"Is the following construction even legal:
e:Error = {}"
Short answer: No
Longer answer:
(If I remember it correct), the "e:Error = { }" construct has already been
debated. Unfortunately it seems that this debate is not documented at all.
That discussion started with the same construct. The use of a empty filter
"{ }" stranded on the choice to use for signature or name matching for an
empty filter. The discussion ended with the conclusion that the only filter
were an empty could have handy was the Error filter.
The option to only allow "e:Error = { }" has been mentioned, but has been
rejected because that would mean that not all the filter types would behave
the same grammarwise and the Error filter would be hard coded in the grammar
(thus Filtertype = "Error" | String, instead of just String).
The best option to get a more clear grammar would be removing the option to
not use any braches. However this breaks backwards compatibility, which has
been a strong argument not to change such things. This to keep all the third
party Compose* examples in the world working and to try to keep the look of
the parent languages of Compose*.
I hope this answers your question, or at least that the other project
members remember this discussion.
Dirk
On 10/3/07, Michiel Hendriks <[EMAIL PROTECTED]> wrote:
>
> Is the following construction even legal:
>
> e:Error = {}
>
> The ARM only lists the following valid constructions:
>
> e:Error = { matching }
> e:Error = { [matching] subst }
> e:Error = { cond => matching }
> e:Error = { cond => [matching] subst }
>
> So, at least one filter element is required. COPPER also requires a
> filter element.
>
> A missing filter element could be interpreted as: <inner.*>
> So:
>
> e:Error = {}
>
> Would be equal to:
>
> e:Error = { <inner.*> }
>
> What is the desired construction? At least one filter element or
> syntactic sugar for <inner.*>?
>
>
> ps, I'm not happy with the "e:Error = { matching }"
> construction. Currently the matching (without any braches) expands to
> signature matching. But I think it's a bit confusing.
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Composestar-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/composestar-developers
>
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Composestar-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/composestar-developers
