World: r4wp

With the following code:

#include %GTK.reds

with gdk [
	err: declare struct! [value [g/g-error!]]
	myimg: load-image "image.jpg" err
]

I get the following error:


*** Compilation Error: invalid literal syntax: [value [g/g-error!]]

If I change it to:

err: declare struct! [value [c-string!]]

I get the following error:


*** Compilation Error: argument type mismatch on calling: gdk/load-image

*** expected: [struct! [value [g/g-error!]]], found: [struct! [value 
[c-string!]]]

What am I missing?

You can't use path notation on types. I've been trying to coerce 
Doc into implementing that, but he doesn't want to. :-/

So you have to use an extra WITH g [...]. You can combine that as 
WITH [gdk g] [...]

Aha!  That did the trick!

By the way, if you only use images, you could do with including only 
GDK.reds. That would cut down on the dependencies

Instead of your struct definition, you can use g-error-reference! 
which is already defined in GLib.reds

There's an example of using load-image in GTK.reds in the icon function

Thanks!  When I use load-image, it looks like it is returning a pointer. 
 Is that correct?

For instance: 00DFD2B8

If 'load-image returns a pointer, how do I know when I have reached 
the end of the image?

Better yet, assuming 'load-image returns a pointer, how do I access 
the first memory location referenced by the pointer?

See the function spec in GDK.reds. It returns a pointer to an image! 
struct, so if you want to access the image internally, you have to 
go by that struct definition

As you can see in the same file, I haven't finished the definition 
of that struct, because I don't have to look inside the image

What is your goal?

Doc, could you bring lex-scope up to date with master? Then I can 
start testing it

Done.

Thanks! Been anxiously awaiting those fixes

I'm working on fixing #405 too in the lex-scope branch. Once that 
done, I will merge it in master if there is no regression.

I figured so

I might add #374 too the list too as it is related to scoping.

I hoped so :-)

No build regressions in a full build run

Oddly, there's a build regression in a GTK program just for Syllable, 
which is not even a valid combination

I can't replicate it

I want to be able to compare two images pixel-by-pixel.

Ah, that's relatively easy, because you don't have to know the image 
format. At least not if you can compare byte values instead of strict 
pixels

So you only have to look in the GDK/GLib headers to finish the image! 
struct definition enough to know the data length

Have to run

Do you only want two know if two images are the same? Then why not 
just read the files and compare them?

Actually, I want to know which pixels are different between the two 
images. Thanks for the tips!

You'd have to know GDK's internal image format, then

Perhaps there are more suitable functions in Oldes' ImageMagick binding

I'll take a look.

Looks like ImageMagick will do the trick, if I can just figure how 
to translate the tutorials from C to Red/System.

Kaj: are we good with the fixes in lex-scope branch? If all is fine, 
I will merge it tonight.

Yep, it's starting to look good. I'll build the examples once more

Great, I'll add a few binding tests from the tickets, then I'll do 
the merge.

No build problems

I'm happy to report that literal arguments work now. For example:

cycle: func ["Cycle a series through its index."
	'series [word!]
	/local s
][
	either tail? s: get series [
		set series  next head s
		first head s
	][
		set series  next s
		s/1
	]
]

Nice one!

Branch `lex-scope` merged in master.

It seems that the optional attributes and function description are 
swapped in Red/System compared to REBOL. Is that intentional?

http://www.rebol.com/r3/docs/concepts/funcs-defining.html#section-1

Yes, it makes it easier and faster to find optional attributes this 
way.

We could change it though, it's no significant impact on the compiler.

I believe REBOL allows either order, so it's just the docs that say 
it has to be that way.

I like the REBOL order better in the formatting of the program

I can see arguments both ways. Attrs are short, and having them first 
makes them easier to see, because they will be used far less than 
doc strings. OTOH, being used less makes them feel "more optional" 
than doc strings, hence could come after the doc string.