[Ecls-list] si::c-uint-max & si::c-ulong-max

[Juan Jose Garcia Ripoll]

Juan Jose Garcia Ripoll wrote:

> Hundo wrote:
>
>> By the way, do you also get the behavior I reported in
>> my other message with evaluating '+nand causing a "Not
>> a number" error message? I'm seeing it on a couple of
>> different machines so I suspect it's not a local
>> problem or perhaps there's something I'm missing. 
>
> Yes, I noticed this. To save memory and avoid duplicated code, ECL 
> uses the C library for parsing numbers and it seems some libraries 
> accept "nan" as a valid floating point number :-) I am looking for a 
> workaround. 

Done. The number parser has been fixed, as it also has been the 
dectector of potential numbers, which now instructs ECL to write the 
symbol +1a32 in escaped form: |+1a32|.