Add `test_reject_invalid_min_final_cltv_delta` which is supposed to test
that the peer rejects incoming htlcs with final cltv delta differing
from what has been requested in the lightning invoice.
refactor `htlc_switch` to new architecture to make it more robust
against partial settlement of htlc sets and increase maintainability.
Htlcs are now processed in two steps, first the htlcs are collected into
sets from the channels, and potentially failed on their own already.
Then a second loop iterates over the htlc sets and finalizes only on
whole sets.
# Conflicts:
# electrum/lnpeer.py
Add unittest to TestPeerForwarding which sends a multi trampoline
payment.
Wait another htlc_switch iteration in tests because trampolines might have different delays
There is a race when initiating multiple lightning payments concurrently
(e.g. when doing a reverse swap with prepayment + swap payment).
suggest_splits might overallocate
split amounts for a channel as the splitting of both invoice amounts runs
concurrently and before acutal htlcs that reduce the channels balance
have been added to the channel yet. This results in a "not
enough balance" PaymentFailure once we try to send the htlcs and the
other payment attempt already reduced the available balance of the
channel.
This fix takes a lock from splitting the amount until the htlcs are
put on the channel, so suggest_splits always acts on the correct channel
balance.
Implement logic to claim a reverse swap funding output to any given
address. This allows to do onchain payments to external recipients
through a submarine swap.
Feerate is passed to `Commands._get_fee_policy()` as str which then
tried to multiply the string by 1000. Now it first casts the string to
`Decimal` and multiplies the decimal.
Fixes https://github.com/spesmilo/electrum/issues/10315
I needed such a tx to test something in electrumx and wanted to copy-paste one from the electrum unit tests. Weird that we were lacking such a test case, I was fairly certain there was one already...
Instead of storing its own path, each StoredDict element stores
its own key and a pointer to its parent. If a dict is removed
from the db, its parent pointer is set to None. This makes
self.path return None for all branches that have been pruned.
This passes tests/tests_json_db.py and fixes issue #10000
This is an optimisation and possible hardening against traffic analysis.
After a new block is mined, we sometimes receive "blockchain.scripthash.subscribe" notifications. If so, this is often due to the just mined block including mempool txs we already knew about.
Normally we would call the "blockchain.scripthash.get_history" RPC, to get full history of the affected scripthash.
Instead now we first optimistically guess that all mempool txs touching this scripthash just got mined and see if by assuming that we can reproduce the announced sh status.
- if yes, we saved a network RTT by not having to call "blockchain.scripthash.get_history"
- if no, we request the history from the server using the RPC
Some regtest tests depend on manual fee injection to simulate certain
mempool conditions (e.g. lnwatcher_waits_until_fees_go_down). This is
done by manually injecting fee estimates into the `Network` object using
the `test_inject_fee_etas` cli command. However it can still happen that
the Network automatically updates its fee estimates from the connected
electrum server in the time between injecting the fee and the actual
tested logic making decisions based on the fee. This causes the test to
fail sometimes.
By setting the `test_disable_automatic_fee_eta_update` true the Network
will stop automatically updating the fee estimates and the test will
behave as expected.
Prevents the client from accidentally connecting to a server on a
different network.
I noticed its possible to connect to mainnet servers on a signet
instance causing the recent peers to get populated with mainnet peers
rendering the wallet instance barely usable. Doing this check should
prevent this and similar issues.
..also export preimage in check_hold_invoice return value if available.
I intentionally did not return the preimage in the returned dict of
wallet.export_requests as this seems risky to do considering some users
of the cli might forward the response to a payer and the payserver
exposes it too.
Closes https://github.com/spesmilo/electrum/issues/10176
Enforce that the information used to create a bolt11 invoice using
`get_bolt11_invoice()` is similar to the related instance of PaymentInfo
by requiring a PaymentInfo as argument for `get_bolt11_invoice()`.
This way the invoice cannot differ from the created PaymentInfo.
This allows to use the information in PaymentInfo for validation of
incoming htlcs more reliably.
To cover all required information for the creation of a b11 invoice the
PaymentInfo class has to be extended with a expiry and
min_final_cltv_expiry. This requires a db upgrade.
Renames RecvMPPResolution.ACCEPTED to .COMPLETE as .ACCEPTED is somewhat
misleading. Accepted could imply that the preimage for this set has been
revealed or that the set has been settled, however it only means that we
have received the full set (it is complete), but the set still can be
failed (e.g. through cltv timeout) and has not been claimed yet.
util.callback_mgr.callbacks was not getting properly cleared between tests.
Every time an Abstract_Wallet or an LNWorker (or many other subclasses of EventListener) is instantiated,
self.register_callbacks() is called in __init__, which puts callbacks into util.callback_mgr.callbacks.
These are only cleaned up if we explicitly call Abstract_Wallet.stop() or LNWorker.stop() later, which we usually do not do in the tests.
As a result, when running multiple unit tests in a row, lots of objects created in a given testcase are never GC-ed and leak into subsequent tests. This is not only a memory leak, but wastes compute too: when events are triggered and cbs get called, these old objects also have their cbs called.
After running all (~1061) unit tests, I observe util.callback_mgr.callbacks had 30 events with a total of 3156 callbacks stored.
On my laptop, running all unit tests previously took ~115 sec, and now it takes ~73 sec.
