AltME groups: search

rebol way 
 a: " " 
call/output "set" a
probe a

ie. I see [A

In windows, it starts up another instance with the values passed 
as command line arguments
eg, launch "test" where test is a command line argument

In Linux, it appears to treat "test" as a script which it can't find.

Of course, there are still many reasons why one could not use Syllable 
for a given job, but once we make it suitable for a job, it stays 
that way. It's much easier to say if it's suitable or not, while 
with Linux, one usually fixes all problems until one hits a showstopper, 
then switches distros, only to find that one of the other problems 
is a showstopper there

And then there are the regressions, where a showstopper comes back 
or suddenly appears in a new version of the same Linux distro. We 
have had regressions in Syllable, but not many, and they're usually 
easy to fix

And isn't ghostscript a standard install for most linux distros ?

maybe I try installing a windows fonts package.

I gave Syllable to Carl a year ago in Milan. He was toying with the 
thought of trying to compile REBOL, but he probably didn't

R3 will definitely be a better base to port, and I will try it when 
the open-source layer is released

Syllable uses ELF executable format like modern Linux, so there's 
a chance we would be able to use the R3 library for x86 Linux

Setting the resolution has traditionally been a problem on Linux, 
so that could be the case with that SUSE

Have you tried that Syllable LivePC? Does it have a problem?

Do you have an idea of how hard it would be to do a port ?

Also, I see a vmware image

maybe a year.

0.6.2 is a lot better :-)

Syllable requires a processor that supports i586 (Pentium) instructions, 
this is a leftover from AtheOS; Kurt wrote some lines of i586 instructions 
in assembler. An installation uses up several hundred megabytes of 
hard disc. About 20 to 24mb of ram is needed to boot. So, a Pentium 
60 with 24mb of ram and a small hard disc should be the minimum currently 
able to 'run' Syllable. A Pentium 166MMX with 64mb of ram and a 1gb 
hard drive can run Syllable quite comfortably. PPP is not supported 
yet, so you can only connect to the internet via a local network 
at the present time.

Syllable is currently hardcoded to require an i586 x86 processor 
or compatible (it will [to quote Kurt] "crash and burn on a 486"). 
The current developers are

The real boot code is GrUB. There must be assembler in there, but 
it's out of our hands. We have a bit of assembler code that are probably 
small pieces of our part of the boot process, and maybe things like 
the scheduler

The one from cilinder.be? That's a much more complete image, with 
the developer tools

Yes, included in 0.6.1, and a newer version in 0.6.2

In windows, launch value .. value is taken as a parameter for the 
encapped application.  In Iinux, it demands a script.

it launches a new process .. but can't find the script.

I didn't realise that such a fundamental thing was broken on the 
linux sdk :(

anyway, i'll test this and make sure it works. IIRC the Detective 
has issues with launch/quit on Linux too, but it hasn't been a huge 
problem for me yet.

this will eventually be fixed but it's not easy (needs a different 
build for each distro... so the final solution is R3)

when you specify a font, you need to specifiy the abs path to that 
font ?

if you know you are on kde, you can also call kprint (or was it kprinter) 
and it even displays a nice ui for print options.

don't know if gnome has a print command. lpr should work almost anywhere 
though.

I'll have a look at all this. Thanks.

On the 23rd March, Cyphre posted an example of using true type fonts. 
 I tried it on Fedora Core 6, substituting other true type fonts 
I had, and correcting the paths as appropriate and get a blank screen 
:(

I'll give that a go.

It would be good to have a page which detailed all the issues relating 
to various flavours of Linux - particularly for those new to the 
game.

What would it take for RT to port View to the Zaurus 3200 ?  It can 
run a form of Debian.

the Linux file requester needs some arrow widgets on the scroller. 
 It's very difficult to navigate a directory with lots of files.

