2022-03-16 03:39:21 +03:00
|
|
|
{-# LANGUAGE TemplateHaskell #-}
|
|
|
|
|
2019-03-25 20:10:52 +03:00
|
|
|
module Hasura.RQL.DDL.EventTrigger
|
2021-09-24 01:56:37 +03:00
|
|
|
( CreateEventTriggerQuery,
|
|
|
|
runCreateEventTriggerQuery,
|
|
|
|
DeleteEventTriggerQuery,
|
|
|
|
runDeleteEventTriggerQuery,
|
|
|
|
dropEventTriggerInMetadata,
|
|
|
|
RedeliverEventQuery,
|
|
|
|
runRedeliverEvent,
|
|
|
|
InvokeEventTriggerQuery,
|
|
|
|
runInvokeEventTrigger,
|
|
|
|
-- TODO(from master): review
|
2023-03-14 15:27:17 +03:00
|
|
|
ResolveHeaderError (..),
|
2021-09-24 01:56:37 +03:00
|
|
|
getHeaderInfosFromConf,
|
2023-03-14 15:27:17 +03:00
|
|
|
getHeaderInfosFromConfEither,
|
2021-09-24 01:56:37 +03:00
|
|
|
getWebhookInfoFromConf,
|
|
|
|
buildEventTriggerInfo,
|
2023-02-20 17:19:14 +03:00
|
|
|
getSourceTableAndTriggers,
|
2022-04-11 14:24:11 +03:00
|
|
|
getTriggerNames,
|
|
|
|
getTriggersMap,
|
2022-04-21 10:19:37 +03:00
|
|
|
getTableNameFromTrigger,
|
2022-03-11 02:22:54 +03:00
|
|
|
cetqSource,
|
|
|
|
cetqName,
|
|
|
|
cetqTable,
|
|
|
|
cetqInsert,
|
|
|
|
cetqUpdate,
|
|
|
|
cetqDelete,
|
|
|
|
cetqEnableManual,
|
|
|
|
cetqRetryConf,
|
|
|
|
cetqWebhook,
|
|
|
|
cetqWebhookFromEnv,
|
|
|
|
cetqHeaders,
|
|
|
|
cetqReplace,
|
|
|
|
cetqRequestTransform,
|
|
|
|
cetqResponseTrasnform,
|
2022-09-09 11:26:44 +03:00
|
|
|
cteqCleanupConfig,
|
2022-11-29 20:41:41 +03:00
|
|
|
cteqTriggerOnReplication,
|
2022-09-09 11:26:44 +03:00
|
|
|
runCleanupEventTriggerLog,
|
2022-09-21 08:59:14 +03:00
|
|
|
runEventTriggerResumeCleanup,
|
2022-09-13 11:33:44 +03:00
|
|
|
runEventTriggerPauseCleanup,
|
2022-09-09 11:26:44 +03:00
|
|
|
MonadEventLogCleanup (..),
|
2022-09-13 11:33:44 +03:00
|
|
|
getAllEventTriggersWithCleanupConfig,
|
2022-09-15 14:45:14 +03:00
|
|
|
getAllETWithCleanupConfigInTableMetadata,
|
2023-04-25 14:22:27 +03:00
|
|
|
runGetEventLogs,
|
|
|
|
runGetEventInvocationLogs,
|
|
|
|
runGetEventById,
|
2021-09-24 01:56:37 +03:00
|
|
|
)
|
|
|
|
where
|
|
|
|
|
2022-09-13 11:33:44 +03:00
|
|
|
import Control.Lens (ifor_, makeLenses, (.~))
|
2021-09-24 01:56:37 +03:00
|
|
|
import Data.Aeson
|
2023-03-14 15:27:17 +03:00
|
|
|
import Data.Either.Combinators
|
2021-09-24 01:56:37 +03:00
|
|
|
import Data.Environment qualified as Env
|
2022-09-15 14:45:14 +03:00
|
|
|
import Data.Has (Has)
|
2021-09-24 01:56:37 +03:00
|
|
|
import Data.HashMap.Strict qualified as HM
|
2022-09-13 11:33:44 +03:00
|
|
|
import Data.HashMap.Strict qualified as Map
|
2021-09-24 01:56:37 +03:00
|
|
|
import Data.HashMap.Strict.InsOrd qualified as OMap
|
2022-03-15 11:41:03 +03:00
|
|
|
import Data.HashSet qualified as Set
|
2022-11-30 21:12:14 +03:00
|
|
|
import Data.Sequence qualified as Seq
|
2021-09-24 01:56:37 +03:00
|
|
|
import Data.Text qualified as T
|
|
|
|
import Data.Text.Extended
|
2022-06-05 23:27:09 +03:00
|
|
|
import Data.URL.Template (printURLTemplate)
|
2021-09-24 01:56:37 +03:00
|
|
|
import Hasura.Base.Error
|
|
|
|
import Hasura.EncJSON
|
2023-04-25 20:16:53 +03:00
|
|
|
import Hasura.Eventing.Backend
|
2022-09-15 14:45:14 +03:00
|
|
|
import Hasura.Eventing.EventTrigger (logQErr)
|
|
|
|
import Hasura.Logging qualified as L
|
2021-09-24 01:56:37 +03:00
|
|
|
import Hasura.Prelude
|
2022-03-08 03:42:06 +03:00
|
|
|
import Hasura.RQL.DDL.Webhook.Transform (MetadataResponseTransform, RequestTransform)
