Commit Graph

431 Commits

Author SHA1 Message Date
niftynei
6030d0542a bcli: print error message in debug log for broadcast failures 2020-12-24 11:38:49 +01:00
niftynei
47d9ff88df bcli: only print error if exitcode is non-zero 2020-12-24 11:38:49 +01:00
niftynei
08c8581461 bcli: allocate temp string off of short-lived cmd
`cmd` gets cleaned up almost immediately, and is the right scope for a
temporary string allocation
2020-12-24 11:38:49 +01:00
niftynei
f9a2489424 bcli: print sendrawtx every time (even on failures) 2020-12-24 11:38:49 +01:00
Rusty Russell
52af729641 plugins/offer and plugins/fetchinvoice: send and recv errors.
This also lets us extend our testing to cover error cases.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
96caf9f4ab fetchinvoice: return the next period for recurring offers.
This is useful for the caller to know when to call again.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
69ddf818c7 plugins/fetchinvoice: catch invoice_request reply.
Using onionmessage hook, we get the response and either present it
to the user (invoice) or return the error to the user.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
3eada358e9 plugins/fetchinvoice: plugin to send an invoice_request for a given offer
Doesn't catch the reply yet, but prepares the invoice request based on
the offer and sends it.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
9d75fbe237 plugins/offers: add code to catch invoice_request onion messages.
We prepare an invoice, but there's no code to send an error reply yet
(future patch).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
08e110b568 JSON: offers plugin offer command.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
26e0cbeb7c plugin/offers: skeleton for offers plugin (EXPERIMENTAL_FEATURES)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-23 14:24:39 +01:00
Rusty Russell
723c16072a cleanups: feedback from Christian Decker review.
1. Hoist 7200 constant into the bolt12 heade2.
2. Make preimage the last createinvoice arg, so we could make it optional.
3. Check the validity of the preimage in createinvoice.
4. Always output used flag in listoffers.
5. Rename wallet offer iterators to offer_id iterators.
6. Fix paramter typos.
7. Rename `local_offer_id` parameter to `localofferid`.
8. Add reference constraints on local_offer_id db fields.
9. Remove cut/paste comment.
10. Clarify source of fatal() messages in wallet.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-15 12:05:02 +01:00
Rusty Russell
9fe612db20 plugins/pay: allow bolt12 invoices (EXPERIMENTAL_FEATURES).
Note that we remove the redundant "is this the correct chain?"
check, since bolt11_decode and bolt12_decode do that internally
anyway (this was changed in 924cc04bd2).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-15 12:05:02 +01:00
Rusty Russell
a33e39b7e8 pay, wallet: rename internal bolt11 vars to invstring.
And handle bolt12 strings if EXPERIMENTAL_FEATURES.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-15 12:05:02 +01:00
Christian Decker
3b599b846e keysend: Check that the destination supports keysend upon init
We were blindly initiating the keysend payment, which could lead to
confusing outcomes. This adds a very specific error message to the
error returned.

Changelog-Fixed: keysend: Keysend now checks whether the destination supports keysend before attempting a payment. If not a more informative error is returned.
2020-12-11 15:50:09 +01:00
Christian Decker
e186b2620a txprepare: Verify that outputs arg is an array
We were not checking that outputs is indeed an array, and just going
ahead creating the array of outputs. Since `tok->size` for a string is
0 we ended up ignoring the argument altogether and thus the created
transaction would end up only with a single change output.

Fixes #4258
2020-12-09 06:56:21 +10:30
Christian Decker
55d9492d9e txprepare: Use param_outpoint_arr helper to validate input 2020-12-09 06:56:21 +10:30
Rusty Russell
c5a2e64fd4 plugins/libplugin-pay: expose gossmap for plugin special effects.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-04 20:16:54 -06:00
Rusty Russell
54c57e7495 libplugin-pay: don't expose bolt11 details.
When we support bolt12, this won't exist.  We only need min_final_cltv_expiry,
routes and features, so put them into struct payment explicitly.

