World: r3wp

In R2 View, Alt is not supported. If you look at the event datatype, 
there are fields only for control and shift.

ah, my bad, Anton is right ....

Robert, you're mentioning Excel, so you're probably trying to do 
ALT+RETURN hard line-breaks with Excel thru comlib?

Maybe you can use the actual character code that Excel uses whenever 
one presses Alt+Enter. The web says Excel just inserts a line feed 
#"^/", have you tried that?

Alt+RETURN is not a character

Christian, right, I try to insert a hard line-break in Excel. I tried 
inserting NEWLINE, which doesn't work. Will try just the LN. Thanks.

Apart from chucking an error, what is the best way to determine if 
a path exists inside an object?  So, the path might be 10 deep ...

As far as I know, Graham, there is not a simpler way than provoking 
an error.
If you are already at the bottom of the structure FIND can help:

    a: make object! [b: make object! [c: 1]]
    find next first a/b 'd     ;; is a/b/d valid?
    == none
    find next first a/b 'c     ;; is a.b.c valid?
   == [c]


But recursing to the bottom may be slower than just trying the error?:
   error? try [a/b/d]
   error? try [a/b/c]

Seems ugly to have to use an error ...

It's a poorly specified way of obtaining some information.

perhaps there is merit for a feature to quickly determine the existence 
of a deeply stored value?

can't even do this

in object 'path/to/rebol

because 'in expects a word

How is this done in R3 ??

Same as in R2:
in object/path/to 'rebol

The problem is that at some point the field names can turn into function 
refinements, depending on the values assigned...

And why is that such a problem?

I mean, it's a solvable problem. IN could theoretically be enhanced 
to also accept paths, but, iIrc, Carl wasn't keen on the idea because 
he wanted IN to remain simple and fast. He said something like "it's 
for words".

The only thing we can do is make our own mezz version of IN that 
accepts paths.

(Ok, it's not a solvable problem for paths with dynamic elements 
in them like functions, but of course, there are many more "static" 
paths which this would be darned useful for.)

GET and SET work with paths in R3. Perhaps as you describe could 
be added as well, to handle the IN function for paths. And functions 
wouldn't be traced through.

the problem with object/path/to 'rebol is that it assumes object/path/to 
exists

Yup.

So you have to check for 'to in object/path first, and so on. I didn't 
say it would be a fast mezzanine :(

The way this kind of thing is resolved in R3 is through liberal use 
of the ASSERT/type function, and not representing XML as objects.

I guess there's a lot of advantages to just using blocks ... as it 
is now, some elements in my object are blocks and some are objects 
which makes it very messy

OTOH, blocks could get messy too ...


at present I have structures like ccr/body/alerts/alert where alert 
can be either a single alert object, or a block of alert objects

Ideally if I were to use blocks, it would end up that alert is always 
a block .. 0 ... n blocks

The R3 GUI is structured as a mix of objects, maps, blocks and gobs, 
but it is very consistent and not messy.

I presume it can't change though ... if a particular item is an object!, 
it won't be a block of objects in another instance ....

very consistent
, yes :)

That's my issue ... because the standard for the data has not been 
tightly defined, there are varying implementations leading to somewhat 
variable structures

Sounds like you need a tighter definition, or to resign yourself 
to slow code.

It's an ASTM standard ... I can't change the XSD!

But you can use a consistent REBOL representation of it, and don't 
need objects.

Google health has solved the issue by only working with a subset 
of the XSD.

Well, you can clean it up to a usable structure on read, process 
it nicely, then regenerate the bad XML on write.

Yeah .. I guess I was getting to that.

I suspect this is a very common issue and not bad xml.  So, if the 
XSD states an element can be 0 ... n and each element can be of a 
different type, then it's going to be difficult to work with no matter 
what.

If an XSD states that; it should always be a block even in the single 
object case.

I guess the xml-to-object.r script doesn't know that!

I never liked that script :(

(no offence intended to whoever wrote it)

Saw a new language the other day that has a native, literal data 
structure which matches the XML object model, without the syntax.

Nice language - clean and small, pretty to look at, compiles to native 
code.

Given 


>> obj: make object! [ Description: make object! [ Text: "REBOL" 
]]
>> path: 'Description/Text
== Description/Text


How can I find "REBOL" without breaking the path into it's constituents?

I'm guessing that there's something very simple ... but in the meantime


get-value: func [ obj [object!] path [path!]
	/local id
][
	if not empty? path [
		either all [ id: in obj path/1  obj: get id  object? obj ][
			remove path
			get-value obj path
		][
			obj
		]
	]
]

>> get-value obj path
== "REBOL"

Is this a bug or a feature:

>> to integer! ""
== 0

>> to decimal! ""
** Script Error: Invalid argument:
** Near: to decimal! ""

R3 returns a bug with the first one, but it could be intentional 
in both R2 and R3.

It was an intentional change. TO has been cleaned up a lot in R3 
- the error there tells why:
>> to integer! ""
** Script error: content too short (or just whitespace)