|
2022-04-27 16:57:28 +03:00
|
|
|
import Hasura.RQL.Types.Backend
|
2023-04-24 21:35:48 +03:00
|
|
|
import Hasura.RQL.Types.BackendType
|
2022-04-27 16:57:28 +03:00
|
|
|
import Hasura.RQL.Types.Common
|
|
|
|
import Hasura.RQL.Types.EventTrigger
|
|
|
|
import Hasura.RQL.Types.Eventing
|
2023-04-24 19:44:21 +03:00
|
|
|
import Hasura.RQL.Types.Headers (HeaderValue (..))
|
2022-04-27 16:57:28 +03:00
|
|
|
import Hasura.RQL.Types.Metadata
|
|
|
|
import Hasura.RQL.Types.Metadata.Backend
|
|
|
|
import Hasura.RQL.Types.Metadata.Object
|
|
|
|
import Hasura.RQL.Types.SchemaCache
|
|
|
|
import Hasura.RQL.Types.SchemaCache.Build
|
|
|
|
import Hasura.RQL.Types.SchemaCacheTypes
|
|
|
|
import Hasura.RQL.Types.Source
|
|
|
|
import Hasura.RQL.Types.Table
|
2021-09-24 01:56:37 +03:00
|
|
|
import Hasura.SQL.AnyBackend qualified as AB
|
|
|
|
import Hasura.Session
|
2022-09-09 11:26:44 +03:00
|
|
|
import Hasura.Tracing (TraceT)
|
2021-09-24 01:56:37 +03:00
|
|
|
import Hasura.Tracing qualified as Tracing
|
|
|
|
|
|
|
|
data CreateEventTriggerQuery (b :: BackendType) = CreateEventTriggerQuery
|
2022-08-01 12:32:04 +03:00
|
|
|
{ _cetqSource :: SourceName,
|
|
|
|
_cetqName :: TriggerName,
|
|
|
|
_cetqTable :: TableName b,
|
|
|
|
_cetqInsert :: Maybe (SubscribeOpSpec b),
|
|
|
|
_cetqUpdate :: Maybe (SubscribeOpSpec b),
|
|
|
|
_cetqDelete :: Maybe (SubscribeOpSpec b),
|
|
|
|
_cetqEnableManual :: Maybe Bool,
|
|
|
|
_cetqRetryConf :: Maybe RetryConf,
|
|
|
|
_cetqWebhook :: Maybe InputWebhook,
|
|
|
|
_cetqWebhookFromEnv :: Maybe Text,
|
|
|
|
_cetqHeaders :: Maybe [HeaderConf],
|
|
|
|
_cetqReplace :: Bool,
|
|
|
|
_cetqRequestTransform :: Maybe RequestTransform,
|
2022-09-09 11:26:44 +03:00
|
|
|
_cetqResponseTrasnform :: Maybe MetadataResponseTransform,
|
2022-11-29 20:41:41 +03:00
|
|
|
_cteqCleanupConfig :: Maybe AutoTriggerLogCleanupConfig,
|
|
|
|
_cteqTriggerOnReplication :: TriggerOnReplication
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
}
|
|
|
|
|
2022-03-11 02:22:54 +03:00
|
|
|
$(makeLenses ''CreateEventTriggerQuery)
|
|
|
|
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
instance Backend b => FromJSON (CreateEventTriggerQuery b) where
|
2021-09-24 01:56:37 +03:00
|
|
|
parseJSON = withObject "CreateEventTriggerQuery" \o -> do
|
|
|
|
sourceName <- o .:? "source" .!= defaultSource
|
|
|
|
name <- o .: "name"
|
|
|
|
table <- o .: "table"
|
|
|
|
insert <- o .:? "insert"
|
|
|
|
update <- o .:? "update"
|
|
|
|
delete <- o .:? "delete"
|
|
|
|
enableManual <- o .:? "enable_manual" .!= False
|
|
|
|
retryConf <- o .:? "retry_conf"
|
|
|
|
webhook <- o .:? "webhook"
|
|
|
|
webhookFromEnv <- o .:? "webhook_from_env"
|
|
|
|
headers <- o .:? "headers"
|
|
|
|
replace <- o .:? "replace" .!= False
|
2021-09-29 11:13:30 +03:00
|
|
|
requestTransform <- o .:? "request_transform"
|
2022-01-19 07:46:42 +03:00
|
|
|
responseTransform <- o .:? "response_transform"
|
2022-09-09 11:26:44 +03:00
|
|
|
cleanupConfig <- o .:? "cleanup_config"
|
2023-02-20 17:19:14 +03:00
|
|
|
when (isIllegalTriggerName name) $
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
fail "only alphanumeric and underscore and hyphens allowed for name"
|
|
|
|
unless (T.length (triggerNameToTxt name) <= maxTriggerNameLength) $
|
|
|
|
fail "event trigger name can be at most 42 characters"
|
|
|
|
