AltME groups: search

Serialised function! not loadable 

in Rebol/View 2.7.5.3.1 on WinXP

>> load {#[function! [][]]}
** Syntax Error: Invalid construct -- #[
** Near: (line 1) #[function! [][]]

>> load/all {#[function! [][]]}
** Syntax Error: Invalid construct -- #[
** Near: (line 1) #[function! [][]]

The above are loadable on View 1.3.2

It is the responsibility of the mail client.  You control the mail 
client.  If you construct a BCC field in your header, Rebol will 
ignore that when using send.

Ladislav's PIF function appears to be functionally similar to the 
new case construct

The old domain still seems to work

http://www.fm.vslib.cz/%7Eladislav/rebol/pif.r

THIRD *does* work for words, as long as you use CONSTRUCT to rebuild 
the object.

To illustrate for those who may not know what the issue is with third 
/ make

>> c: make object! [a: 'print b: "hi"]
>> ? c
C is an object of value:
   a               word!     print
   b               string!   "hi"

>> spec: third c
== [a: print b: "hi"]

;Using make object! fails
>> e: make object! spec
hi
** Script Error: a needs a value
** Near: a: print b: "hi"

;Contruct works
>> f: construct spec
>> ? f
F is an object of value:
   a               word!     print
   b               string!   "hi"

rebol does not recognise filetype .... so it uses application/octec-stream 
.... too many gotchas here, maybe it would be better to construct 
complete headers. Will look into some alternative attachment sending 
scripts on rebol.org ....

If I understand your question, Charles, you may be looking for construct/with......It's 
a way of using an existing object as a template for a new one, eg:
 my-obj1: make object! [a: 99 b: 2 c: "string"]
my-obj2: construct/with  [b: 9999] my-obj

a: make object! [b: 445 c: 444]
 b: make object! [b: 999]
b: construct/with third a  b  ;; set b to a's values and fields

>> blk: [ delete none ]
== [delete none]
>> type? blk/1
== word!
>> type? blk/2
== word!  
;
; none is not of the none! type, unless it's been evaluated. 
; none is so-called indirect value in the REBOL/CORE doc at
; http://www.rebol.com/docs/core23/rebolcore-4.html
;
; In http://www.rebol.net/r3blogs/0034.html, 
; there is a so-called "Scant Evaluation"
;
>> obj: construct [ n: none d: delete ]
>> type? obj/d
== word!
>> type? obj/n
== none! 
; 
; WHY obj/n IS NOT OF THE WORD! TYPE ???
; The "Scant Evaluation" should not evaluate none here.

in headers (and in the construct function) both word lookup and function 
evaluation are disabled, but set-word (assignment) is not.

I see what you mean. The actuality seems to contradict those documents.

>> obj: construct [ n: none d: delete ]
>> probe obj
make object! [
    n: none
    d: 'delete
]


It seems that only potentially dangerous evaluation is prevented, 
and not all evaluation.


Scant Evaluation: A minimal form of evaluation used for headers and 
other data blocks that do not allow any level of deep evaluation.
  Perhaps the evaluation of none is not considered "deep."

Jerry, I found this:


The CONSTRUCT function will perform evaluation on the words TRUE, 
FALSE, NONE, ON, and OFF to produce their expected values. Literal 
words and paths will also be evaluated to produce their respective 
words and paths. For example:


    obj: construct [
        a: true
        b: none
        c: 'word
    ]


The obj/a value would be logical TRUE, obj/b would be NONE, and obj/c 
would be WORD.

file:///C:/SDK/docs/changes.html

>> obj: construct [ a: on b: yes ]
>> probe obj
make object! [
    a: true
    b: 'yes
]

Nothing should be evaluated in CONSTRUCT, except SET-WORD!, which 
is an inconsistency.


NONE, TRUE, FALSE, ON, and OFF are not SET-WORD!, but they are evaluated 
in CONSTRUCT, which is another inconsistency in an inconsistency.


YES and NO are not evaluated in CONSTRUCT, which is yet another inconsistency 
in another inconsistency in an inconsistency.

Inconsistency is no good.

Jerry, this is a "feature" of construct. the words NONE, TRUE and 
FALSE are converted to the respective values. it is done to allow 
construct to work correctly when /all is not used with mold.

in general, it is much better to use mold/all instead, but i know 
construct can give you headaches if you really want to have a word 
in there.

