Re: Re (tilly) 3: processing large files

The problem is that if your naming scheme is 32-bits, then it only has about 4 billion names.

I agree with the point that you make, but what is this about "names?" I thought we were discussing the number of available addresses (and thus bytes) in a file, not the number of names.

Comment on Re: Re (tilly) 3: processing large files

Replies are listed 'Best First'.
Re (tilly) 5: processing large files by tilly (Archbishop) on Jul 04, 2001 at 23:48 UTC
I consider a pointer to be the name of a byte in memory. Thus my mental model of the problem of seeking to a random spot in a large file is that you don't know how to name the location that you want. YMMV.	[reply]
Re: Re (tilly) 5: processing large files by sierrathedog04 (Hermit) on Jul 05, 2001 at 00:19 UTC
Good enough. Since Perl does not permit pointer arithmetic any distinction between addresses and names is meaningless. While I understand that internally Perl uses C uses pointers and even pointer arithmetic it is, as you say, only a mental map. Ultimately it is all just electrical charges on a circuit anyway. Perl itself doesn't really exist at runtime.	[reply]