unless (any isJust [insert, update, delete] || enableManual) $
|
|
|
|
fail "atleast one amongst insert/update/delete/enable_manual spec must be provided"
|
|
|
|
case (webhook, webhookFromEnv) of
|
|
|
|
(Just _, Nothing) -> return ()
|
|
|
|
(Nothing, Just _) -> return ()
|
2021-09-24 01:56:37 +03:00
|
|
|
(Just _, Just _) -> fail "only one of webhook or webhook_from_env should be given"
|
|
|
|
_ -> fail "must provide webhook or webhook_from_env"
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
mapM_ checkEmptyCols [insert, update, delete]
|
2022-12-21 20:14:07 +03:00
|
|
|
defTOR <- case defaultTriggerOnReplication @b of
|
|
|
|
Just (_, dt) -> pure dt
|
|
|
|
Nothing -> fail "No default setting for trigger_on_replication is defined for backend type."
|
|
|
|
triggerOnReplication <- o .:? "trigger_on_replication" .!= defTOR
|
2022-11-29 20:41:41 +03:00
|
|
|
return $ CreateEventTriggerQuery sourceName name table insert update delete (Just enableManual) retryConf webhook webhookFromEnv headers replace requestTransform responseTransform cleanupConfig triggerOnReplication
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
where
|
2021-09-24 01:56:37 +03:00
|
|
|
checkEmptyCols spec =
|
|
|
|
case spec of
|
|
|
|
Just (SubscribeOpSpec (SubCArray cols) _) -> when (null cols) (fail "found empty column specification")
|
|
|
|
Just (SubscribeOpSpec _ (Just (SubCArray cols))) -> when (null cols) (fail "found empty payload specification")
|
|
|
|
_ -> return ()
|
|
|
|
|
|
|
|
data DeleteEventTriggerQuery (b :: BackendType) = DeleteEventTriggerQuery
|
2022-08-01 12:32:04 +03:00
|
|
|
{ _detqSource :: SourceName,
|
|
|
|
_detqName :: TriggerName
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
instance FromJSON (DeleteEventTriggerQuery b) where
|
2021-09-20 22:49:33 +03:00
|
|
|
parseJSON = withObject "DeleteEventTriggerQuery" $ \o ->
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
DeleteEventTriggerQuery
|
|
|
|
<$> o .:? "source" .!= defaultSource
|
|
|
|
<*> o .: "name"
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
data RedeliverEventQuery (b :: BackendType) = RedeliverEventQuery
|
2022-08-01 12:32:04 +03:00
|
|
|
{ _rdeqEventId :: EventId,
|
|
|
|
_rdeqSource :: SourceName
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
instance FromJSON (RedeliverEventQuery b) where
|
2021-09-20 22:49:33 +03:00
|
|
|
parseJSON = withObject "RedeliverEventQuery" $ \o ->
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
RedeliverEventQuery
|
|
|
|
<$> o .: "event_id"
|
|
|
|
<*> o .:? "source" .!= defaultSource
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
data InvokeEventTriggerQuery (b :: BackendType) = InvokeEventTriggerQuery
|
2022-08-01 12:32:04 +03:00
|
|
|
{ _ietqName :: TriggerName,
|
|
|
|
_ietqSource :: SourceName,
|
|
|
|
_ietqPayload :: Value
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
instance Backend b => FromJSON (InvokeEventTriggerQuery b) where
|
2021-09-20 22:49:33 +03:00
|
|
|
parseJSON = withObject "InvokeEventTriggerQuery" $ \o ->
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
InvokeEventTriggerQuery
|
|
|
|
<$> o .: "name"
|
|
|
|
<*> o .:? "source" .!= defaultSource
|
|
|
|
<*> o .: "payload"
|
|
|
|
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
-- | This typeclass have the implementation logic for the event trigger log cleanup.