Ah yes, there is a 24 hour cycle problem with my approach. I could 
construct a more unique decimal taking year, month and day into account. 
Hm, random/seed returns a value, if called with a decimal!? Called 
with an integer, it doesn't. A reason for this?
>> random/seed 100
>> random/seed 100.0
== 41.0

Yes, random/seed with a decimal doesn't use the seed:

>> loop 10 [random/seed 1 random/seed to decimal! now/time/precise 
prin [random 100 ""]]                                            
                           
52 52 52 52 52 52 52 52 52 52

So I need to construct an integer from the time.

Yes, valid points, but it's not what I described at first.

If I should construct a random password given the rules, my output 
will land in a pool of 60 ** 8 possible passwords. I don't have to 
actual do it. The statement holds anyway.

If a computer should construct a random password given the rules 
(using any deterministic computer and any algorithm, but only with 
a 32-bit integer input, as in the case of REBOL random/seed), the 
output will land i a pool of 2 ** 32 possible passwords at most.


Of course we can change the frame and get a better result from the 
computer, but then we change the 'experiment'. In general, I would 
say the pool from human thoughts and decisions is infinite. It's 
not from a deterministic computer. So we need true random input and 
true analog computing with infinite states, if we want our computers 
to be as good as our brains.

A 'bind question: how do I construct an object containing blocks 
without changing the blocks' contexts?  This bad:

>> id: #foo
== #foo
>> blk: [id]
== [id]
>> reduce blk
== [#foo]
>> ctx: context append/only [id: #bar bk:] blk
>> reduce ctx/bk
== [#bar]

This good:

>> id: #foo
== #foo
>> blk: [id]
== [id]
>> reduce blk
== [#foo]
>> ctx: magic-context append/only [id: #bar bk:] blk
>> reduce ctx/bk
== [#foo]

How to 'magic-context ?

I was hoping 'construct would do it, but alas not.

fun: func [spec body /local me][
	me: construct append copy/deep [
		clear-all: [set bind copy next next first self self none]
	] body	
	me/clear-all: does me/clear-all
	clear find spec /local
	func spec bind body me 
]

f: fun [/local c d ][ c: d: 1 probe self print [c d ] clear-all print 
[c d]]
f

so you can't use the construct function on the primary bock...

BACKPORTS! Hot off the presses! Get your new R3 functions, now here 
for R2! Available now, before the next release!

funct: make function! [
    [catch]
    "Defines a user function assuming all set-words are locals."

    spec [block!] "Help string (opt) followed by arg words (and opt type 
    and string)"
    body [block!] "The body block of the function"
    /local r ws wb a
][
    spec: copy/deep spec
    body: copy/deep body
    ws: make block! length? spec
    parse spec [any [

        set a [word! | refinement!] (insert tail ws to-word a) | skip
    ]]
    wb: copy ws
    parse body r: [any [
        set a set-word! (
            unless find wb a: to-word a [insert tail wb a]
        ) |
        hash! | into r | skip
    ]]
    unless empty? wb: difference ws wb [
        remove find wb 'local
        unless find spec /local [insert tail spec /local]
        insert tail spec wb
    ]
    throw-on-error [make function! spec body]
]

functor: make function! [
    [catch]

    "Defines a user function with all set-words collected into a persistent 
    object (self)."

    spec [block!] "Help string (opt) followed by arg words (and opt type 
    and string)"
    body [block!] "The body block of the function"
    /local r wb a

][ ; Note: Words in the spec override the bindings of the object 
words.
    wb: copy []
    parse body r: [any [
        set a set-word! (unless find wb a [insert tail wb a]) |
        hash! | into r | skip
    ]]
    remove find wb [self:]

    throw-on-error [make function! copy/deep spec  bind/copy body construct 
    wb]
]

paul, I never used construct. can it serve nested constructs also?

TomBon, here is the construct docs http://www.rebol.com/docs/changes.html#section-4.1

TomBon, here is the purpose of a construct:

>> ct: construct [int: (1 + 1)]
>> ct/int
== (1 + 1)
>> do ct/int
== 2
>>


Notice that it doesn't automatically evaluate /int as an object would 
do.

