Subj : Re: Dartmouth BASIC to C
To   : comp.programming
From : John Smith
Date : Thu Aug 18 2005 11:16 pm

spinoza1111@yahoo.com wrote:
> John Smith wrote:
> 
>>spinoza1111@yahoo.com wrote:
>>
>>>Looks great, with a minor cavil.
>>>
>>>You simulate Dartmouth Basic's READ/DATA facility which would give a
>>>system error message if there was not enough DATA to READ in the pool
>>>of DATA.
>>
>> From page 6 of the manual:
>>
>>"After commpleting the computation for the second set of right hand
>>sides and printing the answers, the computer will give the last values,
>>4 and -7, to the variable B1 and B2, compute and print the third set of
>>answers, and then stop, because there is no more data when the READ
>>statement 30 is encountered for the fourth time."
>>
>>and page 27:
>>
>>"Each time a READ statement is encountered anywhere in the program, the
>>data block supplies the next available number or numbers. If the data
>>block runs out of data, with a READ statement still asking for more, the
>>program is assumed to be done."
>>
>>My understanding of this is that the result of a READ after data are
>>exhausted is normal termination, not an error. I don't remember whether
>>this was the case with later flavors of BASIC, for example, the MS
>>Basics. I'll have to trot out my copy of qBasic and give it a try.
> 
> 
> Hmm, the compiler in my book (Build Your Own .Net Language and
> Compiler) produces an error message.
> 
> You'll have to explain to me how you can terminate "normally" when a
> READ DATA statement is executed, for example inside a For loop. 

You'll have to get Kemeny and Kurtz to explain it. I'm only quoting 
their manual. It may have been among the rough edges that got fixed in 
later versions.

If no
> error message is provided, this isn't normal termination. Instead, it's
> partial output and an unexplained termination.
> 
> However, this is better than the C program which will, I think, read
> values from random memory locations.
> 
>>JS
> 
>

.