This commit demonstrates that currently the Addresses field of a
LightningNode is not populated correctly if the node has no addresses if
the SQL graph store is being used. This will be fixed in an upcoming
commit.
If a method returns an error, we should assume all other parameters to
be nil unless the documentation explicitly says otherwise. So here, we
fix a log line where a dereference is made to an object that will be nil
due to an error being returned.
The comment of FetchChannelEdgesByID says that if the ErrZombieEdge
error is returned, then the ChannelEdgeInfo return parameter will also
still be populated and returned (ie, wont be nil). This commit updates
the SQLStore implementation of FetchChannelEdgesByID to do this. This is
needed to prevent nil dereference panics at any call-sites that depend
on the method working as it describes.
On startup, we currently blindly overwrite our node announcement info
for our source node. This includes overwriting the existing LastUpdate
timestamp. This can cause an issue with our new SQL backends in the case
that the new timestamp is not greater than the previously written
timestamp. So here, we first fetch the source node and make sure to use
a timestamp that is greater than the previosly set one.
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.
In this commit, a new `test_native_sql` build flag is defined. If this
build flag is used along with the `--use-native-sql` config option, then
the SQLStore implementation of the graphdb.V1Store will be initialised.
This is then used to run our itest suite against the new SQLStore graph
implementation.
NOTE that this only works for new nodes currently - the migration from
kv-to-sql is yet to be implemeneted.
And delete the old MakeTestGraph function.
Here we need to update some of our unit tests to ensure that any updates
to existing nodes or channel policies have newer timestamps. This is so
that we don't violate our SQL DB constraints that prevent updates to a
node or channel policy record if the new update is not newer (has a
newer LastUpdate time) than the currently presisted record).
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.
