Re^2: unpacking unsigned long longs

Replies are listed 'Best First'.
Re^3: unpacking unsigned long longs by ikegami (Patriarch) on Jul 06, 2009 at 16:12 UTC
The necessary info is in Config. `$ perl -MConfig -le'print $Config{ivsize}' 4` [download] You can see that value in `perl -V` as well. Of course, you could just try `unpack 'Q'`. `$ perl -le'unpack "Q", chr(0)x8; print "ok"' Invalid type 'Q' in unpack at -e line 1.` [download]	[reply] [d/l] [select]
Re^4: unpacking unsigned long longs by questionnaire (Novice) on Jul 06, 2009 at 17:11 UTC
I take it ivsize==4 is good for what I want?	[reply]
Re^5: unpacking unsigned long longs by ikegami (Patriarch) on Jul 06, 2009 at 17:23 UTC
Sorry. Like mine, your Perl uses 32-bit integers (8 bits/byte * 4 byte = 32 bits). It can't hold a 64-bit number (64 bits / 8 bits/byte = 8 bytes) unless you use something like Math::BigInt. You won't be able to use `unpack 'Q'`. Tell me which cell of my table you want, and I'll provide some code to do the same thing.	[reply] [d/l]
Re^4: unpacking unsigned long longs by questionnaire (Novice) on Jul 07, 2009 at 17:29 UTC
Alright, so I tried unpacking Q (and q) to no avail. But perl didn't spit out any errors when I try to assign 2^45 to a variable and then print it. Does that still mean perl isn't compiled to support 64 bit numbers? If so, how can I go about fixing this?	[reply]
Re^5: unpacking unsigned long longs by ikegami (Patriarch) on Jul 07, 2009 at 17:44 UTC
Did you really mean 2^45 (47) or 245 (35184372088832). A double-precision floating point number can handle numbers much bigger than 245. And with 53 bits of precision, it can represent 2*45 faithfully. If so, how can I go about fixing this?* See the "64 bit support" section in the perl distribution's `INSTALL` file for how to build such a perl. Or like I said already, tell me which cell of my table you want, and I'll provide some code to do the same thing on a 32-bit build of Perl.	[reply] [d/l]