Re: How to read a line of text?

[Ron Stodden]

Re 4 below, I have just discovered the neat (swi-prolog) ability to 
specify a literal character code in the form 0\'z.   This makes things a 
lot easier!      Except in editing, where kwrite under Prolog 
highlighting wrongly takes the ' as the start of a quote.    For any 
line with an odd number of 0\'x character code representations, add a ' 
in a comment to fix this.

Unfortunately, I cannot locate any similar facility in GNU Prolog. 

Anybody, how are literal character codes expressed in GNU Prolog source 
in terms of the ASCII character they represent?


Ron Stodden wrote:

> Thanks, Alex. but your code has problems:
>
> 1.     It will not deal with an empty stream.  IOW the test for (not) 
> end of stream must immediately follow the repeat.  gprolog does not 
> support a 'not' operator.
>
> 2.     My experiments show that list decomposition is limited to the 
> head item only.    IOW [Head|Tail] works, but [Item1|Item2|Tail] will 
> not produce your expected result.
>
> 3.     Your case is limited to the processing of single lines.   For 
> example, using your technique one could not output a  list of all the 
> lines in a file because all data is lost at the fail of a repeat fail 
> combination.   That would require the use of persistent storage, ie a 
> database or global variables.
>
> 4.     The requirement to specify words as code lists is insufferably 
> tedious.

--
Ron. [Melbourne, Australia]
"If you keep a green bough in your heart, the singing bird will come"
Get Fastest Mandrake downloader, English-only, from:
http://members.optusnet.com.au/ronst/    <--- Last Change 2nd November!