Next one: I need to construct a HREF="..." with some dynamic code.

you mea, you want to construct a block with ANY unbound to a context 
?

I don't know how the values evaluated are stacked by the VM. But 
i see the advantage of having a POP function.
We could easly create postfix functions.
e.g:
CONCAT: func [v][join pop v]

>> "a" concat "b"
=="ab"

All sort of new operators would be easy to construct

I'd like users to construct their own sql as well and send it to 
the server ... but I don't

when i construct ports with a spec block, i would rather prefer to 
use dialect (using delect) instead of passing a prototype of the 
spec object

There is another aspect of this. Questions on functions names probably 
occur too often to not answer them by a proper language construct. 

I was always thinking that something like a SELF word in functions 
would suffice. A function would then be able to envoke itself, and 
only itself, without worrying about it's "name", whatever that is.

Is this a bug?

>> o: construct [ a: true b: yes c: yes ]
>> probe o
make object! [
    a: true
    b: 'yes
    c: 'yes
]


http://www.rebolforum.com/index.cgi?f=printtopic&topicnumber=47&archiveflag=new

>> o: construct [ a: true b: yes c: on ]
>> probe o
make object! [
    a: true
    b: 'yes
    c: true
]

replied there...

I usually mold/all for such things...
    
o: construct [ a: true b: yes c: on ]
    
>> print mold/all o
#[object! [
     a: #[true]
     b: yes
     c: #[true]
]]

construct should not evaluate .. but it is

construct exiplictely looks for the words none, true, false, yes, 
no   and replaces them with their value equivalents.

probably because Carl forgot to include yes/no into the construct 
special case... its possible he did it on purpose.

this is r3

>> a: construct [ a: true b: on c: yes ]
== make object! [
    a: true
    b: true
    c: true
]

yep, it is still there, I use 2.7.7.
I use this function to test:
f: func ['w] [type? get in construct compose [t: (:w)] 't]
>> f yes
== word!
>> f off
== logic!

forum topic is "Careful when Construct an object" on http://rebolforum.com, 
above link goes another topic. (a bug on forum, topic numbers changing)

The yes and no keywords of R3's CONSTRUCT were added at my request. 
We'll have to see whether the enhanced function can be backported 
to R2 safely. There are other changes as well, mostly safety changes, 
and a CONSTRUCT/only option that turns all of the tricks off.

Bug#651 details the changes - yes and no keywords, and unset-to-none 
translation - and bug#687 explains CONSTRUCT/only.

We'll have to see if there is any R2 code that will be broken by 
the change; if so, it won't happen. As a rule, backwards-incompatible 
changes are only made to R2 when they can be proven to not harm legacy 
code. There have been a few exceptions, and at least one mistake 
that predated (and inspired) the rule, but the exceptions have all 
been added functionality that doesn't change old functionality that 
is relied on in code. If nothing relies on yes and no not being keywords, 
or unset! values not being translated to none, then we're cool. As 
a counter-example, CONSTRUCT/only could be backported without question 
because it's a new option.

This is the construct I saw which I don't like

either true [ tail s ][s ]

which I could  then replace with

 if/only true [ tail ] s

It's less verbose and possibly clearer than using the 'either construct

some people prefer this construct

all [
	find blk val
	append blk val
]

I know - but RebGUI will not explain me how to construct styles, 
no? Those thigs like init method description etc. IIRC there was 
some attempt to write it down, so, links apreciated :-)

then use /header refinement and construct header yourself ...

hi all !
i try to detect move events in a feel/over func.

But i receive a flow of move events in my face instead of only 2 
events  (1 when entering, 1 when leaving).

I precise: 
I did'nt put the option [all-over] in the window.

I don't use VID but i construct my own layout with a serie of make 
face.

Basically you have to construct the whole row as 3 separate faces 
and keep appending them to a much larger face.

I fetch the text asynchronously ie. I don't have the text at the 
time I construct the layout .. oh well...

layout [a: text b: button]
v1: context [a: b: none]
set bind copy next first v1 v1 reduce [a b]

now you can construct v2, v3, V4 using the same layout

; I need 100 rows like these both samples with unshared data access 
for each row.