Post the REQUEST-FILE source then, we can have a look.

request-file: func [

    {Requests a file using a popup list of files and directories.} 
    /title "Change heading on request." 
    title-line "Title line of request" 
    button-text "Button text for selection" 
    /file name "Default file name or block of file names" 
    /filter filt "Filter or block of filters" 
    /keep "Keep previous settings and results" 
    /only "Return only a single file, not a block." 
    /path "Return absolute path followed by relative files." 
    /save "Request file for saving, otherwise loading." 
    /local where data filt-names filt-values
][
    if none? out start-out 
    either file [

        either block? name [picked: copy name] [picked: reduce [to-file name]]
    ] [
        if not keep [picked: copy []]
    ] 
    if none? picked [picked: copy []] 
    if file: picked/1 [where: first split-path file] 
    while [not tail? picked] [
        set [name file] split-path first picked 
        either name <> where [remove picked] [
            change picked file 
            picked: next picked
        ]
    ] 
    picked: head picked 
    if any [not where not exists? where] [where: clean-path %.] 
    if not keep [
        fp/data: head fp/data 
        so/data: head so/data 
        si: 1
    ] 
    either filter [
        filters: either block? filt [filt] [reduce [filt]]
    ] [if any [not keep not block? filters] [pick-filter]] 
    ff/text: form filters 
    tt/text: either title [copy title-line] ["Select a File:"] 
    ob/text: either title [copy button-text] ["Select"] 
    if all [
        error? done: try [
            filt-names: copy head fp/data 
            filt-values: copy filter-list 
            either filter [
                insert head filt-names "Custom" 
                insert/only filt-values filters
            ] [
                filt-names: at filt-names index? fp/data
            ] 
            done: local-request-file data: reduce 

            [tt/text ob/text clean-path where picked filt-names filt-values found? 
            any [only] found? any [save]] 
            if done [
                dir-path: data/3 
                picked: data/4 

                if not filter [fp/data: at head fp/data index? data/5]
            ] 
            done
        ] 
        (get in disarm done 'code) = 328
    ] [
        done: false 
        read-dir/full either where [where] [dir-path] 
        show-pick 
        inform out 
        unfocus
    ] 
    if error? done [done] 
    if all [done picked any [path not empty? picked]] [
        either path [
            done: insert copy picked copy dir-path 
            either only [done/1] [head done]
        ] [
            foreach file picked [insert file dir-path] 
            either only [picked/1] [picked]
        ]
    ]
]

This is the same source as on Windows.

REQUEST-FILE is just a wrapper for LOCAL-REQUEST-FILE, which is native 
on Windows and I suppose might be native on Linux.

You must post source of LOCAL-REQUEST-FILE or rambo the problem if 
it is native.

It's a native

A native ? That's seems strange.

request-file: func [

    {Requests a file using a popup list of files and directories.}
    /title "Change heading on request."
    title-line "Title line of request"
    button-text "Button text for selection"
    /file name "Default file name or block of file names"
    /filter filt "Filter or block of filters"
    /keep "Keep previous settings and results"
    /only "Return only a single file, not a block."
    /path "Return absolute path followed by relative files."
    /local where
][
    if none? out start-out
    done: false
    either file [

        either block? name [picked: copy name] [picked: reduce [to-file name]]
    ] [
        if not keep [picked: copy []]
    ]
    if none? picked [picked: copy []]
    if file: picked/1 [where: first split-path file]
    while [not tail? picked] [
        set [name file] split-path first picked
        either name <> where [remove picked] [
            change picked file
            picked: next picked
        ]
    ]
    picked: head picked
    if any [not where not exists? where] [where: clean-path %.]
    if not keep [
        fp/data: head fp/data
        so/data: head so/data
        si: 1
    ]
    either filter [
        filters: either block? filt [filt] [reduce [filt]]
    ] [if any [not keep not block? filters] [pick-filter]]
    ff/text: form filters
    tt/text: either title [copy title-line] ["Select a File:"]
    ob/texts/1: either title [copy button-text] ["Select"]
    read-dir/full either where [where] [dir-path]
    show-pick
    inform out
    unfocus
    if all [done picked any [path not empty? picked]] [
        either path [
            done: insert copy picked copy dir-path
            either only [done/1] [head done]
        ] [
            foreach file picked [insert file dir-path]
            either only [picked/1] [picked]
        ]
    ]
]

Debian Etch RC1 is working nicely with 1.3.2, as is AltMe 1.1.29. 
 Ive only been etched a few weeks now.

