php.net |  support |  documentation |  report a bug |  advanced search |  search howto |  statistics |  random bug |  login

go to bug id or search bugs for

Request #68480 unpack unsigned large number becoming negative
Submitted: 2014-11-22 12:07 UTC Modified: 2014-11-25 17:52 UTC
From: ignasimg at gmail dot com Assigned:
Status: Wont fix Package: *General Issues
PHP Version: Irrelevant OS: Irrelevant
Private report: No CVE-ID: None
View Developer Edit
 [2014-11-22 12:07 UTC] ignasimg at gmail dot com 
Description:
------------
According to the documentation page about unpack:
http://php.net/manual/en/function.unpack.php

Caution
[...] If you unpack a large unsigned long and it is of the same size as PHP internally stored values the result will be a negative number even though unsigned unpacking was specified.



My question (beside wtf?) is would it be so hard to specify inside the unpack function that if unsigned was required make it become unsigned? 
I mean... it might sound stupidly ridicoulous but why is there an unsigned format if it acts the same as with signed?

There's even some funny solution on the same page comments:

if ($unpackedVal <0)
{
      $unpackedVal += 4294967296;
}

which is platform dependent and probably inefficient.

But since I have no idea about the inner workings of PHP my question again is would it be so difficult to make it behave properly?

Patches

Pull Requests

History

AllCommentsChangesGit/SVN commitsRelated reports
 [2014-11-25 17:52 UTC] aharvey@php.net
-Status: Open +Status: Wont fix
 [2014-11-25 17:52 UTC] aharvey@php.net 
PHP doesn't have an unsigned integer (or bigint) type, so there's no way to return it as an unsigned integer unless the number gets returned as a float, which obviously risks losing precision. (This is effectively what the "solution" does — on a 32 bit architecture, that addition will cause $unpackedVal to be converted to a float.)

Closing, since this requires a language change (which then requires an RFC, a patch, et cetera).
 [2014-12-01 00:49 UTC] ignasimg at gmail dot com 
"so there's no way to return it as an unsigned integer unless the number gets returned as a float" so do it!

The only problem you point out is the possible loss of precission, but afaik loss of precission only occurs with fractional values and values that are much larger than the maximum integer. 

And even if it was the case that some big integer value say X did loss precission when converting to float. This precission would be lost aswell when you do:

$a = X;

because as said php would store it as a float.


Maybe I'm missing something but I really can't understand why it can't return a float type.
 
PHP Copyright © 2001-2025 The PHP Group
All rights reserved. 		Last updated: Wed Dec 17 19:00:02 2025 UTC