|
|
|
|
--
|
|
|
|
-- TODO: this doesn't belong here in the DDL folder, but should be part of
|
|
|
|
-- Hasura.Eventing. It could even be made a Service, since the whole point of it
|
|
|
|
-- is to implement features differently between OSS and Pro.
|
2022-09-09 11:26:44 +03:00
|
|
|
class Monad m => MonadEventLogCleanup m where
|
2022-09-15 14:45:14 +03:00
|
|
|
-- Deletes the logs of event triggers
|
2022-09-09 11:26:44 +03:00
|
|
|
runLogCleaner ::
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
SourceCache -> TriggerLogCleanupConfig -> m (Either QErr EncJSON)
|
2022-09-09 11:26:44 +03:00
|
|
|
|
2022-09-15 14:45:14 +03:00
|
|
|
-- Generates the cleanup schedules for event triggers which have log cleaners installed
|
|
|
|
generateCleanupSchedules ::
|
|
|
|
AB.AnyBackend SourceInfo -> TriggerName -> AutoTriggerLogCleanupConfig -> m (Either QErr ())
|
|
|
|
|
2023-02-03 15:27:53 +03:00
|
|
|
-- | `updateTriggerCleanupSchedules` is primarily used to update the
|
|
|
|
-- cleanup schedules associated with an event trigger in case the cleanup
|
|
|
|
-- config has changed while replacing the metadata.
|
|
|
|
--
|
|
|
|
-- In case,
|
|
|
|
-- i. a source has been dropped -
|
|
|
|
-- We don't need to clear the cleanup schedules
|
|
|
|
-- because the event log cleanup table is dropped as part
|
|
|
|
-- of the post drop source hook.
|
|
|
|
-- ii. a table or an event trigger has been dropped/updated -
|
|
|
|
-- Older cleanup events will be deleted first and in case of
|
|
|
|
-- an update, new cleanup events will be generated and inserted
|
|
|
|
-- into the table.
|
|
|
|
-- iii. a new event trigger with cleanup config has been added -
|
|
|
|
-- Generate the cleanup events and insert it.
|
|
|
|
-- iv. a new source has been added -
|
|
|
|
-- Generate the cleanup events and insert it.
|
|
|
|
-- v. the cron schedule for event trigger cleanup config has changed -
|
|
|
|
-- Delete cleanup events with older cron schedule and generate
|
|
|
|
-- cleanup events with new cron schedule.