I'm planning to set up a linux box as my main workstation soon.

Debian's package list is impressive.  And there is nothing stopping 
you from Dl'ing unpackaged goods.  Ubuntu comes no cost for free 
they pay postage...free.  And it's based on Debian, and perhaps making 
mama and papa a little jealous.  :)

I've gotta get a new machine first.

request-file has a premade layout somewhere. Use that. You have the 
source for request-file in the sdk-sources. Needs some binding-tricks 
if you dont use the real source.

I've got a samba share but although I can see files there from the 
linux side, they don't show up on the windows side.  If I drop files 
to the share from the windows side, I can see them.  Is this a permissions 
thing?

where should I start up a script that needs to run with root priviliges 
on boot ?

generally you place a bash script in /etc/init.d/

however... you can always have a very simple shell script that calls 
rebol yourscript.r

(scripts in init.d should take certain standard args, so if your 
rebol script doesn't you're going to need a wrapper script anyway)

avoiding giving anything more permission than it need is a good thing

can the files be world readable (perhaps a different security concern

when the fax is received, it invokes a bash script.  But I just remembered 
that I also call my rebol script from that bash script that creates 
the file in question.  So, I should be able to make it readable to 
my other script to ftp it to me.

Just a shout out to <b>Quanta+<b>


I think I'm starting to like KDE, and just last month I was a Gnome 
that thought it was KiddiE.

In the  had to say it  department
I think I'm settling on an environment, at least until I blink...

Debian GNU/Linux 4.0 - Thanks Ian
KDE with kdm - Thanks Matthias
REBOL - Thanks Carl
Cheyenne - Thanks Nenad
Quanta+ - Thanks Eric
Konsole - Thanks Lars
Whatever browser pops up...and the 'tousands of other net apps
Ahh, livin'.

Been Playing with Ubuntu .... managed to get a Rebol console running 
....

I'm not sure if it's very cross-unix, so there may be a better way.

Hi Phil,


*nix doesn't use drive-letters, so _all_ drives show up somewhere 
under the root as directories. _Where_ they show up is up to you 
;-)

Drives are "named" /dev/hda1 (first paritiion on first disk), /dev/hdb3 
(third partitiion on second drive), etc ...

There's a file, which describes the mapping, where your drives are 
put, in Rebol

print read %/etc/fstab


the first two columns show where ich drive is put in the directory 
tree (and then some more info)

or you can do 

call "mount"

which displays the currently mounted drives.

I've lost several messages that way today, despite being on a broadband 
line.

Seems either the new server is not stable, or the latest Altme is 
buggy.

Either way, this remains one of the most annoying Altme usability/reliability 
issues.

I've got this odd problem.  I am running fedora core 6 as a vmware 
image.  the time keeps dropping back 12 hours.  I have the right 
time zone, and ntpd is running, and I'm pointing to a local time 
server.

I don't think a vmware image can alter the host clock ...

happening at boot?
does hwclock --hctosys fix the time?
I'm fishing here...could it be a cron job "helping"

oh .. is hwclock a program?

don't have a tzconfig program

do a cat /etc/timezone

Write a little script like   forever [print now/exact wait 00:05:00] 
 and log it or just monitor

must be a vmware bug

perhaps I should run a script that fixes the date every minute!

A google of   fedora time drift vmware  shows a lot of info

looks like this is a common problem

seems it may be a config issue

I'm going to leave a nasty comment to the person who uploaded this 
image!

found a how on source-forge

This is getting silly.  I have to run a daytime server on windows 
2003 server so that my cron job on the fedora side can read the time 
and then set the time.

damn I was there 1 hour ago and it wasn't ready yet.So I just reinstalled 
Mandriva,some thing happened to my home dir.First thing after a reboot;install 
View/Altme :)

Alan does your Mandriva have a  sensible-browser  command?

Alan; the sensible-browser seems to be a Debain branch feature.  
It's GPL so we can work that out.  The hint I left for the RAMBO 
folk is to internally call sensible-browser.  This won't work for 
us until RT does a build.  Once/if that is done, we can talk/plan 
more, about documenting getting browse native functioning.


