How can I work out how many IP addresses there are in a given range?

Question

I'm wondering if there is a formula to work out the amount of IP addresses in a given range.

I have a system that takes either a single IP address or a start and and end IP address:

• A single one might be: 145.16.23.241
• A range might be: 145.16.23.122 - 145.16.23.144

The people inputting these aren't technical, nor are they receiving these details from technical people, so suggesting using CIDR notation isn't an answer for me.

I want to be able to highlight that when entering a range such as: 122.100.10.12 - 128.10.200.140 it might not actually be a range and rather 2 singular IP addresses since such a range would comprise of x (large number of) IP addresses.

Is there a basic formula to help me highlight this?

Answer 1

IP addresses are a 32 bit integer which we typically express as four octets for human purposes. You could get the number of addresses in a range by turning both into their 32 bit integer representations and subtracting.

Answer 2

You won't be able to get an accurate count without any information about whether the range crosses subnet boundaries and what those subnets are. For example, 10.10.7.0 - 10.10.8.255 is a contiguous range of 512 addresses in the context of a /16. If that range covers two /24 blocks, the count is 508 because each block has a network address and a broadcast address.

Making sure the entered upper bound is greater than or equal to the lower bound and using subtraction to measure the distance between them is good for a quick sanity check. You'll need to decide what's a reasonable difference before you raise the red flag.

After you've converted each octet to an 8-bit integer, converting the lot of them to a single 32-bit integer is a matter of doing this:

(octet1 << 24) | (octet2 << 16) | (octet3 << 8) | octet4

Answer 3

You have to be clear how you are defining these ranges. Maybe you meant all of the addresses whose 32-bit representations fall between the two numbers, and that's what the other answers here seem to assume. So for example, 1.0.0.0 - 1.0.2.0 includes all of the following IP addresses:

1.0.0.0 - 1.0.0.255
1.0.1.0 - 1.0.1.255
1.0.2.0

However, when I first read your question, what I thought you meant was defining separate ranges for each byte, in which case the range 1.0.0.0 - 1.0.2.0 would mean only the following three addresses:

1.0.0.0
1.0.1.0
1.0.2.0

For the first case, as others stated the correct answer is

abs(ip2-ip1)+1 

But in the second case, the correct answer would be

product(abs(ip2[i] - ip1[i]) + 1) 

where [i] means the ith byte of that IP, and the product is taken over the four bytes.

Make sure it's clear to your users which of these you mean.

Answer 4

Note that to properly convert dotted IP addresses to a 32-bit number and do mathy things on them reliably, you need to use an unsigned integer type. If the only integer types available to you are signed, then...

Another way to calculate it is this: starting with the first octet that is different, multiply each octet's difference from that point on by the following numbers, and then add these differences together:

1: 16777216
2: 65536
3: 256
4: 1

For example, using your range 122.100.10.12 - 128.10.200.140:

  128  10   200  140
- 122  100   10   12
=   6 -90   190  128

= 16777216 * 6 
+ 65536 * -90
+ 256 * 190
+ 1 * 128
= 94813824

This is the same value you'd get doing the whole conversion of

0x800AC88C = 2148190348
0x7A640A0C = 2053376524

and then subtracting.

Additionally, you can also perform the same operation this way:

((a[0] - b[0]) * 256 + a[1] - b[1]) * 256 + a[2] - b[2]) * 256 + a[3] - b[3]

This is easily susceptible to a reduce that collapses each next pair of values repeatedly.