We move the default final ctlv out to the caller, too, which is clearer.
e.g. keysend was using this value, but it was hard to tell.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-04 20:16:54 -06:00
João Paulo
7514d7fb48 plugins: add txprepare to .gitignore
Changelog-None
2020-12-04 11:30:58 +10:30
niftynei
4185a86d90 df-mfc: set the minimum input weight to 110 iif v2
We only need a bumped up input weight if we're talking to nodes that
require v2.
2020-12-02 14:19:08 +10:30
niftynei
b01734087d df-mfc: set protocol type when we know it
We'll use it for figuring out whether or not to set a utxo witness
minimum, which comes much before we were setting this field.

Now we set the protocol as soon as we can reasonably deduce it.
2020-12-02 14:19:08 +10:30
Rusty Russell
9361b62e3a libplugin: add command_hook_success helper.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-12-02 10:38:04 +10:30
niftynei
29c3532856 mfc-psbt: mark all of our inputs as "ours", then only sign those
we only want to sign the inputs that we've reserved via utxopsbt or
fundpsbt. we mark them with a flag (reusing the now defunct max-len
flag is fine), then look for inputs with that flag to pass to signonly
2020-11-23 12:41:05 -06:00
niftynei
6077eca660 df: pass back 'close_to' for completed/commitment secured channels
When commitments are secured, also return the 'close_to' script if we've
got a local_upfront_shutdown_script set.
2020-11-23 12:41:05 -06:00
niftynei
c6b45e052b mfc-df: after openchannel_signed is finished, we call finished
We done!?
2020-11-23 12:41:05 -06:00
niftynei
3e19b6c8f5 mfc-df: after psbt signed, send to openchannel_signed if v2s
If there's an v2 destinations, they'll broadcast the tx for us
2020-11-23 12:41:05 -06:00
niftynei
a34425abd1 mfc-df: after sigs are collected, go sign the psbt 2020-11-23 12:41:05 -06:00
niftynei
d7c06b5b0e mfc-df: once we've gotten the PSBT finalized, we wait for peer sigs
We need our peer's signatures to arrive before we sign/broadcast the
funding transaction (but only if there's v2 peers present)
2020-11-23 12:41:05 -06:00
niftynei
c90a19f739 mfc-df: only add outputs for v1 outs; go to openchannel_update if v2s
We only have output scripts for v1 protocols after the
fundchannel_start/openchannel_init round. We need to add them before
we get into the openchannel_update rounds, however, so we do that here.
2020-11-23 12:41:05 -06:00
niftynei
3d2c0951d5 mfc-df: rework how openchannel_update works, callbacks 2020-11-23 12:41:05 -06:00
niftynei
381658dee6 mfc-df: merge openchannel_init/fundchannel_start results
These happen simultaneously, and should resolve to the same place when
they're finished.
2020-11-23 12:41:05 -06:00
niftynei
95de8b1174 mfc-df: have both paths use redo_multifundchannel
Still need handling of failure for v2 opens (no rpc exists?!)
2020-11-23 12:41:05 -06:00
niftynei
a31b078721 mfc: consolidate to a single FAILED state 2020-11-23 12:41:05 -06:00
niftynei
991ae65e9e mfc-df: track destination's openchannel version
Plus method to help count/tabulate how many of each we've got
around still.
2020-11-23 12:41:05 -06:00
niftynei
e0c4865eea mfc: add a 'fail_destination' helper
Caches state at which we failed + sets error
2020-11-23 12:41:05 -06:00
niftynei
5b6b012af9 mfc: add happy path-way for v2 in multifundchannel
Tested and works with both sides funding! Yay.