|
|
|
|
updateTriggerCleanupSchedules ::
|
|
|
|
L.Logger L.Hasura ->
|
|
|
|
InsOrdHashMap SourceName BackendSourceMetadata ->
|
|
|
|
InsOrdHashMap SourceName BackendSourceMetadata ->
|
|
|
|
SchemaCache ->
|
|
|
|
m (Either QErr ())
|
|
|
|
|
2022-09-09 11:26:44 +03:00
|
|
|
instance (MonadEventLogCleanup m) => MonadEventLogCleanup (ReaderT r m) where
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runLogCleaner sourceCache conf = lift $ runLogCleaner sourceCache conf
|
2022-09-15 14:45:14 +03:00
|
|
|
generateCleanupSchedules sourceInfo triggerName cleanupConfig = lift $ generateCleanupSchedules sourceInfo triggerName cleanupConfig
|
2023-02-03 15:27:53 +03:00
|
|
|
updateTriggerCleanupSchedules logger oldSources newSources schemaCache = lift $ updateTriggerCleanupSchedules logger oldSources newSources schemaCache
|
2022-09-09 11:26:44 +03:00
|
|
|
|
2023-02-03 04:03:23 +03:00
|
|
|
instance (MonadEventLogCleanup m) => MonadEventLogCleanup (ExceptT e m) where
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runLogCleaner sourceCache conf = lift $ runLogCleaner sourceCache conf
|
2022-09-15 14:45:14 +03:00
|
|
|
generateCleanupSchedules sourceInfo triggerName cleanupConfig = lift $ generateCleanupSchedules sourceInfo triggerName cleanupConfig
|
2023-02-03 15:27:53 +03:00
|
|
|
updateTriggerCleanupSchedules logger oldSources newSources schemaCache = lift $ updateTriggerCleanupSchedules logger oldSources newSources schemaCache
|
2022-09-09 11:26:44 +03:00
|
|
|
|
2023-02-03 04:03:23 +03:00
|
|
|
instance (MonadEventLogCleanup m) => MonadEventLogCleanup (MetadataT m) where
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runLogCleaner sourceCache conf = lift $ runLogCleaner sourceCache conf
|
2022-09-15 14:45:14 +03:00
|
|
|
generateCleanupSchedules sourceInfo triggerName cleanupConfig = lift $ generateCleanupSchedules sourceInfo triggerName cleanupConfig
|
2023-02-03 15:27:53 +03:00
|
|
|
updateTriggerCleanupSchedules logger oldSources newSources schemaCache = lift $ updateTriggerCleanupSchedules logger oldSources newSources schemaCache
|
2022-09-09 11:26:44 +03:00
|
|
|
|
|
|
|
instance (MonadEventLogCleanup m) => MonadEventLogCleanup (TraceT m) where
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runLogCleaner sourceCache conf = lift $ runLogCleaner sourceCache conf
|
2022-09-15 14:45:14 +03:00
|
|
|
generateCleanupSchedules sourceInfo triggerName cleanupConfig = lift $ generateCleanupSchedules sourceInfo triggerName cleanupConfig
|
2023-02-03 15:27:53 +03:00
|
|
|
updateTriggerCleanupSchedules logger oldSources newSources schemaCache = lift $ updateTriggerCleanupSchedules logger oldSources newSources schemaCache
|
2022-09-09 11:26:44 +03:00
|
|
|
|
2023-02-20 17:19:14 +03:00
|
|
|
instance (MonadEventLogCleanup m) => MonadEventLogCleanup (StateT w m) where
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runLogCleaner sourceCache conf = lift $ runLogCleaner sourceCache conf
|
2023-02-20 17:19:14 +03:00
|
|
|
generateCleanupSchedules sourceInfo triggerName cleanupConfig = lift $ generateCleanupSchedules sourceInfo triggerName cleanupConfig
|
|
|
|
updateTriggerCleanupSchedules logger oldSources newSources schemaCache = lift $ updateTriggerCleanupSchedules logger oldSources newSources schemaCache
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
resolveEventTriggerQuery ::
|
|
|
|
forall b m.
