[PD-dev] loaderHEX bug report

[Hans-Christoph Steiner]

On Nov 23, 2005, at 10:38 AM, geiger wrote:

>
> On Tue, 22 Nov 2005, IOhannes m zmoelnig wrote:
>> günter geiger wrote:
>>> On Tue, 22 Nov 2005, IOhannes m zmoelnig wrote:
>>>
>>>> but probably it would be nice to hear somebody else's opinion on  
>>>> this
>>>> (at least from those who are not entirely bored of this thread)
>>>
>>>
>>> Not an opinion,but an alternative proposal would be to use something
>>> non-numeric instead of 0x, like "$" for example.
>>>
>>
>> looking into kernighan-ritchie it says (if i read the c-syntax-graph
>> correctly) that the function-name is an identifier which in turn has  
>> to
>> match the regular expression "[_A-Za-z]{[_A-Za-z0-9]}*" (transcription
>> to regex done by me, so it might be erroneous)
>>
>> it doesn't say anything about non-alphanumeric characters (except for
>> the underscore), so i am a bit afraid of that.
>
> Well, a bad example then, what about "x0" instead of "0x", or whatever
> doesnt start as a numeric value.

0x was chosen since it is the way that C declares hex values.  Instead  
of creating some arbitrary syntax, why don't we stick to the well known  
ones and save the brain space for other details.  HTML/XML style  
&entities; won't work, nor will emacs style \201 characters.

.hc


________________________________________________________________________ 
____

If you are not part of the solution, you are part of the problem.
                                                                          
                            - Eldridge Cleaver