World: r3wp

all dialect driven

I think what they should have done here, is lift ordinary keyboard 
commands. You would simply say the key names, along with a few extra 
commands, like "again". This way you don't have to remember an entirely 
separate command set.

This is also a problem in MacOSX, because the command set is very 
limited and nearly useless.

R3 function for creating objects with default values. Neat!
      object [a: b: c:]
      == make object! [
            a: none
            b: none
            c: none
      ]

... and slow!


>> do [st: now/precise loop 10000 [object [a: b: c:]] et: now/precise]
== 11-Jul-2009/23:55:57.406+8:00

>> difference et st
== 0:00:04.36


>> do [st: now/precise loop 10000 [make object! [a: #[none] b: #[none] 
c:#[none]
]] et: now/precise]
== 11-Jul-2009/23:56:13.578+8:00

>> difference et st
== 0:00:00.032

That's functions for you :-)

well, how often do you create 10K objects at once? :-)

I find it a little strange that it is so slow:

>> source object
object: make function! [[
    "Defines a unique object."
    blk [block!] "Object words and values."
][
    make object! append blk none
]]

P: well, how often do you create 10K objects at once? :-)
frequently

It is slow because he reuses the same spec block over and over, which 
means that with every loop iteration another #[none] is added to 
the spec block. This means that the spec block has to be expanded 
many times, with reallocation overhead.


The OBJECT function was meant to be used with specs that aren't reused 
- it's a convenience function for high-level objects, which may be 
replaced by a change to MAKE object! to have an implicit none on 
the end of the spec instead of needing to add an actual one.

Why not just have it use JOIN internally, rather than APPEND?

Append is native too, no?

But APPEND changes the spec block, which caused the speed issue Peter 
pointed out.

At this point, I don't see the value in the OBJECT func myself.

Because of the speed issue?

No, that's incidental.

Yes, looking at object, there are two problems:

  1. it is slow .... though as Gregg says, that looks like it can be 
  fixed by changing its pith to:
                       make object! join blk none

  2, it is unnecessary. I never realised until now that this works 
  fine to set default values:
        make object! [a: b: c: d: none]

Sunanda, the OBJECT function was one of the language usability tweaks 
we made while writing the R3 GUI code. It is likely that the functionality 
of OBJECT will be added to the MAKE object! handler, at which point 
OBJECT will just be another word for the CONTEXT function. In the 
meanwhile we have already seen the value of it in the GUI source.

Sub-panels do no longer work with the demo.
The error messages is

** Script error: append2 has no value

** Where: repeat do switch applier apply if foreach if do-face if 
foreach do-tr

ggers switch-panel view-sub-panel do switch applier apply if foreach 
if do-face

if actor all do-style all until do-event do-event do-event either 
applier wake-

p loop applier wait do-events if view catch either either applier 
do try demo
** Near: repeat n 60 [
    append2 blk 'label ajoin ["Field " n]
    ...

** Note: use WHY? for more about this error

Yeah, I saw that too, I tracked it down to the APPEND2 command which 
is no longer used in the GUI.R

Instead, it looks like the REPEND command took it's place, but was 
not changed in the DEMO.R

This work-around should fix the demo for sub-panels:


Run R3 as you normally would then type this command:  APPEND2: :REPEND

Then you can run the DEMO.  It is just a quick fix until the DEMO.R 
script is changed.

That may not work the way you think since REPEND (which has 2 arguments) 
didn't actually take the place of APPEND2 (3 arguments).

Ah, Carl is preparing to split user and system context. What will 
that mean? Respectively - what is the purpose of such change, to 
have those isolated? Better modularisation?

BrianH,  I thought it was a straigt replacement, but looking closer 
at the code, i do see the change.  Disregard my work-around suggestion.

Patrick, you can replace APPEND2 with APPEND APPEND.

It's a global search-and-replace in the source rather than a simple 
assignment, but it works.

Thanks Brian,  I had changed the APPEND2 to REPEND with the proper 
[ ] which seems to do the same thing as APPEND APPEND.
				repend blk ['label ajoin ["Field " n]]
				repend blk ['field form first+ fields]
				repend blk ['button "Change"]
Thanks!

REPEND has more overhead than chained APPEND.

Didn't know that!!!

REPEND creates an intermediate block - it doesn't REDUCE in place 
like APPLY, since it is mezzanine.

Brian, If I wanted APPEND2 to take on the value of APPEND APPEND, 
how would I do that?
I tried APPEND2:  [append append] but it is not what I wanted.
I also tried APPEND2: 'append 'append but still not right?

What is the proper way to have APPEND2 take on the APPEND APPEND 
so that when you run DEMO, it resolves correctly?

Make a function named append2:

append2: func [blk val1 val2] [append/only append/only blk :val1 
:val2]

Oh so you first need to take a look at the append function like HELP 
APPEND,

then define another function to pass along the same parms etc with 
the APPEND function in it.
so then you could also say this:

append2: func [blk val1 val2] [append append blk :val1 :val2]    
 (without the /only refinement)
I mean to say is that the above could also work,  right?

Yes, but I think you want the /only refinement in this case.

got it!   -- Thank you

You don't need to replicate the type spec of the arguments since 
this is just a helper wrapper function.

Can R3 be run in CGI mode? There seems to be -c command line switch, 
but no CGI helper functions, nor system/options/cgi path ....

IIRC Carl posted some short notice on that, but can't find it. I 
would like to start testing it, e.g. the speed of page build in comparison 
to R2.

Pekr:

http://www.rebol.net/r3blogs/0182.html

Thanks, just found it too :-)

Disappointed by outcome of latest blog. Binary conversion issues 
are still a catastrophe. We were supposed to get it right and easy, 
ok?

Writing a reaction ...

Using issue! to encode binaries is a disaster. We need better conversion 
functions, not syntax tricks.

then issue is completly usless type - remove it

It's useful as a syntax type. We are thinking of making issue! a 
unique and immutable string as well, with supposed modifications 
generating a new issue!. It would be like words without binding or 
as many syntax limitations.

I put my reasoning into blog post ...

hmm, Carl dismissed #1113. I wonder if 'bind is used so often, that 
he was afraid of overloading it, and introduced new mezzanine instead? 
Was that really necessary? Another one to learn ...

Apparently he considers initializaton to be out-of-scope for BIND. 
Oh well, it might have been cleaner - we'll see.

BrianH: is there a bug ticket for?:

>> to-issue #{0001}
== #??


... I did not understand from your blog post, if it is supposed to 
be fixed already, so the ticket exists, or not ....

I mistook the problem you were proposing for one that had already 
been fixed :)

yes, but is above a bug? Should be reported, no? :-)