Steven Allen
f518e34131
fix: atomically get head when registering an observer
...
This lets us always call check (accurately).
2021-08-30 16:43:21 -07:00
Steven Allen
82ac0a24a0
test: improve chain event tests
2021-08-30 16:43:21 -07:00
Steven Allen
3846170302
refactor events system
2021-08-30 16:43:21 -07:00
Steven Allen
a875e9ba73
fix: check parents when adding tipsets to the "cache"
2021-08-30 16:43:21 -07:00
Dirk McCormick
7b7a5b0b21
revert: changes to OnDealExpiredOrChanged in #5431 #7201
2021-08-30 10:42:41 +02:00
dirkmc
77a19774cf
fix events API timeout handling for nil blocks ( #7184 )
2021-08-27 09:05:00 +02:00
Peter Rabbitson
ed387b43cc
Bump go-multihash, adjust test for supported version
2021-07-26 17:50:19 +02:00
hunjixin
eef3fd5d7a
cache loaded block messages
2021-07-19 15:13:53 +08:00
Łukasz Magiera
3ea39f76e1
events: Fix handling of multiple matched events per epoch
2021-05-30 17:20:14 +02:00
Łukasz Magiera
81bd27911f
Propagate StateMsg api changes
2021-04-05 19:56:53 +02:00
Łukasz Magiera
deb2b90b6a
Fix lotus/miner build
2021-04-05 13:23:46 +02:00
frrist
0b579c1e03
feat(events): define Observer intreface for events
...
- allows tipset apply and revert to be observed
2021-03-22 15:37:44 -07:00
Raúl Kripalani
7f0f7d0b36
Merge branch 'master' into refactor/lib/blockstore
2021-02-28 19:55:23 +00:00
Łukasz Magiera
c59e2fea5b
Use correct contexts in storageadapter
2021-02-02 18:43:49 +01:00
Raúl Kripalani
d1104fec4c
rename blockstores for consistency.
2021-01-29 23:17:25 +00:00
Raúl Kripalani
b0cbc932bd
consolidate all blockstores in blockstore package.
2021-01-29 20:01:00 +00:00
Łukasz Magiera
e1be89b442
Merge pull request #5159 from iand/fix/event-hang
...
fix(events): avoid potential hang when starting event listener
2020-12-09 19:40:02 +01:00
Ian Davis
c6f5214657
fix(events): unlock mutex before error returns
2020-12-09 13:37:56 +00:00
Ian Davis
696469aae7
fix: avoid potential hang when starting event listener
...
It was possible for NewEvents to never return, blocked on waiting for a WaitGroup to be done.
The call to Done was in a goroutine that could exit before reaching the Done call.
Replace the WaitGroup with a channel that is closed to signal that initialisation is complete.
Also, while we are waiting on the channel, wait on the context so we can exit clealy if the
context is canceled.
2020-12-09 11:29:48 +00:00
Łukasz Magiera
eb2be69ace
events: use %v to log errors
2020-11-24 17:51:19 +01:00
yaohcn
7c0b6f41d8
fix log format
2020-11-24 19:09:48 +08:00
frrist
d25e5d1c08
polish: add DeffAddressMap method to init wrapper
...
- replace logic in predicates
2020-11-16 11:45:25 -08:00
Łukasz Magiera
961a25298b
predicates: Fast StateGetActor wrapper
2020-11-12 23:45:06 +01:00
Łukasz Magiera
c6a8efff7d
Merge pull request #4623 from filecoin-project/feat/cache-deal-state-matcher
...
Cache deal states for most recent old/new tipset
2020-10-31 01:00:11 +01:00
Dirk McCormick
79a8ff04fd
refactor: simplify chain event Called API
2020-10-30 14:00:32 +01:00
Dirk McCormick
f79652c28c
feat: cache deal states for most recent old/new tipset
2020-10-28 16:55:48 +01:00
Steven Allen
4e730b5ec8
port to v2 imports
2020-10-21 12:16:23 -07:00
Steven Allen
f32b3bf99b
fix a race in tipset cache usage
...
