AaronFiore/electrum
1
0
Fork 0
Code Activity
13,433 Commits 2 Branches 181 Tags
7ca64ebbd8872e8ebe0b0c3849e2ec078e0dfcab
Commit Graph

13433 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
ThomasV
7ca64ebbd8 fix #7078 2021-03-03 15:35:38 +01:00
SomberNight
95b7c976e0 lnrouter.get_distances: fix exception due to rare race 
if the graph is being updated while the pathfinding is running,
channel_info might be None here
 2021-03-03 14:10:57 +01:00
ThomasV
3fa1aed8cd add unit test for basic mpp 2021-03-03 12:52:52 +01:00
ThomasV
ca6ecd56f2 fix route for non-trampoline MPP 2021-03-03 12:30:29 +01:00
ThomasV
1adde4c54a qt channels_list: use monospace font for channel capacity 2021-03-03 11:20:24 +01:00
ThomasV
6cf79dcfb2 qt: update all tabs when base_unit changes 2021-03-03 11:19:41 +01:00
ThomasV
222c70ada6 qt swap_dialog: use a single button to toggle direction 2021-03-03 10:45:46 +01:00
ThomasV
51f3c613a5 (minor) fix typos and indentation 2021-03-03 10:03:29 +01:00
SomberNight
44059ec116 lnpeer: more detailed logging in maybe_fulfill_htlc 2021-03-02 18:53:08 +01:00
SomberNight
9310e9023e test_lnpeer: add test: multihop payment that routes around failure 
This would have caught the overpayment bug fixed in
693583edc5
 2021-03-02 18:35:07 +01:00
SomberNight
750d8cfab5 lnworker: run create_route_for_payment end-to-end, incl private edges 
We pass the private edges to lnrouter, and let it find routes end-to-end.
Previously the edge_cost heuristics didn't apply to the private edges
and we were just randomly picking one of the route hints and use that.
So e.g. cheaper private edges were not preferred, but they are now.

PathEdge now stores both start_node and end_node; not just end_node.
 2021-03-02 18:00:31 +01:00
SomberNight
4445cef033 lnutil: turn global forwarding fee params into Channel attributes 
useful for unit testing, and it is the conceptually correct thing anyway
 2021-03-02 14:54:47 +01:00
ThomasV
259dacd56f Trampoline MPP aggregation: 
- trampoline node is the final recipient of MPP
 - each trampoline receives a bucket of HTLCs
 - if a HTLC from a bucket fails, wait for the entire bucket to fail
 - move trampoline route and onion code into trampoline module
 2021-03-02 14:28:15 +01:00
ThomasV
2da90add8f maybe_fulfill_htlc: add failure reason to logs 2021-03-02 14:26:16 +01:00
ThomasV
693583edc5 lnworker: amount passed in HtlcLog must be without fees 2021-03-02 10:23:30 +01:00
SomberNight
06ea06f7d3 _calc_routing_hints_for_invoice: incl max 15 chans to avoid qr overflow 
Prioritise channels that are likely to be able to receive the payment.
 2021-03-01 21:55:41 +01:00
SomberNight
6f894b9f88 _calc_routing_hints_for_invoice: also incl chans < invoice amount 
now that we can receive MPP
 2021-03-01 21:51:17 +01:00
SomberNight
1139720b58 lnworker: fix handle_error_code_from_failed_htlc for private channels 
if the last (private) edge of the route errors, we need to try other route hints (if any)
 2021-03-01 21:26:05 +01:00
SomberNight
b3b87555dc qt/kivy: lightning_tx_dialog: show LN invoice 2021-03-01 20:56:45 +01:00
SomberNight
c3ae1c0965 android build: (trivial) build arm64 apk first 
Build the apk I use to test with first :P
This is also the one mentioned in the README copy-paste code snippets.
 2021-03-01 20:45:52 +01:00
SomberNight
a9185b0846 follow-up prev 
kivy infers the type of the property from the default value.
without this, it was converting the str(Decimal(x)) to float
 2021-03-01 18:20:33 +01:00
SomberNight
6094f2751e kivy channel dialog: fix unit of displayed feerate 
The amount shown was in sat/kw, incorrectly labeled as sat/kbyte.
Show sat/vbyte instead.
 2021-03-01 17:09:04 +01:00
SomberNight
7d7dcf0795 qt/kivy ChannelsList: if node alias is unknown, display node id 
instead of "unknown"
 2021-03-01 16:41:54 +01:00
SomberNight
dd37151d65 qt ChannelsList: (trivial) format_fields should not know column order 2021-03-01 16:28:18 +01:00
SomberNight
bf7129d57e synchronizer/verifier: ensure fairness between wallets (follow-up) 
follow-up to 4346d2fc76

