By introducing a small limitation: only allowing entries in the change
graph to be addressable, instead of individual operations; we
drastically simplify the CRDT implementation.
There are four advantages:
1. Op ids are just regular SHA-1s
2. There's no need for relative IDs, ops never refer to other ops within the
same commit
3. There's no need for a nonce, since commits can't collide, and neither can op IDs
4. `OpId` can just be an alias of `EntryId`
The disadvantage of course, is that we have to be mindful of how we
create op transactions, since each transaction creates an addressable
unit. For example, we must not include multiple patch revisions in the
same transaction.
Before this change, the identity document would be loaded for the
COB signer. This meant that a fork would be needed for that signer.
After this change, it's no longer necessary, since the identity doc
head is computed from available remotes.
While making those changes, it was also apparent that some of the
identity-related functions had bad names and error types, this
was fixed as well.
It's fairly easy for a user to (by mistake or intentionally) create two
operations with the same OpId. This patch makes it much less likely,
and ensures that if the OpId is equal, it's because the operations
are identical.
The macro in the trailers module existed since there were multiple
trailers before the refactoring, so it minimised boilerplate code.
After the refactoring of the code there was only one trailer
left. Remove the macro generation code and replace it with just the
necessary code for the ResourceCommitTrailer.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
Rust 1.67 was announced[0]. Update the necessary files for running
1.67 and fix the clippy suggestions for string interpolation.
[0]: https://blog.rust-lang.org/2023/01/26/Rust-1.67.0.html
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
We were getting warnings in release mode, since the
debug code is not being run, and some vars are mutable
instead of immutable.
Signed-off-by: Alexis Sellier <alexis@radicle.xyz>
By using `0` as the clock value for an empty state (without operations),
we simplify transaction code. Previously, there was no clock value for
empty states, which meant that we had to sometimes use an `Option`.
* Move storage impl into `storage::cob`
* Rename `change::Create` to `change::Template`
* Rename `change::Storage::create` to `change::Storage::store`
Some of the renames are to further distinguish COBs from Changes, and
the different types of "create" operations within cobs.
Signed-off-by: Alexis Sellier <alexis@radicle.xyz>
We're using petgraph to encode an operation based CRDT (each graph node
is an operation) as a DAG, and one of the important things to ensure
is that the final state is not influenced by the order of concurrent
operations. In this case a concurrent operation would mean two potential
graph traversal orders, as there's no direct edge between the two
concurrent ops. This is essential a "partial order" of operations.
To be able to test this, we'd want some kind of control over how
neighbors are iterated over I guess, when there's more than one possible
sort order. Say we have a graph like this:
┌────────b◄─────┐
▼ │
a d
▲ │
└────────c◄─────┘
There are two possible topological traverse orders: [a, b, c, d] and [a,
c, b, d]. Having a way to go through these different orders would
be super handy.
One option would be to allow random order traversal. This would allow us
to test all orders by running the test enough times to likely test all
permutations.
Since petgraph doesn't support this, we implement our own simple DAG in
`radicle-dag`, which implements random-order topological orders.
For now, we don't make explicit use of these improvements, we simply
replace the underlying graph with our own.
Panic to clear to the operator something is wrong. Panic'ing like this
is ok in debug mode, but must be avoided in production.
Signed-off-by: Slack Coder <slackcoder@server.ky>
Collaborative object IDs are derived from their commit oid. These
IDs must be reproducible to support testing the client's example. All but
one of the parameters used to derive the Object ID is the commit's Author
and committer timestamp.
Introduce an environment variable 'RAD_COMMIT_TIME' to set both Author and
Committer timestamp when creating a commit in the cob module.
Use single variable, instead of Git's two (for author and committer), to
keep the changes simple. The function being modified returns a single
timestamp which is derived from the author's.
Signed-off-by: Slack Coder <slackcoder@server.ky>
We weren't accounting for batches ops in an entry.
This change should account for it by adding N-1
to the clock value of an entry, where N is the number
of entries.
Signed-off-by: Alexis Sellier <alexis@radicle.xyz>
Sometimes it is more efficient to perform multiple operations at once
and store those in the collaborative object graph together.
Change the contents to be a non-empty sequence of byte sequences,
i.e. NonEmpty<Vec<u8>>. These are stored as blobs in a git tree in
order by their index in the non-empty sequence. They are, in turn,
retrieved in order of their original index.
Change the conversion of an `EntryWithClock` to return a non-empty
sequence of `Op`s, which are in turn used to apply many changes in
each `from_history` implementation.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
The latter is a fork of the former, which adds const generics
and a bunch of other quality of life improvements that aren't
in the original crate due to it being unmaintained.
Signed-off-by: Alexis Sellier <alexis@radicle.xyz>
Add quickcheck tests for parse_refstr to ensure the correct
behaviour.
This required adding Arbitrary instances for TypeName and ObjectId, as
well as adding a mechanism for generating Invalid ObjectIds.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
Added the `parse_refname` function for parsing a reference into its
`TypeName` and `ObjectId`, if possible.
This replaces the `cob_suffix` function in `radicle`.
This change also fixes the `FromStr` instance of `ObjectId` to use
`from_str` instead of `TryFrom` for bytes.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
* Use the existing conventions for passing a signer to functions.
* Allow replacing git refs when updating cobs.
* Fix the parsing of a cob id.
Signed-off-by: Alexis Sellier <alexis@radicle.xyz>
Objects does not need to be separated by `local` and `remotes`, it
uses `iter` internally, so all references are used regardless.
Simplify Objects by having it be a simple vector of references.
This also means that the `identifier` is not needed. This was only
required to identify the local reference, so it can now be removed.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
The `Store` trait had a paramter for a `Resource`. This is too
restrictive for implementations, particularly `heartwood`.
Instead, where a unique identifier is needed for the
`object::Storage`, i.e. in `create` and `update`, the identifier
becomes a parameter.
This allows the removal of passing the resource in `list`, `get` and
`changegraph`.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
It will be common for Reference to be converted from a
`git2::Reference` when one is implementing `object::Storage`.
Provide a `TryFrom` implementation between `git2::Reference` and
`Reference`.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
`TypeName` and `ObjectId` are expected to be used in reference
names.
Add `From` conversions to `Component` for both of these types to make
it easier for using them with `git-ref-format`.
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
X-Clacks-Overhead: GNU Terry Pratchett
This is a minimal port of the collaborative objects code from
radicle-link[0].
A `CollaborativeObject` is made up of a `History` of `Change`s that
are expected to be stored in a `git` backend.
The library exposes a CRU series of functions `create`, `get`, `list`,
and `update`.
This implementation differs, to previous one, in that it does not
assume any kind of identity system -- its only assumption is that
identities have an identifier and can be addressed in the `git`
database. This means that it does not require an `author` to be passed
in and it only uses the `resource` (previously `authorizing_identity`)
to store its content-address in the `Change`.
It also foregoes any caching to simplify the port and not make any
early optimisation assumptions.
[0]: https://github.com/radicle-dev/radicle-link/tree/master/cob
Signed-off-by: Fintan Halpenny <fintan.halpenny@gmail.com>
Alexis Sellier
Fintan Halpenny
xphoniex
Alexis Sellier
Slack Coder
Dr Maxim Orlovsky
