World: r3wp

Your question should be posted in "Source Control" channel.

Maybe you should have look at a Git documentation first?

ah, I looked

Trying to push my first modifications currently...

Doc, do you see something ? 
I pulled a request.

yep, here it is:
https://github.com/dockimbel/Red/pull/1

why did you comment out the original code, instead of just deleting 
it?

(refering to: https://github.com/SteeveGit/Red/commit/d5f6c9f)

i'd suggest to keep the TBD comments and delete the rest

You mean, you don't want to see the old code as comments ?

Exactly.

Steeve: yes, got it, looks good to me. I'll pull it in a few minutes. 
Andreas is right, you don't need to comment the old code, just delete 
it (but keep the TBD comments).

Your pull request disappeared, I guess you'll publish a new one.

In one minute...

sorry...

Oh, my bad, looking in the wrong repo...still need to adjust to the 
DCVS approach.

The pull requests will appear in your repo (dockimbel/Red).

Steeve's commits of course go to Steeve's repo :)

Sure, it's clear to me now. :-)

I repulled a request.

Ok, merging it to my Red repo...

But, instead of seeing 2 commitments , I should have deleted the 
first one.
Don't know how to do that though....

Done

Steeve: have you tested the changes locally by compiling the hello 
script?

The power-of-2? function is not retro-compatible with the previous 
version.

eh ?

Let me test that more closely...

I've not tested with the compiler but the function itself

The compiler chokes on the new version of this function, but works 
with the old one.

It was missing a to-integer conversion: power-of-2?: func [n [integer!]][if 
zero? n - 1 and n [to integer! log-2 n]]

ah, yes sorry

Should have tested with the compiler.
sorry

Fixed. Thanks for the nice code reduction ;-)

well it should be a little little faster aswell ;-)

I saw some rooms for code factorization also, but you want it as 
fast as it can, I guess

For now, it's a prototype code, it doesn't need to be much micro-optimized 
as it will be fully rewritten in a few months using the Red language.

I see that you use a lot, nested EITHER structures.
Personnaly I prefer flat CASE structures, more readable.
And i also think CASE could be faster.

Yeah, it's temporary.

I guess you're looking at IA32.r/emit-path code?

yes

I need to factorize the code first, there's some redundancies, but 
it's also untested code. So, once it will work ok, I'll clean it 
up.

Go Go Gadget

:-)

Btw, you've opened an issue ticket too...good opportunity for me 
to test github's tickets management options.

Hm... I did nothing though.

It seems a ticket is opened for each pull request...

I guess it's automated.

I just click ramdomly like a foul

lol

But how can I resync my fork, I wonder...
I don't see your modification