

Cheers
Bruce

[Pauly]

Hmmm.. that reminds me of the guys that say 'I do not need to comment my code as it is self documenting' - Two years later they are wondering what the hell they were trying to achieve in their own code! :-)

But seriously folks.. a small but relevant example... p_web.message() How do I find out what the parameters valid values, intentions, and limitations are - For that matter - how do I find out if it's even what I think it is?

I hope you get what I'm, saying - not trashing anything, just wanting information.

Thanks
Pauly

 

[Bruce]

sure I understand Pauly.
There certainly could be more in the way of technical docs - but frankly I only have the 4 pairs of fingers, and you lot keep me sooo busy with new features....

I wasn't being too flippant with regard to the source. netweb.inc and netweb.clw contain the source code, and the best way of figuring out what a method id good for is to
a) check out the source for it and
b) check to source to see how it is used.

Message is not a function I would recommend you to use - it probably doesn't do what you want.

Remember always, there's a request/response model. The browser makes a request, you send back a response.
It only _starts_ processing the response when a complete reply has been received. So you don't "interact" with the browser. you just give it a response, stand back, and wait for the next request.

It takes a lot of getting used to, because you're used to code being very interactive - do this, wait for that input, do some more, wait for some more input and so on. Whereas here you have to keep thinking - what's the request? what's the response?

Also, you can be fooled - especially while you're learning - by the familiarity of what's in front of you. Hey, it's a menu, browse, form, report - so you start thinking it's like a normal Windows program. But it's not, and in some cases there's a huge amount of work that goes into mimicking a "simple" interaction. The further you step outside the box, without understanding the box, the more likely you are to get very confused as to what will happen.

>> 'I do not need to comment my code as it is self documenting' - Two years later they are wondering what the hell they were trying to achieve in their own code! :-)

Comments in the code can be important in helping you understand what's going on - and certainly where it's being "clever" then a comment is warranted. But the code is what the computer will run, so writing clean code, and more importantly learning to read code, is a big step forward in understanding what each bit does.

Cheers
Bruce