; I have also tried with stylize/master (panel with) to construct/compose 
a master 

; row like this but doesn't succeed. any change to make this work?


list-func: func [] [ layout/tight [across x: text "x"  y: text "y" 
z: progress 0.5   w: box 40x20 effect [ draw arrow blue rotate  60 
]   ] ]

view layout [
	a: box 300x50
	b: box 300x50
	do [a/pane: list-func]
	do [b/pane: list-func]

 btn "set a" [a/data: [x/text: "1" y/text: "2"  z/data: 0.8 w/effect: 
 [ draw arrow red rotate  180 ]   ]  show a ]

 btn "set b" [b/data: [x/text: "1" y/text: "2"  z/data: 0.8 w/effect: 
 [ draw arrow red rotate  180 ]   ]  show b ]

 btn "set"   [x/text: "1" y/text: "2"  z/data: 0.8 w/effect: [ draw 
 arrow red rotate  180 ] show b  ]
]

quit

you're starting from some points i guess, and you want to construct 
a smouth curve with them...

This is not obvious with VID, because you might build such a layout 
in a special setup, but in the VID Extension Kit, you have face constructs. 
A face construct is simply a style that accepts a particular dialect 
as input. The dialect is parsed into VID which then in turn generates 
an internal face tree. This means you can build styles that do something 
like:

view layout [
	paper [
		size a4
		dpi 300
		at 20x20 image bay.jpg
	]
]


The point is that every time you feed the paper new data, using SET-FACE, 
it's re-rendered from scratch using VID.


You can also build very complicated list views that dramatically 
change appearance and number of faces, every time you feed it new 
data.

(you have to tell Parse what the alternatives are *before* using 
any cycle construct like TO, SOME, etc.)

I have done this
template: make object! [
i2: reduce [[vlan 0][disabled " "][name " "]]

i1: reduce [[i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2][i2]]

data: reduce [[i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1]]
]
record: make template []

 probe record/data/13/48  returns none ol
but >> probe record/data/13/48/vlan
** Script Error: Cannot use path on none! value

do I need to do somthing to make it construct deeper levels?

I think I need to learn more before I can use this to the effect 
I need.   I was expecting your construct to all me to do this:
foreach [key val] b1 [print b1/:val/1]

Thanks PeterWood. I like to think I am ok with the most basic concepts, 
so now I am trying to learn things that will help me some my real 
life probelms in a better way.  I use parse pretty much every day 
& always have a rebol console up on my work PC, but ANY SOME & OPT 
& |  I do not understand in context.  I understand them in abstract 
terms, but not how to apply them in conjuction with [] . I do understand 
your examples of some & any (these examples are usefull to me). skipping 
an un-known number of chars to get to the next match is the bit I 
find hard to understand how to construct, paticularly if it needs 
to be done in the context of a previous match.

[ and ] are a grouping construct.

I'm also working on very similar to your case right now. I don't 
know if its useful for you but here how I do (on Windows)


command: {csvde -u -f export.ldap -d "ou=myou" -r "(objectClass=user)" 
-s 10.1.31.2 -a "" "" -l "DN,sn,uid,l,givenName,telephoneNumber,mail"}

call/wait/console/shell/error command %export.err  ;export all users, 
bind annonymous

if 0 < get in info? %export.err 'size [print "error" editor %export.err 
halt]
lines: read/lines %export.ldap

;create an object from the first line (field names, order may differ 
from what you give in the batch)

ldap-object: construct append map-each v parse first content none 
[to-set-word v] 'none
foreach line lines [
	(
		set words-of o: make ldap-object []  parse/all line {,}
		append users: [] o
	)
] ;append all valid users as an object to a block
probe users

I hope it gives some idea.

>> parse ["something" #[datatype string!]] [string! datatype!]
** Syntax Error: Invalid construct -- #[

** Near: (line 1) parse ["something" #[datatype string!]] [string! 
datatype!]

eFishAnt, the {} nesting rules only apply to strings entered in {}, 
not "". They are also only a REBOL syntax trick. Strings that you 
construct or read from somewhere don't have any special escaping 
rules or syntax.