Doesn't do any amount of reasonable cleanup or handling of errors.
2020-11-23 12:41:05 -06:00
Christian Decker
313976e2f4 pay: Cleanup the route applicability checks for channel hints
I previously mistyped the rather lengthy conditions for failures, so
let's dissect it into its smaller components and add rationale behind
the individual parts of the decision.
2020-11-17 14:04:19 +10:30
Christian Decker
4d6b4a0445 pay: Retry the route computation if we could not apply the chanhints
This adds a new state `PAYMENT_STEP_RETRY_GETROUTE` which is used to
retry just that one step, without spawning a completely new
attempt. It's a new state so that modifiers do not act on it twice.

Changelog-Fixed: pay: Improved the performance of the `pay` command considerably by avoiding conflicting changes to our local network view.
2020-11-17 14:04:19 +10:30
Christian Decker
544e110c96 pay: Add a pre-apply check to channel_hint updates
This allows us to atomically update all channel_hints and determine if
we had a collision and therefore should retry.
2020-11-17 14:04:19 +10:30
Christian Decker
83f57ac300 pay: Move the chanhint update up to payment_getroute
We were delaying the channel_hint update till after the `createonion`
call which gave us the same situation we had with concurrent
`getroute` calls. Now we update the hints as soon as the plugins have
had their say in the route construction. If we still fail, either
because a modifier changed the route causing the failure, or because
we interleaved the route computation for multiple parts, we reset the
attempt and retry inline (i.e., without creating a new sub-payment).

Notice that interleaved route computations now only happen if the
modifier makes an async call to some RPC or similar.
2020-11-17 14:04:19 +10:30
Rusty Russell
62c52fe868 libplugin: add support for before and after deps on hooks.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-11-09 15:22:33 -06:00
Rusty Russell
cd5a93d0bd gossmap: fix reutrn of gossmap_xxx_has_feature, rename.
1. One place returned false instead of -1.
2. The names implied it returned a bool, and it doesn't.

Fix both, and curse C's loose typing a little.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-10-28 15:32:12 +10:30
Rusty Russell
572c849873 libplugin: ignore incoming notifications.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-10-23 13:53:16 +10:30
Rusty Russell
127811757f libplugin: support for sending notifications.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: libplugin: routines to send notification updates and progress.
2020-10-23 13:53:16 +10:30
Rusty Russell
a8177e9013 Makefile: make check-includes check all the non-generated files.
Note that check-whitespace and check-bolt already do this, so we
can eliminate redundant lines in common/Makefile and bitcoin/Makefile.

We also include the plugin headers in ALL_C_HEADERS so they get
checked.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-10-22 12:14:34 +10:30
Rusty Russell
5e865ce42b Makefile: unify generated files definition.
We change gen_ to _gen, because filtering on gen_% doesn't work if they're
in subdirectories :(

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2020-10-22 12:14:34 +10:30
niftynei
f8c4cc73ae mfc: use declared constants
reduce, reuse, recycle for a greener world
2020-10-21 11:24:16 +10:30
niftynei
e66937e012 mfc: Add a commitment_feerate optional parameter to multifundchannel
Technically there *are* two feerates that we need to know:
  - the feerate to use for the funding transaction, and
  - the feerate to tell our peer to use for our commitment txs/htlc txs

As written, `multifundchannel` uses the same feerate for both. This
optional parameter will allow us to differentiate between the two, which
will be exceedingly handy for anchor output worlds. ;)

FIXME: test this

Changelog-Added: JSON API: `multifundchannel` has a new optional argument, 'commitment_feerate', which can be used to differentiate between the funding feerate and the channel's initial commitment feerate
2020-10-21 11:24:16 +10:30
Rusty Russell
b470ae2c73 plugins/libplugin-pay: use gossmap.
This is a fairly direct translation.  Even so, it should be faster in
most cases, and and we can do more sophisticated things if we want.

This also handles disabled channels better.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: plugins: `pay` will now try disabled channels as a last resort.
2020-10-21 08:58:34 +10:30