World: r3wp

It's at the bottom of the page as a file attachment

BTW, nice use of Deki wiki.

It looks pretty but is slower than docbase :)

and there's no REBOL formatting .. so I use ruby or whatever ...

Runs really fast here. Where's the server?

I think EC2

It was a free hosting offer I picked up a couple of years ago ...

That is Mindtouch were offering free wikis to people limited to 200Mb 
I think

Of course what I really like about it is that there is a REST based 
API that I can use with REBOL.

Carl - guys found out, that we have just one shared buffer - why? 
You have to read everything out during read phase, or you can get 
corrupted data ....

pekr, what protocols are full duplex?

none ... but wasn't it you, after all, who just wondered, why the 
in-out buffer is shared?

No .. I just pointed it out.

And wondering if that imposes a lot of GC ...

It's one of those things that's in the "balance".

It turns out, I think you can allocate separate buffers, but you 
must know how to do it... and I've not tried it yet, so have not 
mentioned it either.

But, I did give full duplex consideration during implementation. 
 I think more modern IM systems benefit from it.

Also, it turns out that one must be careful when estimating GC overhead. 
Sometimes it is more efficient, becuase otherwise you must add extra 
code to manage memory usage... that is constantly being run. You 
know what I mean?

It wouldn't be a big change to the schemes if that gets implemented 
right ?

Right. No change to schemes.

lunch time, bbl

I think in my ftp scheme I access port/data directly so I can stream 
to the file system .. so it would require a tiny change if there 
were two sep buffers

Graham, instead of
    write imap-port [ dialect stuff ]
    wait [ imap-port timeout ]
could you try:
    write imap-port [ dialect stuff ]
    read imap-port
    wait imap-port
I think that should work.

Graham, before looking at your new schemes, I decided to review some 
of DocBase. But... I must say that I have a difficult time finding 
things there... I'm not sure if the user who was reorganizing it 
got a chance to finish the updates.

On http://www.rebol.net/wiki/Table_Of_Contents-- I've added some 
links to ports and schemes. Did I miss any?

Hard to say ...

If I find anymore I'll update the wiki 
Andreas, going to try that now.

You did have an example somewhere of doing a large file transfer 
between client/server

http://www.rebol.net/wiki/TCP_Port_Examples

http://www.rebol.net/wiki/Ports:_Synchronous_and_Asynchronous_Operations
may also make sense to be linked up

Ports:_Synchronous_and_Asynchronous_Operations -- that is

A: do you have a wiki account?

Not yet

OK, created an account. I'll simple add those links myself, then

I tossed a few up just now. Go ahead and add or edit as needed.

looks like we side tracked into a docbase update ...

Already debugging the IMAP thingie

Not a sidetrack. See my note. This is an opportunity to check the 
docs against the code.

Ok, fixed the IMAP proto so that no timeout is needed anymore. Stumbled 
over what seems to be a bug in R3's USE, though:

>> foo: func [] [use [] [return 42] 0]
>> foo
== 0

Looks like this bug is what was causing me grief ...

Reported as bug#1436

Updated to 0.0.4 
working http://rebol.wik.is/Rebol3/Schemes/Imap4

There's some optimizing of parsing the input and checking for incomplete 
data returned on a read to be done.

I've updated it to v 0.0.5 which hopefully checks now for incomplete 
read data, and factors out some of the parsing

Back to separate buffers - just a note. I think that we eventually 
don't need them. But we should be carefull to user errors. Because 
with one shared buffer, if you write to it, before you read-out all 
incoming data, you just destroyed a data. And what if you don't need 
to act on the data yet? You will temporarily store it in some other 
storage (buffer) anyway. But I never wrote a scheme, so maybe this 
is just plain theory, and in fact having one buffer will not cause 
any problems to anyone ...

well, I think you normally read the data and then decide what to 
do with it.  So, you have saved it somewhere ...

Anyway I haven't had any issues so far .. and users unless they are 
writing schemes won't either.

I mean - what if data comes, but you don't want/need to react on 
it, but you need to write? You have to read-out the buffer first, 
before you write, no? This constant small chunk read/wirte could 
cause unnecessary overhead, or not?

Anyway - good to know, there is still some secret mode to reveal 
:-) So maybe one day Carl describes how to utilise two separate buffers 
:-)