On Fri, Nov 20, 2009 at 8:57 AM, David Ashley <[email protected]>wrote:
> Rick -
>
> I have been having a long running discussion with Fernando on some other
> topics and this came up. This is definitely a bug in 4.0 IMHO. Any time you
> use a quoted string as the stem tail it trys to pass that string to the
> address environment for processing. Here is some sample output from
> rexxtry.rex.
>
> a."xxx yy z" = 'hello'
> sh: 0: command not found
> rc = 127 ...................................... rexxtry.rex on LINUX
>
> Am I just off my rocker? This has got to be a bug.
>
No, it does not have to be a bug. That is not a valid syntax for stem
evaluation, but is just a concatenation between the stem value and the
literal string. To use a literal string, you need to use the brackets to
cause the stem lookup, as others have already pointed out.
Rick
>
> David Ashley
>
>
>
> -------- Original Message -------- Subject: Re: Fwd: stems Date: Fri, 20
> Nov 2009 11:41:05 +0100 From: FERNANDO MANSITO CABALLERO
> <[email protected]> <[email protected]> To: David
> Ashley <[email protected]> <[email protected]>
>
> Dear Mr. Ashley,
>
> Thanks you very much for your patient and kind explanation.
>
> If you allow me, although the two expressions are indeed completely
> different the result of the two of them is the same: at present you can use
> any string enclosed in square brackets as a tail component, but *only if
> it is the final component:*
> **
> *a.[o sole mio] = "che si fa" --> che si fa*, but *a.[o sole mio].1 = "a
> la una" --> 0 is not a command*
> *a.["o sole mio"] = "e non si fa" --> e non si fa*, but *a.["o sole
> mio"].a = "y a las dos" --> 0 is not a command*
>
> Although I would never imagine that there be no pertinent reasons for the
> present situation, for my purposes I would prefer that issue solved
> the "tolerant" way (use any string as a tail component) and please not the
> reverse (illegalise the use of any bracketed string as the ending
> component).
>
> By the way, rule 1 (quoted strings become indexes) does not seem to work:
> *a."o sole mio" = "hello" --> 0 is not a command, a."pppp" = "hello" --> 0
> is not a command* (!)
>
> If I may, I have also noticed that if you EXPOSE a stem from within a
> PROCEDURE, USE ARG it and change some components, the stem reverts to
> uninitialised or, there being a default, to that default for all the
> unchanged components unless you recur to RETURNing the stem, which may be a
> bit rigid.
>
> Thank you very much again for your patience and best regards
> fernando mansito
>
> On Wed, Nov 18, 2009 at 9:57 PM, David Ashley
> <[email protected]>wrote:
>
>> Fernando -
>>
>> I believe your problem in the understanding of what is really going on
>> with the stem tail evaluations.
>>
>> a.[o sol mia] is NOT equivalent to a.["o sol mia"]
>>
>> When Rexx evaluates a stem tail one of two things happen.
>>
>> 1. If the tail contains quotes strings then then the quoted string become
>> the index.
>> 2. If the tail contains unquoted strings then they are considered to be
>> variable references and the value of the variable is substituted. If that
>> variable is not defined then the name of the variable is converted to upper
>> case and then substituted.
>>
>> Thus for your example
>>
>> a.[o sol mia] is equivalent to a.["O SOL MIA"] since none of those
>> tail variables has been defined. Thus a.["O SOL MIA"] is not equal to
>> a.["o sol mia"]. Those are two different indexes.
>>
>> I hope this helps. If this is unclear please let me know.
>>
>> David Ashley
>>
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus
> on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Oorexx-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
Oorexx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oorexx-devel
