World: r3wp

serriously the rendering is slow beacause it's a hudge cheat ( Ahley 
compone on fly a multi composed widget layout) make a change and 
click ave to see how long it take to be refreshed...

I have problems with s it's 2:30 and i need to woke up at 5:00

http://shadwolf.free.fr/mdp-gui13.jpg-> image of MDP-GUI 1.3

it's like adding line selection displacement (that eFishAnt suggested 
to me for MDP-GUI...) well thinking of it it's cool !!! but I'm affraid 
to say that the text displacement  motion will be so slow that this 
feature will not be very shinny for users ...

I fund very best hope for all this kind of highly advanced features 
with rebol View AGG engine (rendering those kind of advanced widgets 
 that need lot of refresh and quick rerendering to be visually smooth 
and less processor calculation could be very a good thing if we make 
them in AGG) but that's a hudge work ...

maybe I'm over estimating the capabilities of AGG on it's actual 
form too... because it's still depending on View layout rendering 
engine

it will be great to have Carl's opinion on those topics ;)

it's like on fly colored text  in area it"s a very good idea too 
but for the same reason it's actually not in the plans

who want to wait 3 minutes is code to be refresh every time he insert 
a new caracter on an existing colores text interpreted area ??? (3 
minutes okay I'm over estimating the time that takes but maybe not 
for hudge text in fact ...more you have text displayed in the area 
more slow will be the refreshment on new caracters insertion)

If I may make a suggestion (though this may already have been considered) 
-- use a display toggle (with keyboard shortcut) instead of split 
screen.  Upon activating the toggle, you could then scan-doc -> gen-doc 
the area text without saving it to disk...

Though the split-screen display may be preferrable for those with 
larger screens than my own :o)

Ctrl-U for 'Update' ?

Also recommended -- http://homepages.paradise.net.nz/left/rebol/scripts/scrolling.r

Great stuff shadwolf! MDP-GUI has a good look and feel to it now. 
A couple of suggestions:


1. The preview pane is out by 40 pixels, page-width: 360 should fix 
it
2. Full-screen (even as an option) would be good
3. Option to use makedoc2.r instead of make-doc-pro.r


Now that the "proof of concept" seems sound I'll be concentrating 
on the VID emitter with a view to making it MDP complient and more 
efficient.

ability to directly edit the elements in the preview pane
 I'm working on that, but its a real pane (pardon the pun).

the rendering is slow because its a huge cheat

 - yes, like you I looked at a "live update" but its just too slowwwww. 
 An intermediate solution might be to have a timer that refreshes 
 the preview every few seconds or so.

yeah .. it would be good to be able to choose which variant of makedoc 
to use.

>> do %mdp-gui-v1.r
Script: "Untitled" (6-Oct-2004)

** Script Error: split-path expected target argument of type: file 
url

** Where: func [face value][browse join (first split-path config-data/current-file) 
%makedocpro.
html]
** Near: first split-path config-data/current-file

where I didn't select a path to mdp, and then clicked on the help 
button

I'm adding some °7° function to MD2 that will parse °7° tags.. this 
would add a global DB for adding links, addresses, bit's of MD code 
etc.   update the °7°, and you instantly update any future MDs.  
You can run the latest MD2 from Carl with this.. 

do load http://powerofq.com/start
°run° {°Make Doc 2°}

Ashley thank you very mutch for the tips ;) I will work on special 
adaptation to MDP format requirement on the redering process. Option 
tu use makedoc 2  sure but in this case your cute MD2IDE will be 
deprecated and I  doesn't want that... MDP and MD2 have similar form 
but diferent kind of rendering processs. I don't want to make a monster 
application ... I want to make a quick usefull little GUI to write 
aesyly MDP formated programs with lot of automation like save on 
quit, recent file list, etc...

graham yes the help is not included into MDPGUI it's included into 
the MakedocPro.r directory (makedocpro.txt) so normally I use the 
MDP.r directory path to find the help file

I will fix this by forcing ppl to put a path for mdp script

Graham: On first run config pan 's cancel button will pop up an alert 
message if the config-file is not existing

that way you doesn't have this kind of problem any more ;)

Graham her you get the patch for this proprlem that allows you to 
not redownload it :
add this code in config-pan: 

btn-cancel "Cancel" [
			either exists?  mdp-config-file [ 
				unview/only config-win 
				if not viewed? main-win [mdp-gui-init]

   ][ alert "It's my first run baby !! ^/ You need to fill properly 
   the config form !!!" ]
			
		]

My first version of NicomDoc format, that is designed from MakeDoc2, 
is close to be completed, and I will publish specifications and scripts 
shortly. For now I've made an example document, so you can see it's 
features: http://home.tiscali.dk/john.niclasen/nicomdoc/example.html

nice examples, Geomol...curious the look of the markup needed to 
generate that.  I think you said you surround the text with nestible 
tags, using a heirarchical structure?

My example is "Valid HTML 4.01 Transitional" btw. (see http://validator.w3.org/) 
for those of you, who like me like to stick to standards.

@eFishAnt Yes, I use blocks - [ and ] - for notes and tables, but 
there's very little syntax in the rest. The way I've implemented 
style changes will assure, that the output is strict hierarchical 
datamodel, but freedom is given to the writer for input.

To clarify: the text-document, the writer produces, is only strict 
hierarchical with notes and tables (but I use block notation for 
that, so it's not a big problem for the writer). Formatting can be 
thrown around freely. That text is first parsed into a RebXML format, 
which is strictly hierarchical (like XML and HTML is, when it's done 
right without error).

The text version of the above example can be seen here: http://home.tiscali.dk/john.niclasen/nicomdoc/example.txt

interesting...the block notation is not too bad.  I might have to 
sleep on it twice (meaing it takes more than a one night stand) to 
sort out suggestions....but I have to admit it is better looking 
than what I thought it would be...

a few more keystrokes for font than the old nroff / troff stuff.

distinguishing between [] and () in a writer's mind seems very logical.

a writer can change "voice" while typing (like this - notice I am 
now talking to you between the lines) using parentheses, while blocks 
are containers for just about anything without the sublime meaning 
that () have...for example here are my favorite groups [Pixies Brian-Eno 
Patti-Smith They-Might-Be-Giants]

I am thinking [] could make English better...in a literal sense.

Lists in English are now: Item 1, Item 2, and Item 3.

(sorry Webster)

I finalise my first stage on adaptation  rendering process !!!

you can Dl it on http://rebol.dev.fr/view.php?sid=140(actual version 
is 1.3.3 see rebol script header for changelog )

MDP-GUI rendering nows feet very more closer the MDP Format allowing 
beatifull doc preview (but processing is slow damn maybe I hace to 
add some kind of progress bar to not bore the user with hangon...)

still only mdp?

graham for MD2 you get MD2IDE from ashley !!!

I'm a competitor not an application / project killer

ok.

MDP and MD2 are to far on rendering method to be correctly handle 
in a single app ...

and has you notice it (cause you are very clever) both format have 
no format heading in files like =MDP or =MD2 at the very begining 
of the .txt file witch countains the

flagged doc ..

has MDP rendering is very distinct from MD2 we can merge them .. 
one little example section on MD2 it's sised text folowed by a red 
line on MDP it's black line  folowed by newline then Num_section 
(example 1.2.4)  folowed by text then newline again ....

has the flag are the same how user can make the difference betwin 
one and other ...

making 2 rendering engine will encrease and complicate the application 
for very not a good result ...