Re: Is it me?

Hi Dimitris

I don't think that's right, but I think I've solved it. Sorry, my code is particularly confusing.

The debug output shows "SETVAR ONETWO 1$two" (which I think is, as you say, the literal contents). It looks like the "1" has been translated from $one, and $two appended literally, with the \ acting not as a stop, but an escape. At the "say" line, this still results in "12", but not because variable onetwo contains "12". I'm sorry, I didn't realise that I could put $ inside a variable contents like that.

It reports
* Variable "two3" was not found

which is presumably as it tries to interpret
onetwothree = "$onetwo\$three".

If onetwo contained "12", it wouldn't be looking for "two3". I think it's doing this:

interpret: $onetwo\$three
(onetwo contains 1$two), so
         = 1$two\$three
         = 1$two3
($two3 = "", unassigned), so
         = 1

I'm afraid that my choice of variable names wasn't helpful. They made sense because I set out intending to concatenate the contents, but used the wrong syntax. Fer said to use $one\#$two, and that seems to work fine. It would be easier to read if I'd just used a, b, c and d.

So, perhaps the answer to this isn't important and I should just not use that syntax, but it's important if a variable can contain the name of another with a dollar prefix. I thought it should all be translated in the assignment. It could be quite useful. However, if that's what's happened, and it's how it should work, it's also quite context sensitive, because in $onetwo\$three it doesn't act to escape the "$three" and add it as a literal string, as it does with $one\$two (presumably because "1$two" begins with a literal itself).