Assuming SRC: contains the source text, it seems to work too:

header-char: complement charset "^/:"
EOL2: rejoin [newline newline]
parse/all src [
	some [
		some [pos: #" " (change pos #"-") | header-char]
		#":" pos: newline (change/part pos " {" 1)
		[to EOL2 | to end] pos: (change pos "} ") skip skip
	]
]
probe construct to block! src

Gregg well ... i'm not against contributing but I think one of the 
good way is for lower ranked programmer to express their problems 
and try to construct a doc helped by gurus to answers their ask then 
if it's readable and understandable by me I assum anyone could understand 
it 

My point is once people end rading the doc they will say "yeah i 
understoud what it's all about so now lets play with parse :P" if 
that only goal can be achieve I think a big step will be done.

But we still have a problem with break not being able encapable in 
a block.
We can't construct reusable rules with such logic.

I can have a look, but the purpose of NOT is not to have better perfs 
than complemented charset, but to allow some simplification when 
writing rules.

Actually, It's the case of most other improvements, easier to write, 
not inevitably faster.

And don't forget that safe complemented charset in R3 are a pain 
in the ass to construct, because of UTF-8

Actually, DO is really easy to simulate, both in R2 and R3.
Just construct the rule on the fly.
>> parse [2][(rule: do [1 + 1]) 1 1 rule]
==true

Much easier with R3.
assuming, 
>> src: {Number : 10017
Name       : Disable Message Partner
Application: MXS
Severity   : Info
Type       : Alarm Event
Date-Time  : 20/09/10 12:39:43
GMT        : Mon Sep 20 10:39:43 2010}

then, 


>> construct parse copy src [return while[and[some #" " #":"] remove[some 
#" "]| skip]]
== make object! [
    Number: "10017"
    Name: "Disable Message Partner"
    Application: "MXS"
    Severity: "Info"
    Type: "Alarm Event"
    Date-Time: "20/09/10 12:39:43"
    GMT: "Mon Sep 20 10:39:43 2010"
]

but to intercativly move or deform a draw block, i need to construct 
a serie of handles (gobs) at some calculated points. and so it's 
currently a tragedy to have to simulate the draw engine to get th

Why? The main dialect parser of RebGUI can now be expressed in one 
simple parse construct:

parse reduce/only spec words [
	any [
		opt [... evaluate () to return a REBOL value...]
		[
			parse rules to handle REBOL values
			...
		]
	]
]


compared to the relatively complex parsing logic in VID (which, to 
be fair, has to handle styles amongst other things).


The main functional difference between what this allows in VID and 
RebGUI is that you don't need to parenthesize your expressions in 
VID (and I kind of like the idea of having to parenthesize them anyway, 
it makes expressions stand out just like they do in compose).


Having said that, if someone can show me a RebGUI parser built along 
the lines of VID that is clean and lean I'll willing to be convinced 
otherwise. ;)

Tab (and shift-tab) changes focus. 'with is a VID construct, not 
RebGUI (unless Cyphre's changes are more far reaching than I'm aware 
of ;) )


The area in tour.r is the only tabable widget so it re-tabs to end 
of text. Try:

	display "" [area area area]


to see how it works when more than one tabable widget is/are present.

I thought there was supposed to be a switch in display to allow you 
to construct a layout without viewing it.

As RebGUI is aproaching 1.0 release, I would like to know your opinion 
on following - how do you construct and optimise your GUI? So far, 
if you look at tour.r, it reminds me of one big dialog configuration 
box. Not sure what to do about it, maybe it is a given, as widgets 
we have do suggest such layouting. This debate could open discussion 
about eventual addition of potentially missing widgets. My questions 
are:


- are you missing kind of MDI application scheme? Parent window, 
containing one or more child windows, which are not able to being 
moved away from its parent. We used that design much, but I am not 
sure anymore it is vital, as with latest system, we use two monitor 
set-up, and by simple accelerator key we can navigate window being 
moved between displays. Having MDI available, we would probably need 
rebol/view native windowing system. So - is anyone missing something 
like that?


- do you somehow optimise your display? Isn't it like following? 
- with using tabs, everything is in memory, whereas it eventually 
is not needed? How do you divide your application, if you would like 
to have kind of load-on-demand aproach?


What styles do we miss, to further help us have more complete GUIs? 
I created screenshot of potentially two usefull widgets:

http://www.xidys.com/widget-drop-tab.jpg
http://www.xidys.com/widget-vertical-tab.jpg


Especially with the second one, I think it could be usefull. It is 
used by many applications. It is kind of mixture of tab and tree, 
but not necessarily with multi-level aproach, just one level of nesting, 
mostly represented by icons, text, or icons plus text. I would like 
to know your opinion.

