Implement subtraction between IpNetworks
#126
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This is a helper type that is created with an initial IPv4 network, represented as a `u32` representation of its IP address and a `u8` with its prefix, and the prefix of a larger IPv4 network which contains the initial IPv4 network. Since one is contained in the other, the IP address shares most bits, so it only needs to keep track of it once. However, proper care must be taken so that the initial network prefix is larger than the larger network prefix. The iterator will produce `Ipv4Network` items representing increasing network ranges that are still inside the larger network but do not include the initial network.
This is also an iterator, and it wraps around the `Ipv4NetworkSubSet` type. However, it also includes entries for when the minuend and the subtrahend don't overlap and for when the subtrahend completly ovelaps the minuend. In the former case, there is no overlap of the networks. Therefore, nothing is subtracted from the original network (the minuend), so the result is just the original network. In the latter case, since the subtrahend completly overlaps the minuend, it is in effect completely removing the original network. Therefore, the result is no networks at all, and the resulting iterator is empty.
The resulting set is returned through a `Ipv4NetworkSubResult` helper iterator.
This allows one to subtract multiple `Ipv4Network`s from a single `Ipv4Network`. This works by having a subtraction chain, where the original minuend `Ipv4Network` is subtracted by the first `Ipv4Network` subtrahend, and the result is then subtracted by the second `Ipv4Network` subtrahend, and so on. The implementation uses dynamic dispatch because it is impossible to know how many subtrahends there will be. This is a simple initial implementation, and tere might be other solutions, such as creating a custom iterator and looping around all subtrahends to provide the results, but it quickly becomes more complicated. This might be doable as future work. Another possible future work is to remove the dynamic dispatch for known sizes of subtrahend arrays, but this would probably require const generics to be stabilized.
This is the IPv6 equivalent of the `Ipv4NetworkSubSet`. It is a helper type that is created with an initial IPv6 network, represented as a `u128` representation of its IP address and a `u8` with its prefix, and the prefix of a larger IPv6 network which contains the initial IPv6 network. Since one is contained in the other, the IP address shares most bits, so it only needs to keep track of it once. However, proper care must be taken so that the initial network prefix is larger than the larger network prefix. The iterator will produce `Ipv6Network` items representing increasing network ranges that are still inside the larger network but do not include the initial network.
This is the IPv6 equivalent to the `Ipv4NetworkSubResult` type. This is also an iterator, and it wraps around the `Ipv6NetworkSubSet` type. However, it also includes entries for when the minuend and the subtrahend don't overlap and for when the subtrahend completly ovelaps the minuend. In the former case, there is no overlap of the networks. Therefore, nothing is subtracted from the original network (the minuend), so the result is just the original network. In the latter case, since the subtrahend completly overlaps the minuend, it is in effect completely removing the original network. Therefore, the result is no networks at all, and the resulting iterator is empty.
The resulting set is returned through a `Ipv6NetworkSubResult` helper iterator.
This allows one to subtract multiple `Ipv6Network`s from a single `Ipv6Network`. This works by having a subtraction chain, where the original minuend `Ipv6Network` is subtracted by the first `Ipv6Network` subtrahend, and the result is then subtracted by the second `Ipv6Network` subtrahend, and so on. The implementation uses dynamic dispatch because it is impossible to know how many subtrahends there will be. This is a simple initial implementation, and tere might be other solutions, such as creating a custom iterator and looping around all subtrahends to provide the results, but it quickly becomes more complicated. This might be doable as future work. Another possible future work is to remove the dynamic dispatch for known sizes of subtrahend arrays, but this would probably require const generics to be stabilized.
This type represents the result of the subtraction of one `IpNetwork` from another. The implementation just wraps the resulting subtraction of the protocol specific (IPv4 or IPv6) subtraction result. Note that this means that it is not currently possible to subtract `IpNetwork`s of different protocols (i.e., either an IPv4 network from an IPv6 network or vice-versa).
The resulting set is returned through a `IpNetworkSubResult` helper iterator. Note that it is not currently possible to subtract `IpNetwork`s of different protocols (i.e., either an IPv4 network from an IPv6 network or vice-versa).
This allows one to subtract multiple `IpNetwork`s from a single `IpNetwork`. This works by having a subtraction chain, where the original minuend `IpNetwork` is subtracted by the first `IpNetwork` subtrahend, and the result is then subtracted by the second `IpNetwork` subtrahend, and so on. Since it isn't currently possible to subtract `IpNetworks` of different protocols, all operands must be of the same protocol (either IPv4 or IPv6). Having one operand of a different protocol will cause the code to panic. The implementation uses dynamic dispatch because it is impossible to know how many subtrahends there will be. This is a simple initial implementation, and tere might be other solutions, such as creating a custom iterator and looping around all subtrahends to provide the results, but it quickly becomes more complicated. This might be doable as future work. Another possible future work is to remove the dynamic dispatch for known sizes of subtrahend arrays, but this would probably require const generics to be stabilized.
Contributor
|
this pr is cool |
Open
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR implements the
Subtrait forIpNetwork,Ipv4NetworkandIpv6Network. There are two possibilities: either a single IP network item is subtracted from the original IP network, or multiple IP network items are subtracted from the original IP network.The result of the subtraction operation is always an iterator, which can contain zero or more IP network items. For subtracting a single item, the iterator has a rather simple representation in memory and does implement
Copy. When subtracting multiple items, a linked list ofBox<dyn Iterator>is returned, so the resulting type has a more complex representation.With this PR it is still not currently possible to subtract between IP networks with different protocols (for example, subtract an IPv4 network from an IPv6 network or vice-versa). Attempting to do so will lead the code to panic.