|
|
|
|
(Backend b, UserInfoM m, QErrM m, CacheRM m) =>
|
|
|
|
CreateEventTriggerQuery b ->
|
2022-03-15 11:41:03 +03:00
|
|
|
m (Bool, EventTriggerConf b)
|
2022-11-29 20:41:41 +03:00
|
|
|
resolveEventTriggerQuery (CreateEventTriggerQuery source name qt insert update delete enableManual retryConf webhook webhookFromEnv mheaders replace reqTransform respTransform cleanupConfig triggerOnReplication) = do
|
2020-12-28 15:56:00 +03:00
|
|
|
ti <- askTableCoreInfo source qt
|
2018-09-19 15:12:57 +03:00
|
|
|
-- can only replace for same table
|
|
|
|
when replace $ do
|
2021-09-06 14:15:36 +03:00
|
|
|
ti' <- _tiCoreInfo <$> askTabInfoFromTrigger @b source name
|
2019-11-20 21:21:30 +03:00
|
|
|
when (_tciName ti' /= _tciName ti) $ throw400 NotSupported "cannot replace table or schema for trigger"
|
2018-09-19 15:12:57 +03:00
|
|
|
|
2018-09-05 14:26:46 +03:00
|
|
|
assertCols ti insert
|
|
|
|
assertCols ti update
|
|
|
|
assertCols ti delete
|
2018-09-19 15:12:57 +03:00
|
|
|
|
2019-02-14 10:37:59 +03:00
|
|
|
let rconf = fromMaybe defaultRetryConf retryConf
|
2022-11-29 20:41:41 +03:00
|
|
|
return (replace, EventTriggerConf name (TriggerOpsDef insert update delete enableManual) webhook webhookFromEnv rconf mheaders reqTransform respTransform cleanupConfig triggerOnReplication)
|
2018-09-05 14:26:46 +03:00
|
|
|
where
|
2021-09-20 10:34:59 +03:00
|
|
|
assertCols :: TableCoreInfo b -> Maybe (SubscribeOpSpec b) -> m ()
|
2022-10-04 00:49:32 +03:00
|
|
|
assertCols ti opSpec = for_ opSpec \sos -> case sosColumns sos of
|
2021-09-24 01:56:37 +03:00
|
|
|
SubCStar -> return ()
|
2021-09-09 14:54:19 +03:00
|
|
|
SubCArray columns -> forM_ columns (assertColumnExists @b (_tciFieldInfoMap ti) "")
|
2018-11-23 16:02:46 +03:00
|
|
|
|
2022-03-15 11:41:03 +03:00
|
|
|
droppedTriggerOps :: TriggerOpsDef b -> TriggerOpsDef b -> HashSet Ops
|
|
|
|
droppedTriggerOps oldEventTriggerOps newEventTriggerOps =
|
|
|
|
Set.fromList $
|
|
|
|
catMaybes $
|
|
|
|
[ (bool Nothing (Just INSERT) (isDroppedOp (tdInsert oldEventTriggerOps) (tdInsert newEventTriggerOps))),
|
|
|
|
(bool Nothing (Just UPDATE) (isDroppedOp (tdUpdate oldEventTriggerOps) (tdUpdate newEventTriggerOps))),
|
|
|
|
(bool Nothing (Just DELETE) (isDroppedOp (tdDelete oldEventTriggerOps) (tdDelete newEventTriggerOps)))
|
|
|
|
]
|
|
|
|
where
|
|
|
|
isDroppedOp old new = isJust old && isNothing new
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
createEventTriggerQueryMetadata ::
|
2022-09-15 14:45:14 +03:00
|
|
|
forall b m r.
|
|
|
|
( BackendMetadata b,
|
|
|
|
QErrM m,
|
|
|
|
UserInfoM m,
|
|
|
|
CacheRWM m,
|
|
|
|
MetadataM m,
|
|
|
|
BackendEventTrigger b,
|
|
|
|
MonadIO m,
|
|
|
|
MonadEventLogCleanup m,
|
|
|
|
MonadReader r m,
|
|
|
|
Has (L.Logger L.Hasura) r
|
|
|
|
) =>
|
2021-09-24 01:56:37 +03:00
|
|
|
CreateEventTriggerQuery b ->
|
2022-03-15 11:41:03 +03:00
|
|
|
m ()
|
2020-12-28 15:56:00 +03:00
|
|
|
createEventTriggerQueryMetadata q = do
|
2022-03-15 11:41:03 +03:00
|
|
|
(replace, triggerConf) <- resolveEventTriggerQuery q
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
let table = _cetqTable q
|
|
|
|
source = _cetqSource q
|
2020-12-28 15:56:00 +03:00
|
|
|
triggerName = etcName triggerConf
|
2021-03-15 16:02:58 +03:00
|
|
|
metadataObj =
|
2021-09-24 01:56:37 +03:00
|
|
|
MOSourceObjId source $
|
|
|
|
AB.mkAnyBackend $
|
|
|
|
SMOTableObj @b table $
|
|
|
|
MTOTrigger triggerName
|
2022-03-15 11:41:03 +03:00
|
|
|
sourceInfo <- askSourceInfo @b source
|
2022-08-23 11:49:51 +03:00
|
|
|
let sourceConfig = (_siConfiguration sourceInfo)
|
2022-09-15 14:45:14 +03:00
|
|
|
newConfig = _cteqCleanupConfig q
|
2022-08-23 11:49:51 +03:00
|
|
|
|
|
|
|
-- Check for existence of a trigger with 'triggerName' only when 'replace' is not set
