Commit Graph

16 Commits

Author SHA1 Message Date
Byron Hambly
4ca1f8da5e
ci: update msrv and github actions toolchains 2023-01-10 14:57:09 +02:00
Byron Hambly
5a3675a06f
Update dependencies and rust-toolchain to 1.59
This commit updates the rust-toolchain to the current stable version
1.59, and fixes a number of new clippy warnings from that change.

Other changes:
    - updates backoff to 0.4
    - updates swap to 2021 edition
    - updates comfy-table to 5.0
    - updates monero-wallet to 2021 edition
    - updates moneor-harness to 2021 edition
    - updates bdk and rust_decimal
    - updates tokio-util to 0.7
    - updates workflow to use actions/setup-python@3
    - updates pem and serde_with
    - adds stable rust toolchain notice to readme
2022-03-12 13:05:36 +02:00
Your Name
665ea503d2 Make price ticker ws url configurable 2021-06-28 03:19:49 -07:00
Philipp Hoenisch
316f95c65b
Always log whole causality chain for errors. 2021-05-11 15:27:53 +10:00
Philipp Hoenisch
c011e95062
Improve logging by adding details as fields instead of into the msg. 2021-05-11 15:27:53 +10:00
Thomas Eizinger
654cfff2a8
Make kraken module emit PriceUpdates instead of Rates 2021-04-06 16:16:56 +10:00
Thomas Eizinger
cfc530e8ab
Make ask field of Rate private 2021-04-06 16:16:53 +10:00
Thomas Eizinger
bc46d95985
Remove unnecessary Serialize implementations 2021-04-06 16:08:42 +10:00
Thomas Eizinger
bac0f11898
Disable backoffs max_elapsed_time functionality
This config setting makes backoff stop retrying if we didn't get an
error within this timeframe.
For us, this results in backoff not actually doing anything.

The connection to kraken is very long-running. It might be active
for hours without failing. However, the default value for
`max_elapsed_time` is set to 15 minutes. As such, once the connection
fails any time after that, backoff doesn't actually retry the operation
but just gives up.

Fixes #303.
2021-03-11 13:33:46 +11:00
Thomas Eizinger
9ad2160c69
Refactor the kraken module to automatically re-connect on errors
In order to be able to re-connect on certain errors, we model
connection errors separately from parsing errors. We also change
the API of the whole module to no longer forward all errors to
the subscribers but instead, only update the subscribers with
either a latest rate or a permanent failure in case we exhausted
all our options to re-connect the websocket.

To model all of this properly, we introduce to sub-modules so that
each submodule can have their own `Error` type.

Resolves #297.
2021-03-09 17:37:37 +11:00
Thomas Eizinger
c560b3b21a
Introduce RateUpdate type alias to reduce duplication 2021-03-09 15:39:08 +11:00
Thomas Eizinger
fd5b38739a
Remove constant in favor of inlining string
This removes an import and simplifies the `connect_async` call.
2021-03-09 15:39:07 +11:00
Thomas Eizinger
142e5e2d3a
Use tracing macros by fully-qualifying them 2021-03-09 15:39:03 +11:00
Thomas Eizinger
1822886cd0
Provide stronger isolation of kraken module
Instead of leaking the tokio::sync::Receiver type in our
return value, we create a newtype that implements the desired
interface. This allows us to get rid of the `RateService` structs
and instead implement `LatestRate` directly on top of this struct.

Given that `LatestRate` is only used within the event_loop module,
we move the definition of this type into there.
2021-03-05 13:56:25 +11:00
Thomas Eizinger
f6ed4d65b5
Properly deal with additional messages sent from kraken 2021-03-05 13:40:39 +11:00
Thomas Eizinger
60d6d50246
Extract kraken websocket client into dedicated module 2021-03-05 13:37:51 +11:00