Hmm.. bit complicated.  Is there a way we can construct a layout 
and point the existing layout to the new one?  this is how panels 
are done in Vid I think.

yeah ... I mean - I run - do %rebgui.r, and then? do I have to construct 
'display, or does it contain runnable example?

yes... I have an app that works in a similar way. the answer was 
to construct a dialect that describes each screen with well-defined 
parts.

I could construct a block of functions, a block of parameters, and 
a data variable . and feed them to the first function, gradually 
removing them from the blocks and passing them in the callbacks?

Taking some time to read and understand the chapter, I would guess, 
it wouldn't be too hard to construct an algorithm, that did the job.

Vlad, Good to know that there is other people involved in old computers 
revival.
I'm always coding for MSX computers . 

Last year, with Rebol, i tried to simluate the PSG AY family (sound 
processor of MSX) and  an old game (GALAGA).

I think Rebol is perfect to construct toolsets for that purpose (graphics/maps/patterns/sprites 
editors, asm preprocessor).
(see http://www.rebol.org/search.r?find=guest2&form=yes)

Hi. Is there a build-in function do do something like this : given 
the number 5 and the string "abc"  construct the string "abcabcabcabcabc" 
(equal 5 times "abc")?

construct doesn't work in this case, because then input-file just 
become the word READ.

Just for fun (it's a slow sunday today) I've wrote a rebol version 
of the code used as benchmark in this page http://tinyurl.com/5nezt9

here's the code:

REBOL []

person: make object! [
	_count: none
	_prev: none
	_next: none
	
	_construct: func [n] [_count: n]
	get-prev: does [_prev]
	set-prev: func [pr] [_prev: pr]
	get-next: does [_next]
	set-next: func [nxt] [_next: nxt]
	shout: func [shout nth /local aux] [
		if shout < nth [
			return shout + 1
		]
		aux: get-prev aux/set-next get-next
		aux: get-next aux/set-prev get-prev
		1		
	]
]

chain: make object! [
	
	_first: none
	_last: none
	
	_construct: func [size /local current] [
		repeat i size [
			current: make person []
			current/_construct i
			if none? _first [_first: current]
			if not none? _last [
				_last/set-next current
				current/set-prev _last 
			]
			_last: current
		]
		_first/set-prev _last
		_last/set-next _first
	]
	
	kill: func [nth /local current shout] [
		current: _first
		shout: 1
		while [not equal? current current/get-next] [
			shout: current/shout shout nth
			current: current/get-next
		]
		_first: current		
	]	
]

start: now/precise
iter: 100000
loop iter [
	current-chain: make chain []
	current-chain/_construct 40
	current-chain/kill 3
]

print ["Time per iteration =" (difference now/precise start) / iter 
]
halt

which give me:
Time per iteration = 0:00:00.00080234

802 microsecond that is the slower time of the benchmark, 

but not SO slow, php is near with 593 microsecond, jython 632 ... 
.(the test system is pretty like mine so i can compare the result)

There's a way to improve the performance?

or just construct the pipes and pumps that would use seawater near 
to constantly water it? They transport oil across continets in pipes 
no matter how far the water is, they sould be able to make a constant 
suply in there.

Moreover, it looks to me, that we are able to "simulate" such a construct 
using PARSE in REBOL2

So this new lazy LOAD is kind of a PARSER of REBOL values + foreign. 
I think it maybe easier to provide  *rebol value rules*, so you can 
construct your PARSER as desired instead of triggering exceptions.

Nah. That is easy to handle if you know how to construct your parsers 
properly. As easy an extra rule.

but what is definitely issue to me is that it is not imo consistent 
with how we construct scripts ...

and task for me is kind of higher level construct as script or modules 
are ... well, kind of "instance" of "environment"

also, FOLD is a generic construct, as opposed to having special handling 
in all the natives.

Here's the current source for LOAD:

load: func [
	{Loads a file, URL, or string.}
	source [file! url! string! any-block! binary!]