In the meantime, try the  browse: [url /only] call reform ["path 
to browser command" rejoin [{"} url {"}]]

redef.  If you know the switches for your browser, you can even code 
this to support /only.

Does anyone think that perhaps we may see a Rebol port to the Playstation 
3?

I'm waiting for the Linux port to the Tungsten E2...that will be 
fun, with a little rebview running off an SD card.

Well, to be honest I'm a Nintendo fanboy.  So no  :)

I waited many moons for a Palm port in the old rebol days... alas 
it was not to be. So then I got a Zaurus and alas again, no rebol. 
: (

brian, yes linux is not too far away they where purchased by the 
no.1 linux handheld distributor... last I heard they are writing 
a palm emulator to be able to run normal palm apps on top of the 
linux kernel.

with linux instead, then we might have a better platform, since having 
64 mb palms is pretty much the minimum for a long time.  and if linux 
can be smart and allow us to use the SD cards as ram... well then 
there is no more problems.

obviously the smaller screens also meen reduced memory footprint, 
since rebol/view has a lot of image clip regions cached... so smaller 
buttons and windows... means much less ram being used for those as 
well  :-)

James; Do you like the Zaurus.  I've promised some of the staffers 
that I'd get them PDA's.  I was thinking about a low end Z 22

I do but I wouldn't recommend it as a PDA.

Max, why don't you use LOAD in PARSE, if you want to? Example:

rule: [
	(result: make block! 0)
	any [
		[
			; trying to load
			pos: skip (
				next-rule: either error? try [
					set [value pos] load/next pos
				] [[end skip]] [[:pos]]
			) next-rule |
			; load didn't succeed, using something else
			copy value skip
		]
		(insert/only tail result get/any 'value)
	]
]

>> parse "1 2 a, 3" rule
== true
>> result
== [1 2 "a" "," 3]

well, its not that we can't right now... like you show, load can 
be called anywhere.


 (though I am going to admit I would not have figured out such an 
 elegent way to do it, *I* am not yet able to use the force... <ahem> 
 parse with such skill, MASTER JEDI  ;-) ... 


but did you see how much is needed to make that work and the twist 
of redefining a rule within the evaluated expression?  parse is supposed 
to scream in speed... having a load directive integrated would be 
MUCH faster IMHO, and would be MUCH simpler for the new guys on the 
block and us oldies too...  try to explain the above to the average 
joe and well... there goes the parser...  I only 'SEE' your example 
cause I've become able to fit parser in my mind, and I realize that 
even within the advanced REBOLer crowd... not everyone really grasps 
the parser...


one question though, does the set [value pos] define the word globally 
even if the entire above code is within a context which has a value 
and pos defined?

twist of redefining a rule...

 - that is just to "tell PARSE" whether the paren operation succeeded 
 or not - [end skip] is failure

Often if you're running a script and you need to send an email, you 
can just use send.  But sometimes that doesn't work for various reasons 
.. like the internet connection being down.  So, I was wondering 
if it makes more sense to hand the email over the mail transport 
agent on your linux system.

hum GPL itself is more like a 50 ton capable steel  cable when compared 
to a string.

Other than Public Domain, I don't think there is a freer license. 
 Except for the fact you aren't free to restrict access to any stuff 
based on GPL...so freedom from / freedom to arises here.

but an mit/bsd license derived product can *become* close source 
in time... whereas a GPL-derived product cannot... but there is nothing 
stopping the "original" copyright holder to release further versions 
closed source.

MySQL being a good example.

I don't know, I like the GPL, but it is definitely a 'headspace' 
issue.

Nope.  Copyrights stay.  A good thing in my opinion. Again 'headspace'.

that's the thing about public domain  :-)  no more rights, as free 
as a rock on the moon.

BSD is a respectable license.  But the GPL does not stop anyone from 
commercializing, it's just that you can't stop the next guy from 
giving the stuff away  for free.

look at wireshark... a very good example.

If you are trying to run a CGI, best to do it with Core, not View
Looks like your script *has* started, but fails.