Re: How to work with dots? When are they read as `./3` when as parts of an atom?

On 02/16/2014 06:20 AM, Shon Feder wrote:
> To clarify and correct my question: I'm mainly curious what the decisive
> factors are for whether the unquoted *a.b* gets read as an atom and when
> it's an operator expecting the *a *to be a dict.

I first foolishly thought it was possible to deal with that distinction
in a meaingful way, but it isn't, so, X.Y is read as .(X,Y), regardless
of the types of X and Y (well, except for floats).  The fact this worked
for floats also tricked me it could work for atoms :-(

	Cheers --- Jan

P.s.	Docs are updated, but the download page still claims a.b is
	and atom.  Will fix that.

>
> Also, I should note that I made the mistake of adding backquotes around *a.b
> *to the quote from the SWI download page. It should have read
>
>> Unquoted atoms may contain internal dots. E.g., a.b is an atom.
>
> Thanks again!
>
>
> On Sat, Feb 15, 2014 at 3:12 PM, Shon Feder <[email protected]> wrote:
>
>> Hi, all,
>>
>> I skimmed through the recent mailing list subjects and didn't see this
>> addressed elsewhere. But I am sorry if I overlooked an existing discussion!
>>
>> I just fired up the newest development version, and am running *SWI-Prolog
>> (Multi-threaded, 64 bits, Version 7.1.8)* on OS X 9.1.
>>
>> According to the note on the page for the latest development release,
>>
>>> Unquoted atoms may contain internal dots. E.g., `a.b` is an atom.
>>
>> But the dot seems to treated exclusively as the operator for functions on
>> dicts:
>>
>>      ?- atom(a.b).
>>      ERROR: Type error: `dict' expected, found `a' (an atom)
>>
>> Same thing if I try and use it as a functor:
>>
>>      ?- compound(a.b(X)).
>>      ERROR: Type error: `dict' expected, found `a' (an atom)
>>
>> Can anyone help clarify for my sake?
>>
>> Many thanks!
>>
>> Shon
>>
>> *P.S. I'm seeing some buggy interface elements on the new (much prettier!)
>> SWI site. Is this a known issue, or would it be helpful for me detail what
>> I'm seeing?*
>>
> -------------- next part --------------
> HTML attachment scrubbed and removed
> _______________________________________________
> SWI-Prolog mailing list
> [email protected]
> https://lists.iai.uni-bonn.de/mailman/listinfo.cgi/swi-prolog
>