In this commit, we move the serialisation details of a channel's
features to the DB layer and change the `models` field to instead use a
more useful `*lnwire.FeatureVector` type.
This makes the features easier to work with and moves the serialisation
to where it is actually used.
Replace all usages of the "github.com/go-errors/errors" and
"github.com/pkg/errors" packages with the standard lib's "errors"
package. This ensures that error wrapping and `errors.Is` checks will
work as expected.
Here we fix a bug which would not include the selected incoming blinded
path chain from being included in the selected path set if it meets the
minimum length requirement. The appropriate unit test and itest is
updated to demonstrate the fix.
Now that we know that the InboundFee on the ChannelEdgePolicy is always
set appropriately, we can update the GraphCache UpdatePolicy method to
take the InboundFee directly from the ChannelEdgePolicy object.
In this commit, we make sure to set the new field wherever appropriate.
This will be any place where the ChannelEdgePolicy is constructed other
than its disk deserialisation.
In this commit, we unify how all unit tests that make use of the graph
create their test ChannelGraph instance. This will make it easier to
ensure that once we plug in different graph DB implementations, that all
unit tests are run against all variants of the graph DB.
With this commit, `NewChannelGraph` is mainly only called via
`MakeTestGraph` for all tests _except_ for `TestGraphLoading` which
needs to be able to reload a ChannelGraph with the same backend. This
will be addressed in a follow-up commit once more helpers are defined.
Note that in all previous packages where we created a test graph using
`kvdb.GetBoltBackend`, we now need to add a `TestMain` function with a
call to `kvdb.RunTest` since the `MakeTestGraph` helper uses
`GetTestBackend` instead of `kvdb.GetBoltBackend` which requires an
embedded postgres instance to be running.
To clearly separate the mission control DB store (which for now will
remain kvdb only) from the graph store, we rename the `graphBackend`
variable to `mcBackend` in the `testGraphInstance` struct.
Create the ChanIDString function to return a string representation
of the route's channel IDs, formatting them in the order they appear
in the route (e.g., "chanID1 -> chanID2").
Discarded routes with a success probability lower than the minimum
threshold are now logged accordingly when finding a blinded path.
In this commit, the blindedPathRestrictions are expanded to include a
list of incoming channels that must be included in any blinded path. The
unit tests are expanded to test this new logic.
Mission control may have outdated success/failure amounts for node pairs
that have channels with differing capacities. In that case we assume to
still find the liquidity as before and rescale the amounts to the
according range.
We skip the evaluation of probabilities when the amount is lower than
the last success amount, as the probability would be evaluated to 1 in
that case.
If the success and fail amounts indicate that a channel doesn't obey a
bimodal distribution, we fall back to a uniform/linear success
probability model. This also helps to avoid numerical normalization
issues with the bimodal model.
This is achieved by adding a very small summand to the balance
distribution P(x) ~ exp(-x/s) + exp((x-c)/s), 1/c that helps to
regularize the probability distribution. The distribution becomes finite
for intermediate balances where the exponentials would be evaluated to
an exact zero (float) otherwise. This regularization is effective in
edge cases and leads to falling back to a uniform model should the
bimodal model fail.
This affects the normalization to be s * (-2 * exp(-c/s) + 2 + 1/s) and
the primitive function to receive an extra term x/(cs).
The previously added fuzz seed is expected to be resolved with this.
This test demonstrates an error found in a fuzz test by adding a
previously found seed, which will be fixed in an upcoming commit.
The following fuzz test is expected to fail:
go test -v -fuzz=Prob ./routing/
The bimodal model doesn't depend on the unit, which is why updating to
more realistic values doesn't require changes in tests.
We pin the scale in the fuzz test to not invalidate the corpus.
We can remove paymentFailureInfo since we can gate the result
interpretation on the source index, meaning that if we don't have a
source index, we deal with an unknown payment outcome because we
couldn't pinpoint the failing hop.
