🚀 go-pugleaf

In article <1992Sep28.003421.1246@RedBrick.COM> jfr@RedBrick.COM () writes:
> I can see the need for memory
> conservation to some degree, but certainly there is no performance
> benefit on most machines today (there may even be a performance penalty
> in obscure cases).  I can also see the need to store short ints on disk
> for databases since memory can multiply radically when you have large
> databases.  But for in-memory structures, particularly counters and indexes
> and thinks like that, there is simply no justification to use short ints.

  Why would you use a long when a short would suffice.  If your reading a file  
in 80 byte increments would you declare a 5000 byte string to hold the data.

>  Why would you EVER use NUMBERS in a program.

  Maybe you had to support multiple versions of functions in one program, don't  
flame me in this it is just a thought; a function that complies to some  
standard name ieeeblah().

  Basically your problem is that the coder used a short int without realizing  
that it only has two bytes and could only store -32768 to 32767.  It wasn't a  
problem with the types.  That's like saying I used a byte and I didn't realize  
that it could only hold 256 values.  Besides you could have made it unsigned  
and increased the size.

--
Robert John Andersen           
(919) 380-6523 (v,w)
andersen%reality@glv.com - NeXT mail!

Outside of a dog, a book is man's best friend.
Inside of a dog it's too dark to read.