Commit graph

6 commits

Author SHA1 Message Date
Torkel Rogstad
4f4636448c Node cleanup (#591)
* Cleanup

In this commit we cleanup a few Scaladocs and change some
trait/impl pairs into just a case class.

* Node cleanup

In this commit we
1) Rename Client to P2PClient. Client is a very generic name.
2) Clean up some Scaladocs
3) Remove some unecessary objects/traits/classes and replace them
    with case classes
4) Add trace logging of bytes received in P2PClient
2019-07-13 09:26:00 -05:00
Torkel Rogstad
b3600f4888 DB loggers exposed in logging configuration 2019-06-27 15:28:08 +02:00
cwaldron97
a457717eb4 scalafmt 2019-06-24 14:12:47 -05:00
Torkel Rogstad
a87858247c Service identifier and node cleanup (#522)
* Log messages more granularely in node

* Add pong message

* Rework P2P service identifier

The old implementation didn't catch the notion
that service identifiers was a bit field where
multiple services could be set at the same time.

* Make Peer take InetSocketAddress

Previously it took NetworkIpAddress.
This doesn't make any sense, as we
need to know the services bitfield
to construct a NetworkIpAddress.

* Clean up logs and toString methods

* Clean up node README and main method

* Make all networks final case objects

* Skip publishing of wallet, node and chain

* Clean up logging of node and chain modules

* Add logging of max height in SpvNodeMain

* Clean up Scaladoc and toStrings

* Add logging configuration as symlinks in node

We'll have to figure out a more stable solution to
configure logging for the SPV node, but for the time
being this is an acceptable solution.

* Fix 2.11 compiler error
2019-06-17 14:27:51 -05:00
Torkel Rogstad
435aef7b0b Pass bloom filter to SPV node (#514)
* Add getAddressInfo to Wallet API

* Add Bloom filter and event callbacks to SPV node

* Move wallet test trait to testkit

* Test for connecting SPV node with wallet

* Tune logging in SPV node

* Clean up BloomFilter

* Proper toString in inventory and inventorymessage

* Actually pass in callbacks to SPV node

* Fix rebase screwup and partially respond to code review

* Default to file based databases in tests

This is due to issues with deadlocks with
in-memory based databases. We need DBs
to stay alive between connections, but
not across tests.

In DB intensive chain validation tests we
use in-memory databases instead. This
seems like a reasonable tradeoff between
simplicity and speed.

* Make SpvNodeCallbacks contain sequence of functions within

* Make Bloom filter non-optional in SpvNode
2019-06-17 06:48:26 -05:00
Torkel Rogstad
2723ef6f5d Move Logback config into test resources (#509)
I though I had solved our logging problem:
    We want projects to pick up logging config
    in core, without it being published.

The previous attempt was filtering out XML files
from unmanagedResources. That caused the XML
files to not get published, and when running
tests through Bloop everything worked fine.
However, when running them through sbt the
config did not get picked up. I suspect this
is because of a bug in how Bloop exports sbt
builds.

In this attempt we instead move the XML files
to test resources in core, and then make the
test projects depend on test sources in core,
with the awkward syntax 'core % "test->test"'.
2019-06-11 06:38:42 -05:00