 /header  {Includes REBOL header object if present. Preempts /all.}

;	/next    {Load the next value only. Return block with value and 
new position.}

;	/library {Force file to be a dynamic library. (Command version)}
;	/markup  {Convert HTML and XML to a block of tags and strings.}
	/all     {Load all values. Does not evaluate REBOL header.}
	/unbound {Do not bind the block.}
	/local data tmp
][
	; Note: Avoid use of ALL func, because of /all option
	if any-block? :source [return :source]

	data: case [
		string? source [to-binary source]
		binary? source [source]
		; Check for special media load cases: (temporary code)
		find [%.jpg %.jpeg %.jpe] suffix? source [
			return load-jpeg read/binary source
		]

  url? source [read source] ; can this possibly return not binary! 
  ?
		file? source [read source] ; binary! or block of file!
	]

 ; At this point, data is binary!, a block of file!, or something 
 weird.

	if binary? :data [
		unless find [0 8] tmp: utf? data [
			cause-error 'script 'no-decode ajoin ['UTF tmp]
		]

		; Only load script data:
		if any [header not all] [ ; Note: refinement /all
			if tmp: script? data [data: tmp]
		]
	]

	unless block? :data [data: to block! :data] ; reduce overhead

 ; data is a block! here, unless something really weird is going on
	tmp: none
	
	; Is there a REBOL script header:
	if any [header not all] [ ; /header preempts /all
		tmp: unless any [

   ;not any [file? source url? source] ; removed: hdr in string is same
			unset? first data ; because <> doesn't work with unset!
			'rebol <> first data
			not block? second data
		][ ; Process header:
			attempt [construct/with second data system/standard/script]
		]
		; tmp is header object or none here
		case [
			tmp [
				remove data
				either header [change data tmp][remove data]
				tmp: tmp/type = 'module ; tmp true if module
			]
			header [cause-error 'syntax 'no-header data]
		]
	]
	; tmp is true if module, false or none if not

 ; data is a block!, with possible header object in first position

	; Bind to current global context if not a module:
	unless any [
		unbound
		tmp ; not a module
	][
		bind/new data system/contexts/current
	]

 ; data is a block! here, unless something really weird is going on

	; If appropriate and possible, return singular data value:
	unless any [ ; avoid use of ALL
		all
		header ; This fixes a design flaw in R2's LOAD
		;not block? :data ; can this ever happen?
		empty? data ; R2 compatibility
		not tail? next data
	][data: first data]
	; If /all or /header, data is a block here

	:data
]

i guess names msg are here to construct file trees

if you construct linked gobs for the same line of text (one gob = 
one style), i say good luck to manage wrapped lines.

load: func [
	{Loads a file, URL, or string.}

 source [file! url! string! binary! block!] {Source or block of sources}

 /header  {Includes REBOL header object if present. Preempts /all.}

 /next    {Load the next value only. Return block with value and new 
 position.}

;	/library {Force file to be a dynamic library. (Command version)}
;	/markup  {Convert HTML and XML to a block of tags and strings.}
	/all     {Load all values. Does not evaluate REBOL header.}
	/unbound {Do not bind the block.}
	/local data content val rst tmp

][  ; Note: Avoid use of ALL and NEXT funcs, because of /all and 
/next options
	content: val: rst: tmp: none ; In case people call LOAD/local
	
	; Retrieve the script data
	data: case [
		block? source [ ; Load all in block
			return map x source [apply :load [:x header next all unbound]]
		]
		string? source [source] ; Will convert to binary! later
		binary? source [source]
		; Otherwise source is file or url
		'else [
			; See if a codec exists for this file type
			tmp: find find system/catalog/file-types suffix? source word!
			; Get the data, script required if /header
			content: read source  ; Must be a value, not unset
			case [
				binary? :content [content] ; Assumed script or decodable
				string? :content [content] ; Assumed script or decodable
				header [cause-error 'syntax 'no-header source]
				block? :content [content]
				'else [content: reduce [:content]]
			] ; Don't LOAD/header non-script data from urls and files.

