The strto{l,ul,ull} functions do not set errno upon a successful return, so a
successful return from a maximally valued input could be misinterpreted as an
overflow error if errno happened already to be set to ERANGE before the call.
To guard against this edge case, always set errno to zero before calling these
functions if checking errno afterward.
Changelog-None
bookkeeper used to generate these as channel events, now lightningd does.
We also add a "journal" event, which we will need later too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We still output the fields, they're just always the currency of the node.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Changed: Plugins: `bookkeeper` now explicitly assumes every transaction is in the same currency as the node (true unless you added manually)
We're going to get rid of this concept, but the main change is that the
account_get_balance API can be drastically simplified:
account_get_credit_debit() accesses the raw fields, never fails, but
returns the a flag which tells us if the account doesn't actually have
any events.
The one place we care about the balance, calculate by hand. Then
account_get_balance() (and struct account_balance) can simply be
moved to th test.
Subtly, without the "GROUP BY" clause, you always get one row, even if
there are no rows (but the SUM are null).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's always set, and in fact we assume it is (journal entries are not
internal to lightningd, so we won't see them in
lightningd/notification.c: that comment is misleading).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This is not particularly relevant now (it's always the current time) but will be
useful when we implement the list commands.
Note that timestamp is set to be "u32" in various schemas. This will
only become a problem on Sun 07 Feb 2106 06:28:15 UTC. I apologize to
my grandchildren in advance.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We're going to store them in the db this way, so I thought I'd see what it looks like if
we lift that interface all the way through.
We use a struct, so that types are checked strictly.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Undocumented, but the first tag in the coin_movement notification is
considered the primary tag, and the others are optional. The
bookkeeper plugin relies on this!
Enforce that this is true, and in the process document in the code which
is the primary tag.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This means we can keep a pointer to the channel directly, *or* a string.
This avoids gratuitous formatting (on creation) and lookups (later).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
More readable for me. Also, change order so we definitely break
compilation on all callers (putting enum before amount).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The part id is *only* unique within a group. The payment_hash / partid / groupid tuple is unique.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Plugins: `coin_movement` notification with `part_id` field now always has `group_id` field.
Rather than converting to a generic coin_mvt struct, use these directly in
the notification, which is more explicit.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This makes our final balance not match our wallet:
1. We only spend the anchor when we need to boost the commitment tx,
which we don't always do (sometimes the peer does, sometimes it's
not worth it).
2. We don't put the UTXO in our wallet, because we don't consider it
"ours": anyone can spend it after 16 blocks.
We used to use the tag "ignored" for this, but that's overly complex
IMHO.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The rare case happened where a lockfile sha-sum contained a "Ctlv" which
spell-check complained about. Stupid lockfiles that don't know it is
actually "cltv"!
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
There was a problem with a ‘highlight’ that was misunderstood as a
spelling mistake in lib-wally. Since ‘hightlight’ is already filtered
out, we simply instruct grep to ignore upper/lower case when filtering.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
Changelog-Added: The `htlc_accepted_hook` now gets the TLV-stream
attached to the HTLC passed through as `extra_tlvs` and can replace it.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
Adds some testcases for custom tlvs, set by a htlc_accepted_hook. We
check that the custom tlvs replace the update_add_htlc_tlvs and get
forwarded to the peer. We also check that a malformed tlv will result in
a **BROKEN** behaviour.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
Add serializing and deserializing of the extra tlvs to to the
htlc_accepted_hook to allow plugin users to replace the tlv stream that
is attached to the update_add_htlc message on forwards.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
This appends the extra_tlvs to the internal channeld_offer_htlc wire
msg. We also recombine the extra_tlvs with the blinded path key for
forwarding htlcs.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
This appends the extra_tlvs to the internal wire htlcs "added" and
"existing" for the extra tlvs to be handed to lightningd.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
We currently only consider known tlv types in the internal
representation of a htlc. This commit adds the remaining unknown tlv
fields to the htlc as well. This is in prepareation to forward these to
the htlc_accepted_hook.
Signed-off-by: Peter Neuroth <pet.v.ne@gmail.com>
json_tok_streq(…) and json_get_member(…) are convenience wrappers for
json_tok_strneq(…) and json_get_membern(…) respectively. Unfortunately, using
them incurs a performance penalty in the common case where they are called with
a string literal argument because the compiler is unable to substitute a
compile-time constant in place of the buried call to strlen(…).
For example,
json_get_member(buf, tok, "example");
…will have worse performance than…
json_get_membern(buf, tok, "example", strlen("example"));
…because the former is forced to scan over "example" at run-time to count its
length whereas the latter is able to elide the strlen(…) call at compile time.
Hoist these convenience functions up into common/json_parse_simple.h and mark
them as inline so that the compiler can elide the strlen(…) call in the common
case of calling these functions with a string literal argument.
Changelog-None
We cannot use subd_req() here: replies will come out of order, and the
we should not simply assign the reponses in FIFO order.
Changelog-Fixed: lightningd: don't get confused with parallel ping commands.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
One reason why ping processing could be slow is that, once we receive
a message from the peer to send to a subdaemon, we don't listen for
others until we've drained that subdaemon queue entirely.
This can happens for reestablish: slow machines can take a while to
set that subdaemon up.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Requested-by: Michael at Boltz
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Changelog-Added: Plugins: `xpay` now publishes `pay_part_start` and `pay_part_end` notifications on every payment send attempt.
Implement the sending of `start_batch` and `protocol_batch_element` from `channeld` to `connectd`.
Each real peer wire message is prefixed with `protocol_batch_element` so connectd can know the size of the message that were batched together.
`connectd` intercepts `protocol_batch_element` messages and eats them (doesn’t forward them to peer) to get individual messages out of the batch.
It needs this to be able to encrypt them individiaully. Afterwards it recombines the now encrypted messages into a single message to send over the wire to the peer.
`channeld` remains responsible for making `start_batch` the first message of the message bundle.