|
|
|
|
if replace
|
|
|
|
then do
|
|
|
|
existingEventTriggerOps <- etiOpsDef <$> askEventTriggerInfo @b source triggerName
|
|
|
|
let droppedOps = droppedTriggerOps existingEventTriggerOps (etcDefinition triggerConf)
|
|
|
|
dropDanglingSQLTrigger @b (_siConfiguration sourceInfo) triggerName table droppedOps
|
2022-09-15 14:45:14 +03:00
|
|
|
|
|
|
|
-- check if cron schedule for the cleanup config has changed then delete the scheduled cleanups
|
|
|
|
oldConfig <- etiCleanupConfig <$> askEventTriggerInfo @b source triggerName
|
|
|
|
when (hasCleanupCronScheduleUpdated oldConfig newConfig) do
|
|
|
|
deleteAllScheduledCleanups @b sourceConfig triggerName
|
2022-10-04 00:49:32 +03:00
|
|
|
for_ newConfig \cleanupConfig -> do
|
2022-09-15 14:45:14 +03:00
|
|
|
(`onLeft` logQErr) =<< generateCleanupSchedules (AB.mkAnyBackend sourceInfo) triggerName cleanupConfig
|
2022-08-23 11:49:51 +03:00
|
|
|
else do
|
|
|
|
doesTriggerExists <- checkIfTriggerExists @b sourceConfig triggerName (Set.fromList [INSERT, UPDATE, DELETE])
|
2022-09-15 14:45:14 +03:00
|
|
|
if doesTriggerExists
|
|
|
|
then throw400 AlreadyExists ("Event trigger with name " <> triggerNameToTxt triggerName <<> " already exists")
|
2022-10-04 00:49:32 +03:00
|
|
|
else for_ newConfig \cleanupConfig -> do
|
2022-09-15 14:45:14 +03:00
|
|
|
(`onLeft` logQErr) =<< generateCleanupSchedules (AB.mkAnyBackend sourceInfo) triggerName cleanupConfig
|
2022-03-15 11:41:03 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
buildSchemaCacheFor metadataObj $
|
|
|
|
MetadataModifier $
|
|
|
|
tableMetadataSetter @b source table . tmEventTriggers
|
|
|
|
%~ if replace
|
|
|
|
then ix triggerName .~ triggerConf
|
|
|
|
else OMap.insert triggerName triggerConf
|
2020-12-28 15:56:00 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
runCreateEventTriggerQuery ::
|
2022-09-15 14:45:14 +03:00
|
|
|
forall b m r.
|
|
|
|
( BackendMetadata b,
|
|
|
|
BackendEventTrigger b,
|
|
|
|
QErrM m,
|
|
|
|
UserInfoM m,
|
|
|
|
CacheRWM m,
|
|
|
|
MetadataM m,
|
|
|
|
MonadIO m,
|
|
|
|
MonadEventLogCleanup m,
|
|
|
|
MonadReader r m,
|
|
|
|
Has (L.Logger L.Hasura) r
|
|
|
|
) =>
|
2021-09-24 01:56:37 +03:00
|
|
|
CreateEventTriggerQuery b ->
|
|
|
|
m EncJSON
|
2018-12-13 10:26:15 +03:00
|
|
|
runCreateEventTriggerQuery q = do
|
2022-03-15 11:41:03 +03:00
|
|
|
createEventTriggerQueryMetadata @b q
|
2020-12-08 17:22:31 +03:00
|
|
|
pure successMsg
|
2018-09-05 14:26:46 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
runDeleteEventTriggerQuery ::
|
|
|
|
forall b m.
|
2021-10-22 17:49:15 +03:00
|
|
|
(BackendEventTrigger b, MonadError QErr m, CacheRWM m, MonadIO m, MetadataM m) =>
|
2021-09-24 01:56:37 +03:00
|
|
|
DeleteEventTriggerQuery b ->
|
|
|
|
m EncJSON
|
2022-04-21 10:19:37 +03:00
|
|
|
runDeleteEventTriggerQuery (DeleteEventTriggerQuery sourceName triggerName) = do
|
|
|
|
sourceConfig <- askSourceConfig @b sourceName
|
|
|
|
tableName <- (_tciName . _tiCoreInfo) <$> askTabInfoFromTrigger @b sourceName triggerName
|
2021-09-24 01:56:37 +03:00
|
|
|
|
|
|
|
withNewInconsistentObjsCheck $
|
|
|
|
buildSchemaCache $
|
|
|
|
MetadataModifier $
|
2022-04-21 10:19:37 +03:00
|
|
|
tableMetadataSetter @b sourceName tableName %~ dropEventTriggerInMetadata triggerName
|
2021-09-09 14:54:19 +03:00
|
|
|
|
2022-04-21 10:19:37 +03:00
|
|
|
dropTriggerAndArchiveEvents @b sourceConfig triggerName tableName
|
2020-12-28 15:56:00 +03:00
|
|
|
|
2022-09-15 14:45:14 +03:00
|
|
|
deleteAllScheduledCleanups @b sourceConfig triggerName
|
|
|
|
|
2019-11-20 21:21:30 +03:00
|
|
|
pure successMsg
|
2018-09-07 14:51:01 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
runRedeliverEvent ::
|
|
|
|
forall b m.
