World: r3wp

Yes the rule can only fail after the first parens.

oldes: it makes no difference whether the first parens is left out 
or not

And to answer your question, the variables are just regular rebol 
words, so they are as local as you make them.

and how looks the code you parse?

if the parse is inside function and you set pos in the function as 
a local - it will be local

I'm parsing:

[image "String" image-word [action block]]

where 'image is always first and the remaining items may come in 
random order

there is no pair and no numbers - the pos must be none

oldes, my code is different from his. I used his as it's structurally 
similar to mine

and what exactly do you want?

I want to assign a variable to each element so I can process them 
later

so for the block above it would assing what?

the block is stored in 'attr.

parse attr [

  (txt: 123 print txt) 'image any [set txt string! (print txt) | set 
  img word! | set action block!] (print txt) | 'face
]

gives:

123
search
none
== true

argh... forgot to show you attr

attr: [image test-image "Search" [print "test"]]

Don't you want to use rather:

parse/all attr [(txt: 123 print txt) 'image opt [set txt string! 
(print txt)] opt [ set img word!] opt [ set action block! (print 
txt)] (print txt) | 'face]

it seems the variable is stored in the last iteration, so it's definitely 
lost after one loop.

Doesn't Henrik want to be able to process the attributes in any order 
?

that's what I want. I'm catching fish :-) or rather information to 
store them safely for processing later.

but let me try Oldes thing to see what it does

it works the exact opposite :-) Only the outer 'txt is set, and I 
can't reach the variable inside the block

Henrik, I parse your test attr block successfully:

>> parse [image test-image "Search" [print "action"]] ['image any 
[set txt string! (?? txt) | set img word! | set action block!]]
txt: "Search"
== true

Need more test data to parse.

anton: yes, but what is txt after processing?

I don't want to handle it inside the parse block but after it

it is none.

right. that's the problem

add a block to control the evaluation.

>> parse [image test-image "Search" [print "action"]] ['image any 
[[set txt string! (?? txt)] | set img word! | set action block!]]
txt: "Search"
== true
>> txt
== "Search"

it always set the variables to none, if it fails to match

You have to be careful in your interpretation of  |

interesting, this is the solution:

 txt: 123 parse/all attr ['image any [[set txt string! (print txt)] 
 | [set img word!] | [set action block! (print txt)]] (print txt) 
 | 'face]

oldes solution seems to work

the wonders of parse... :-)

it's just improved anton's solution:-)

I'm trying to figure out a simple example to show why.

I wonder what the difference is? If it's only for controlling how 
global a variable is, it seems a little backwards to me

without the extra brackets, the parse set the variables to none if 
it failes

the brackets would make it a "real" rule, wouldn't it? it would be 
possible to replace the rule with a variable and have the rule block 
placed elsewhere in your code

You have to think of a rule like this:
	[ integer! | ]
as equivalent to
	[ integer! | none ]
or
	opt [ integer! ]

I don;t know how much complex will be the source to parse, but I 
would use the OPT settings as used above

But Henrick wants it in any order, like LAYOUT code. (Thus, it's 
worth to look how LAYOUT works. :)

>> parse ["hi"] [[set num integer! | ] string!]   ?? num
num: none

I just think he can end with endless loop easily, as he will never 
be able to catch the second 'image

>> parse ["hi"] [[set num [integer! | string!]] ] ?? num
num: "hi"

Note the | .     "set NUM to an integer or NONE, then... oh. it's 
the end of this block..."  but in Henrick's rule above he continues 
to set other variables, which is of no concern to parse, it still 
has successfully set NUM to NONE.

there is only one 'image in the block fortunately

ok, so you will not catch the 'face word:-) as it end up in the img 
variable:-)

I think he might be using 'test-image in place of a real image! for 
this example ?

oldes, nope, because I have many different blocks. they either start 
with 'face or with 'image