It's not just about the Synchronizer, the Verifier should not starve other jobs either...
(previously I thought the Verifier is not too important as it only makes
requests if there are new txs; however with LNWatcher its progress is not persisted)
 2021-03-01 13:08:01 +01:00
SomberNight
e0cfb2179d bech32: another around 10% speedup for bech32_decode 
turns out stdlib ord() is somewhat slow;
also, only lookup data chars once

benchmarked with:
```
import time
import electrum
from electrum.segwit_addr import bech32_decode

electrum.constants.set_testnet()

inv = "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"

def f():
  for _ in range(10000):
    addr = bech32_decode(inv, ignore_long_length=True)

t0 = time.time()
f()
t1 = time.time()
print(f"{t1-t0:.4f}")
```
 2021-02-28 18:32:48 +01:00
SomberNight
b83f7159a9 bech32: around 5% speedup for bech32_decode 
useful for lnaddr.lndecode
 2021-02-28 16:53:21 +01:00
SomberNight
d7597d96d0 lnaddr: 15x speedup for lndecode 
benchmarked with:

```
import time
import electrum
from electrum.lnaddr import lndecode

electrum.constants.set_testnet()

inv = "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"

def f():
  for _ in range(100):
    addr = lndecode(inv)

t0 = time.monotonic()
f()
t1 = time.monotonic()
print(f"{t1-t0:.4f}")
```
 2021-02-28 16:26:15 +01:00
SomberNight
2b693d3498 tests: fix test_lnpeer.test_payment_race 
broke in b6b13217b4
see changes to lnworker.htlc_fulfilled
 2021-02-28 13:58:31 +01:00
ThomasV
bc1ec6ac34 Qt: fix running GUI offline 2021-02-28 13:34:24 +01:00
ThomasV
5175a97671 test_payment_race: increase delay 2021-02-28 09:43:46 +01:00
ThomasV
f32d49b8ca revert 'keep invoice INFLIGHT', check HTLCs before payment attempt 2021-02-28 09:01:31 +01:00
ThomasV
38652cffb0 fix test_lnpeer (follow-up prev commit) 2021-02-27 20:51:19 +01:00
ThomasV
b6b13217b4 lnworker: keep invoice status INFLIGHT as long as HTLCs are inflight 2021-02-27 20:26:58 +01:00
SomberNight
2f223cdf46 qt channels dialog: fix for channel backups 2021-02-27 15:15:09 +01:00
SomberNight
d85e910262 logging: eliminate "fee_estimates" log spam 
only log fee estimates if they changed, instead of ~10 times per minute
 2021-02-27 12:27:26 +01:00
ThomasV
7f61f22857 MPP receive: allow payer to retry after mpp timeout 2021-02-27 11:48:14 +01:00
ghost43
0ce6adffcc Merge pull request #6968 from HardCorePawn/issue6664 
Added fiat fee estimate to Advanced Preview
 2021-02-27 08:31:17 +00:00
SomberNight
f9f49daad7 tx dialog: uniform high fee warnings between GUIs 2021-02-26 19:02:24 +01:00
SomberNight
fc3009918c follow-up prev: some clean-up 2021-02-26 16:44:07 +01:00
hcp
e01a2014b1 qt tx dialog: also display fiat amounts 2021-02-26 16:43:46 +01:00
SomberNight
84326cf1f7 qt tx dialog: add legend for input/output colouring 
based on e1d70bcd98
 2021-02-25 18:54:55 +01:00
ghost43
6fda9add28 Merge pull request #7026 from SomberNight/20210213_wallet_bumpfee 
wallet: refactor bump_fee; add new strategy; change Qt dialog to have "advanced" button
 2021-02-25 14:48:50 +00:00
SomberNight
d2019fd928 qt bump fee: rename "Final" checkbox to "Keep Replace-By-Fee enabled" 
Now that the checkbox is hidden behind an advanced option, there is
no need to be brief about it, better to be explicit.
(terminology unchanged for kivy.)
 2021-02-25 15:39:33 +01:00
SomberNight
4c36c45664 qt bump fee: add "advanced" button, allow choosing strategy 2021-02-25 15:39:29 +01:00
SomberNight
058d9ab6bb wallet.bump_fee: add new strategy: decrease payment amounts 
- Rename bump_fee "methods" to "strategies".
- Refactor strategies so that bump_fee can use any subset of them in any permutation.
- Adds a new strategy which decreases the payment outputs (instead of change).
 2021-02-25 15:32:07 +01:00
SomberNight
8fe7d750f7 qt: move RBF dialog out of main_window.py into its own file 2021-02-25 15:32:03 +01:00
SomberNight
254f57bce5 lnpeer.maybe_fulfill_htlc: fix error case FINAL_INCORRECT_HTLC_AMOUNT 
follow-up ef5a265449
 2021-02-25 12:49:06 +01:00
SomberNight
31bdb5c344 lnpeer.maybe_fulfill_htlc: follow BOLTs re some errors 
related: https://github.com/lightningnetwork/lightning-rfc/pull/608
 2021-02-25 12:12:44 +01:00
ThomasV
61e7f7e75e Merge pull request #7065 from SomberNight/20210224_mpp_recv_amt_sum 
lnpeer: MPP recv: only fulfill htlc if amt sum exact-matches total_msat
 2021-02-25 10:27:55 +01:00