This tipset cache is shared between multiple services and is called from
multiple places.
2020-10-09 17:32:45 -07:00
Steven Allen
46f5b62a76
Remove a misleading miner actor abstraction
...
We shouldn't implement CBOR functions on the "abstract" miner info. Otherwise,
we could end up trying to actually _use_ this "abstract" info when decoding
state (which won't work across version).
Also remove the use of these CBOR functions, and instead explicitly use miner0
types. We'll have to abstract over versions eventually, but we'll probably need
some form of abstract miner builder (or maybe even adding some "add sector",
etc. functions to the current miner abstraction?
2020-09-23 14:49:53 -07:00
Aayush Rajasekaran
819180f739
Implement inefficient OnAddressMapChange predicate
2020-09-23 02:15:01 -04:00
Aayush Rajasekaran
476e7992e8
Add an error return to all actor state interface methods
2020-09-23 01:51:38 -04:00
Steven Allen
5314ba8c6d
remove ptr indirection
2020-09-22 10:55:29 -07:00
Steven Allen
1bf3b4989d
rename imports to match actors code
...
`sed -i 's/\bv0\(\w\)\(\w*\)/\L\1\E\20/g' **/*.go`
2020-09-18 14:59:27 -07:00
Aayush Rajasekaran
694463ffbe
More lint fixes
2020-09-18 03:06:28 -04:00
Aayush Rajasekaran
37de154a7c
Fixup tests
2020-09-18 01:56:21 -04:00
Steven Allen
daa441b989
simplify market diff
2020-09-17 21:48:50 -07:00
Steven Allen
f2a0779bb9
remove specs-actors import
2020-09-17 21:41:36 -07:00
Steven Allen
b2ee59024f
improve diff logic
...
* Make diffing work across versions.
* Start porting more chainwatch logic.
2020-09-17 21:39:34 -07:00
Łukasz Magiera
6eda53565f
Most tests passing
2020-09-17 17:30:24 +02:00
hannahhoward
691bd9f442
feat(markets): complete markets conversion
...
complete markets conversion to using chain/actors types, also replacing DealProposal/DealState
interfaces with structs
2020-09-17 00:43:14 -07:00
Aayush Rajasekaran
9e48dd211a
Fixups
2020-09-17 02:34:15 -04:00
Aayush Rajasekaran
b5ba7a0fad
Miner migration
2020-09-17 02:21:16 -04:00
hannahhoward
80b6994fe2
feat(market): update state diffing for market actor
...
Update to abstract actor for markets state diffing. Also move the diff adt functions inside the
abstract actors
2020-09-16 19:14:07 -07:00
hannahhoward
05c11531b1
feat(paych): convert paych actor
...
build abstraction for paych actor and switch to using it in payment channel manager and state
predicates
2020-09-15 21:06:04 -07:00
Steven Allen
9804310cc8
Update specs-actors to 0.9.10
...
Moves the Keyer type (and helpers) into go-state-types.
2020-09-14 14:32:57 -07:00
Raúl Kripalani
6d29d75724
Merge branch 'master' into inmem-journal
2020-09-14 12:17:45 +01:00
Aayush Rajasekaran
39755a294a
Update to specs v0.9.6
2020-09-07 15:48:41 -04:00
Dirk McCormick
76a1b3286b
fix: if cache best is nil, return chain head
2020-09-07 14:43:06 +02:00
Raúl Kripalani
efdfd3ee3e
Merge branch 'master' into inmem-journal
2020-08-26 16:38:23 +01:00
Steven Allen
5733c71c50
Lint everything
...
We were ignoring quite a few error cases, and had one case where we weren't
actually updating state where we wanted to. Unfortunately, if the linter doesn't
pass, nobody has any reason to actually check lint failures in CI.
There are three remaining XXXs marked in the code for lint.
2020-08-20 20:46:36 -07:00