  ] ; content is data if content doesn't need copying, or none if it 
  does
	]
	;print [1 "data type?" type? :data 'content true? :content]
	if string? :data [data: to-binary data] ; REBOL script is UTF-8

 assert/type [data [binary! block!] content [binary! string! block! 
 none!]]
	assert [any [binary? :data not header]]
	if tmp [ ; Use a codec if found earlier
		set/any 'data decode first tmp :data

  ; See if we can shortcut return the value, or fake a script if we 
  can't
		case [

   block? :data [if header [insert data val: make system/standard/script 
   []]]

   header [data: reduce [val: make system/standard/script [] :data]]

   (to logic! unbound) and not next [return :data] ; Shortcut return

   any [next any-block? :data any-word? :data] [data: reduce [:data]]
			'else [return :data] ; No binding needed, shortcut return
		]
		assert/type [data block!] ; If we get this far
	]
	;print [2 'data mold to-string :data]
	
	if binary? :data [ ; It's a script
		unless find [0 8] tmp: utf? data [ ; Not UTF-8
			cause-error 'script 'no-decode ajoin ["UTF-" abs tmp]
		]
		; Process the header if necessary
		either any [header not all] [
			if tmp: script? data [data: tmp] ; Load script data
			; Check for a REBOL header
			set/any [val rst] transcode/only data
			unless case [
				:val = [rebol] [ ; Possible script-in-a-block
					set/any [val rst] transcode/next/error rst
					if block? :val [ ; Is script-in-a-block
						data: first transcode/next data
						rst: skip data 2
					] ; If true, val is header spec
				]
				:val = 'rebol [ ; Possible REBOL header
					set/any [val rst] transcode/next/error rst
					block? :val ; If true, val is header spec
				]
			] [ ; No REBOL header, use default
				val: [] rst: data
			]
			; val is the header spec block, rst the position afterwards

   assert/type [val block! rst [binary! block!] data [binary! block!]]
			assert [same? head data head rst]
			; Make the header object

   either val: attempt [construct/with :val system/standard/script] 
   [
				if (select val 'content) = true [
					val/content: any [:content copy source]
				]
			] [cause-error 'syntax 'no-header data]
			; val is correct header object! here, or you don't get here
			; Convert the rest of the data if necessary and not /next
			unless any [next block? data] [data: rst: to block! rst]
			if block? data [ ; Script-in-a-block or not /next
				case [

     header [change/part data val rst] ; Replace the header with the object

     not all [remove/part data rst]	; Remove the header from the data
				]
				rst: none ; Determined later
			]
		] [rst: data] ; /all and not /header
	]

 ; val is the header object or none, rst is the binary position after 
 or none

 assert/type [val [object! none!] rst [binary! none!] data [binary! 
 block!]]

 assert [any [none? rst same? head data head rst] any [val not header]]

 ;print [3 'val mold/all :val 'data mold/all :data "type?" type? :data]
	
	; LOAD/next or convert data to block - block either way
	assert [block? data: case [
		not next [ ; Not /next
			unless any [block? data not binary? rst] [data: to block! rst]
			data
		]
		; Otherwise /next

  block? data [reduce pick [[data] [first+ data data]] empty? data]
		header [reduce [val rst]] ; Already transcoded above
		binary? rst [transcode/next rst]
	]]
	
	; Bind to current global context if not a module
	unless any [ ; Note: NOT ANY instead of ALL because of /all
		unbound
		(select val 'type) = 'module
	][
		bind/new data system/contexts/current
	]
	;print [6 'data mold/all :data 'tmp mold/all :tmp]
	
	; If appropriate and possible, return singular data value
	unless any [
		all header next  ; /all /header /next
		empty? data
		1 < length? data
	][set/any 'data first data]
	;print [7 'data mold/all :data]
	
	:data
]

Yeah, and it doesn'tt make sense to ignore unset! with REDUCE because 
that is a value context. With CONSTRUCT it is treated as no-value 
and not inserted. Both of these are the right choice in their respective 
cases.

Might be easier to do with parens instead of do/next, but the case-like 
construct is more readable...

the error message is: "** Syntax Error: Invalid construct -- #[**" 
how does that tell it is a method?

I realise that an on-connection-error callback (in the service) would 
also be required to construct a proper reply, based on the service 
being called.

I sometimes use the following safe construct to avoid reloading some 
longer code on each RSP script call : 
<%
	if not value? 'my-object [
		my-object: context [

   ;--loaded on each new worker process only once, then available in 
   global context
			...
		]
	]
%>