World: r3wp

(I wonder what they use to document the ImageMagick project.. it 
does not look like fully automated documentation. There are also 
some typos in the spec names.)

I'm working on an incremental lexer able to perform line-by-line 
analysis of any plain text documents.

the idea is to allow editing without having to reparse all the document.

The syntactical rules will be regular parse rules easy to understand 
and to modify, to facilitate
the creation of different model of document.

Of course, the first target is a rebol parser, but the make-doc format 
is also in my short range.

If anyone already have deep thoughts about the subject, please share 
your opinions.
I will come with a proto soon enough.

Will the REBOL parser be using the R3 incremental parser, TRANSCODE 
?

Btw, if you are using PARSE for incremental parsing, watch out for 
this: http://issue.cc/r3/1787

what is the equivalent in R3  of disarm ?

Nothing, it's unnecessary in R3.

You have to DO an error explicitly to trigger it. Otherwise, you 
can just reference it directly as if it were an object.

value: disarm :value 
 how i translate that ?

you  don't need to, just use the /error refinemeent with transcode.
>> transcode/next/error
and you get an error object if any

[!REBOL3 Parse group now exists for specific R3 Parse discussions]

ok thx

steeve it's in the colorize function code of area tc ...

either error? :value [
                        value: disarm :value

how do I adapt that ?

value: disarm :value 
Just get rid of the line.

ok ...

and this how i use transcode in it ? instead of load/next here ? 
>> error? set/any [value end] try [load/next start]

Shad, I will do it if you wait a little

I can do it now either way i need to understand the way R3 handle 
things

since the disarm is gone this like don't work anymore either value/arg2/1

It depends on the circumstances. Once an error is triggered it can't 
be referred to again until it is caught by TRY, which disarms it.

You don't need disarm in R3:

>> if error? set/any 'err try [1 / 0][ probe err true]
make error! [
    code: 400
    type: 'Math
    id: 'zero-divide
    arg1: none
    arg2: none
    arg3: none
    near: [/ 0]
    where: [/ try]
]
== true

Do we need the set/any or is enough just:
	if error? err: try [1 / 0][ probe err true]

how to copy a parsed stream into a object instead a string var?
this fails
parse page [thru <tag1>  copy obj-buffer/result   to <tag2>]

** Script Error: Invalid argument: obj-buffer/result

ok, looks like this is working.


parse page [thru <tag1>  copy a  to <tag2> (obj-buffer/result : copy 
a)]

Common parse patterns request - http://www.rebol.com/article/0508.html

Thanks Pekr for drawing that to my attention.

If it's a patterns page, it shouldn't be R3 specific. If patterns 
require different implementations in R2 and R3, those can be filled 
in, or stubbed out with a note as to why it is either not needed 
or not worth the effort in one version or the other.

Every pattern should have an R2 and R3 version, with explanations.

Also, every new R3 operation should have an R2 equivalent in the 
pattern list. The parse proposals page can be mined for examples.

IIRC there's only one new R3 op that doesn't have an R2 equivalent, 
but it might not be implemented yet. Let me check.

I just didn't want to remove the R3-only comments without some discussion, 
in case there was a reason for them being there.

Yup, agreed. Just checked, LIMIT, the one proposal with no R2 equivalent 
hasn't been implemented yet.

Hello

Hello, I am struggling to build a parser of equation. The idea is 
a function who would take sth like "SQRT(1-x^2)/(1-x)^3" and return 
a block of rebol code: [divide square-root subtract 1 power x 2 power 
subtract 1 x 3].
Anyone has done it already?

I know Gabriele has done something along those lines, and I'm sure 
others have as well. I don't know of any that return REBOL code. 
Can you be more specific about what problems you're having?

Geocachi, see

http://www.fm.tul.cz/~ladislav/rebol/evaluate.r

(can return Rebol code)

Thanks ladislav... but I end up with my own implementation...  http://www.rebol.org/view-script.r?script=parse-equation.r
But I will check the your implementation to see how it differs :)

Hello, I put a new version of my parse-equation.r script on rebol.org: 
the parser engine is now completed and finalized. I improved the 
recursive implementation to make the code much more readable, understandable 
and elegant. Future improvements should be now limited to adding 
new math functions and a syntax error handling to return usefull 
error messages to the user when she/he makes a syntax error in the 
equation.  http://www.rebol.org/view-script.r?script=parse-equation.r

very nice work !

Thanks... just uploaded a new file on rebol.org... I mistakenly put 
a old one :)

this does not work with signed numbers as in "1 + -2"

indeed, not yet implemented... in the meantime use "1 - 2" instead

And i am not sure how to implement signed numbers... Usinfg the minus 
sign introduces a level of complexiy I want to avaoid in the code. 
I might decide to force the use of a function like neg()... I am 
still thinking about it...

actually, i found a quite straightforward way to handle signed numbers 
:) i will post later today (for now, I am going to ski... shusss!!!!)

Hello sqlab... Signed numbers and signed sub-expression are now properly 
handled :)
>> parse-equation "+(1+-x)**-sqrt(-1--x)"

== [power add 1.0 negate x negate square-root subtract -1.0 negate 
x]

Thanks for pointing my attention to this issue.

http://www.rebol.org/view-script.r?script=parse-equation.r

Very nice script. ^ character can also be added as power operator 
just like **. As in some languages ^ is power operator.

Oh and English comments in the source code please, if possible :)