Best regards
John

--- In [email protected], Dimitris Keletsekis <gui4cli@...> wrote:
>
> Hi,
> 
> Sorry, its a mistake...
> (I should have tested it first...)
> 
> > onetwo = "$one\$two"
> > say "onetwo: $onetwo"
> 
> The above does correctly result in 12.
> 
> In the first line, g4c will stop reading the variable name at "\", so
> will evaluate it as "$one", ie "1".
> 
> Assignments work the same whether its Say or =
> 
> Also, the debug output are the raw variable contents (AFAIR)
> 
> Best regards
> Dimitris
> 
> 
> On Sun, May 2, 2010 at 10:36 PM, freedeanna <freedeanna@...> wrote:
> >
> >
> >
> > Dimitris, thanks. I'm very confused about all this. If what you say below is true, I don't understand why the first output would be "12" unless variable assignment works completely differently from "say". Looking at the debug output, it says SETVAR ONETWO 1$two. Is that the debug output from my assignment onetwo = "$one\$two"?
> >
> > It appears to have translated the first part to "1", not the second, then attached $two. When this is "say-ed", it appears to comes out as "12", which has a kind of logic, but not the logic you've described.
> >
> > I realise I was also confused because I expected "say" to reveal the untranslated contents of things in this instance, not translate them. If a variable contains "one$two" I expected say to tell me that, not look for a variable one$two. But I don't usually expect that in day-to-day programming - it was just because I wanted to inspect the direct contents of these variables, and I forgot. Normally, if I know that x contains "$$lv.line" and a listview is at line 24, say $x will give me 24. Sorry, I was being dim there.
> >
> > Does "say" do the same as an assignment, translate deeply until no more variables can be translated, or just one layer down?
> > Can I take it that debug output is the direct contents?
> >
> > --- In [email protected], Dimitris Keletsekis <gui4cli@> wrote:
> > >
> > > Hi,
> > >
> > > I think the problem is you are confusing the variable names with the
> > > variable contents (which I also did when I first ran the gui).
> > >
> > > The first "$" applies to the whole variable. The subsequent "$" (if attached
> > > to the variable name) will be translated and attached to the variable's name
> > > (not contents) and then g4c will look for the whole variable.
> > >
> > > So, say "$one$two"
> > > means - look for a variable named "one2"
> > >
> > > and say "$one\$two"
> > > means - look for a variable named "one$two"
> > > which when re-translated will result in - look for variable "one2"
> > >
> > > Dimitris
> > >
> > >
> > >
> > >
> > > On Sat, May 1, 2010 at 8:05 PM, freedeanna <freedeanna@> wrote:
> > >
> > > >
> > > >
> > > > G4C STRING_TEST
> > > >
> > > > /* ===============================================================
> > > > Description: What's happening here? It seems like the same process gives
> > > > different results the second time round and according to which end a
> > > > variable is attached to another.
> > > > =============================================================== */
> > > >
> > > > WINDOW 50 30 300 80 "String Test gui"
> > > > WinAttr style resize
> > > >
> > > > xOnLoad
> > > > GuiOpen #this
> > > >
> > > > one = "1"
> > > > two = "2"
> > > > three = "3"
> > > >
> > > > xOnClose
> > > > GuiQuit #this
> > > >
> > > > xbutton 20 20 -40 30 "concatenate"
> > > > say "\n"
> > > > onetwo = "$one\$two"
> > > > say "onetwo: $onetwo"
> > > > onetwothree = "$onetwo\$three"
> > > > say "onetwothree: $onetwothree"
> > > > further = "$three\$onetwo"
> > > > say "further: $further"
> > > > addabit = "$onetwo\LITERAL"
> > > > say "addabit: $addabit"
> > > > oneplusliteral = "$one\LITERAL"
> > > > say "oneplusliteral: $oneplusliteral"
> > > > literalplusonetwo = "LITERAL\$onetwo"
> > > > say "literalplusonetwo: $literalplusonetwo"
> > > > appvar onetwo " literal appvar-ed ok"
> > > > say "onetwo by appvar: $onetwo"
> > > >
> > > > // END ////////////////////
> > > >
> > > > Here's the output:
> > > >
> > > > onetwo: 12
> > > > onetwothree: 1
> > > > further: 312
> > > > addabit: 1
> > > > oneplusliteral: 1LITERAL
> > > > literalplusonetwo: LITERAL12
> > > > onetwo by appvar: 12 literal appvar-ed ok
> > > >
> > > > and the relevant part of the debug output is:
> > > >
> > > > 20: SAY
> > > >
> > > > 21: SETVAR ONETWO 1$two
> > > > 22: SAY onetwo: 12
> > > > * Variable "two3" was not found
> > > > 23: SETVAR ONETWOTHREE 1
> > > > 24: SAY onetwothree: 1
> > > > 25: SETVAR FURTHER 3$onetwo
> > > > 26: SAY further: 312
> > > > * Variable "twoLITERAL" was not found
> > > > 27: SETVAR ADDABIT 1
> > > > 28: SAY addabit: 1
> > > > 29: SETVAR ONEPLUSLITERAL 1LITERAL
> > > > 30: SAY oneplusliteral: 1LITERAL
> > > > 31: SETVAR LITERALPLUSONETWO LITERAL$onetwo
> > > > 32: SAY literalplusonetwo: LITERAL12
> > > > 33: APPVAR onetwo literal appvar-ed ok
> > > > 34: SAY onetwo by appvar: 12 literal appvar-ed ok
> > > > ====| End of Commands.
> > > >
> > > > At the first setvar, does onetwo contain "1$two" or "12"? If the former, as
> > > > per the help file, the "say" output is translating it, which seems
> > > > unhelpful. Shouldn't the say report it directly as "1$two".
> > > >
> > > > Strangely, to my mind, variable "onetwothree" isn't set to either
> > > > "1$two$three" or "12$three". Instead, the variable three is being translated
> > > > and attached to "$two", and Gui4Cli is looking for variable "two3", which it
> > > > fails to find.
> > > >
> > > >
> > > >
> > >
> >
> >
>




------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/gui4cli/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/gui4cli/join
    (Yahoo! ID required)

<*> To change settings via email:
    [email protected] 
    [email protected]

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/