World: r3wp

This situation happened to me several times when I first started. 
I would report bugs that were my caused by my own custom user.r. 
It makes you unsure and paranoid about your bug reports, so slows 
you down and maybe stops you making the bug report.

So what I did was create an "anton-user.r" file, which my user.r 
calls *optionally*.
It decides whether to call it based on a command line argument.

So, two different command lines can start rebol "clean" (without 
anton-user.r) or "unclean" (with all my customizations in anton-user.r).

For example, two icons I have:

Rebol/View 1.3.1.3.1   ->  D:\Anton\Dev\Rebol\View\rebview1.3.001.3.1.exe 
-- "do-anton-user?: true"

Rebol/View 1.3.1.3.1 (clean)   ->  D:\Anton\Dev\Rebol\View\rebview1.3.001.3.1.exe

Then my user.r contains this:

        use [args] [if args: system/options/args [do first args]] 
        if value? 'do-anton-user? [
            use [site contents url] [
                site: http://www.lexicon.net/antonr/rebol/

                if contents: load-thru http://www.reboltech.com/index.r[
                    site: select contents [folder "Anton"]
                ] 
                clear find site %index.r 
                if none? do load-thru url: site/anton-user.r [
                    print ["Couldn't load-thru" url]
                ]
            ] 
            unset 'do-anton-user?
        ]

nice tip thanks 8)

(The SITE stuff I have there is so I can keep anton-user.r part of 
my public distribution. You may not need all that though.)

You're welcome :)

So I keep my user.r small, and if by chance it should ever be overwritten 
by a new installation of rebol in the same directory, then I haven't 
lost too much.

I'm checking your library for something useful 8)

There's a few useful things in there.

Am I wrong or Romano ftp patch is builtin latest versions?

I think it is, let me check...

Let's see, anton-user.r...
if find form system/product "view" [


 if (system/version * 1.1.1.0.0) < 1.2.110.0.0 [ ; View1.2.110 has 
 Romano's FTP handler built in.
		do load-thru site/patch/ftp-romano-patch/ftp-patch53.r
	]

hmm yes , well thanks again, maybe time to go sleep here 8)

sleep well.

you have also system/options/script in %user.r . That way you can 
make a "raw"-folder.

 if not find/match clean-path system/options/script view-root/public[ 
 my-extra-stuff ]

4021 was already dismissed, so there was no need for a new ticket 
actually.

Volker, your code means "if this script was not run from view-root/public 
(ie. someone else's script), then do my customizations".
Is my interpretation ok ?

Yes.

I posted two new tickets concerning UDP to RAMBO:
- "no-wait UDP ports and blocking COPY"
- "no-wait UDP ports and READ-IO"
I see the blocking problem as critical

nice! So we make networking layer better and better. Maybe then Gabriele 
will be able to get his Chord to work properly, as it uses UDP too?

I don't know if this is technically a bug or not but I REALLY wish 
that 'switch would throw its errors so that I know where to find 
the switch that is giving me headaches!

>> do http://www.lexicon.net/antonr/rebol/gui/demo-percent-progress.r
connecting to: www.lexicon.net

** User Error: Error.  Target url: http://www.lexicon.net/antonr/rebol/gui/demo-percent-progress.r
could not be retrieved.  Server re
sp...

** Near: do http://www.lexicon.net/antonr/rebol/gui/demo-percent-progress.r
>>

In the above console session, the Error message is cut off just at 
"Server reps..."
so I can't see the response from the server.

(I uploaded the file now, so you can't replicate my error message.. 
:)

Can somebody confirm this pair-bug in rebservices? http://polly.rebol.it/test/test/rebservices/pair-bug/

Problemi shere:

mold-packet calls compand-packet always, and thus insert pair! .

but the call to expand-packet happens only if there is extra data. 
if i outcomment "if not zero? num [" it works

Volker, interesting finding, i will look into it. i guess that removing 
that if is the correct fix.

added a new 1.3.2 bug to Rambo:
b: head insert copy [] make list! [1] produces an invalid block

RAMBO Ticket #-578
Exists? corrupts http protocol.

Are Rambo tickets now negative ??

I believe the new/unreviewed submissions have a temporary sequence 
number and it is negative.

Once the submission been checked manually and found not to be spam, 
it'll get a more permanent sequence number.

>> q: query http://www.rebol.com/index.html
== none
>> read http://www.rebol.com/
== {<html>
<head>

<META NAME="Description" CONTENT="Lightweight distributed computing, 
collaboration, and programming systems for t...

>> q: query http://www.rebol.com/asdfsa.r

** User Error: Error.  Target url: http://www.rebol.com/asdfsa.r
could not be retrieved.  Server response: HTTP/1.0
404 Not Found
** Near: q: query http://www.rebol.com/asdfsa.r
>> read http://www.rebol.com/
== ""

looks like 'query is corrupting http protocol.

*but* also fixes it :)

>> q: query http://www.rebol.com/index.html
== none
>> read http://www.rebol.com/
== {<html>
<head>

<META NAME="Description" CONTENT="Lightweight distributed computing, 
collaboration, and programming systems for t...

Nice one Graham.

This looks like an inconsistency:
>> save %afile [[1]]
>> load/all %afile
== [[[1]]
]
>> save %afile [[1] 2]
>> load/all %afile
== [[1] 2
]
>> save %afile [[1] 2 3]
>> load/all %afile
== [[1] 2 3
]]

The first load/all seems to wrap in brackets once too often !

The difference is in whether there is a single item or multiple items 
being loaded.

Mmm, don't see it in RAMBO already. Anybody can confirm ?  By the 
way, SAVE or SAVE/ALL doesn't make a difference, but it must be LOAD/ALL 
to show the inconsistency.

I'm sure we've been over this one before, actually.
Similar behaviour:
>> save %afile [1]
>> load/all %afile
== [[1]
]
>> save %afile [1 2]
>> load/all %afile
== [1 2
]
>> save %afile [1 2 3]
>> load/all %afile
== [1 2 3
]

that's because save is like mold/only, except when the block only 
has one item.

i'm not sure we can classify it as a bug.

It's difficult to know what was saved after the fact.  Or at least, 
I am finding it difficult to see the logic clearly, in order to produce 
a nice loading function that handles all cases consistently.

either use write mold/only, or don't use load/all.

though, i guess this deserves some more thinking. add it as an issue 
ticket? or as note...

Ahh.. Of course ! Excellent suggestion.

Last time I remember Carl talking about it, he seemed to like the 
idea of being able to save/load a single value easily, so it is perhaps 
difficult to convince him.

.. to change it.