|
|
|
|
(BackendEventTrigger b, MonadIO m, CacheRM m, QErrM m, MetadataM m) =>
|
|
|
|
RedeliverEventQuery b ->
|
|
|
|
m EncJSON
|
2020-12-28 15:56:00 +03:00
|
|
|
runRedeliverEvent (RedeliverEventQuery eventId source) = do
|
2021-09-06 14:15:36 +03:00
|
|
|
sourceConfig <- askSourceConfig @b source
|
|
|
|
redeliverEvent @b sourceConfig eventId
|
2020-12-28 15:56:00 +03:00
|
|
|
pure successMsg
|
2018-09-19 15:12:57 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
runInvokeEventTrigger ::
|
|
|
|
forall b m.
|
|
|
|
( MonadIO m,
|
|
|
|
QErrM m,
|
|
|
|
CacheRM m,
|
|
|
|
MetadataM m,
|
|
|
|
Tracing.MonadTrace m,
|
|
|
|
UserInfoM m,
|
|
|
|
BackendEventTrigger b
|
|
|
|
) =>
|
|
|
|
InvokeEventTriggerQuery b ->
|
|
|
|
m EncJSON
|
2020-12-28 15:56:00 +03:00
|
|
|
runInvokeEventTrigger (InvokeEventTriggerQuery name source payload) = do
|
2021-09-06 14:15:36 +03:00
|
|
|
trigInfo <- askEventTriggerInfo @b source name
|
2019-05-13 12:41:07 +03:00
|
|
|
assertManual $ etiOpsDef trigInfo
|
2021-09-24 01:56:37 +03:00
|
|
|
ti <- askTabInfoFromTrigger source name
|
2021-09-06 14:15:36 +03:00
|
|
|
sourceConfig <- askSourceConfig @b source
|
2021-07-27 11:05:33 +03:00
|
|
|
traceCtx <- Tracing.currentContext
|
|
|
|
userInfo <- askUserInfo
|
2021-09-06 14:15:36 +03:00
|
|
|
eid <- insertManualEvent @b sourceConfig (tableInfoName @b ti) name (makePayload payload) userInfo traceCtx
|
2019-05-13 12:41:07 +03:00
|
|
|
return $ encJFromJValue $ object ["event_id" .= eid]
|
|
|
|
where
|
2021-09-24 01:56:37 +03:00
|
|
|
makePayload o = object ["old" .= Null, "new" .= o]
|
2021-08-17 13:21:56 +03:00
|
|
|
|
2019-05-13 12:41:07 +03:00
|
|
|
assertManual (TriggerOpsDef _ _ _ man) = case man of
|
|
|
|
Just True -> return ()
|
2021-09-24 01:56:37 +03:00
|
|
|
_ -> throw400 NotSupported "manual mode is not enabled for event trigger"
|
2019-05-13 12:41:07 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
askTabInfoFromTrigger ::
|
|
|
|
(Backend b, QErrM m, CacheRM m) =>
|
|
|
|
SourceName ->
|
|
|
|
TriggerName ->
|
|
|
|
m (TableInfo b)
|
2021-09-06 14:15:36 +03:00
|
|
|
askTabInfoFromTrigger sourceName triggerName = do
|
2022-04-21 10:19:37 +03:00
|
|
|
schemaCache <- askSchemaCache
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
getTabInfoFromSchemaCache schemaCache sourceName triggerName
|
|
|
|
`onNothing` throw400 NotExists errMsg
|
2022-12-08 19:03:20 +03:00
|
|
|
where
|
|
|
|
errMsg = "event trigger " <> triggerName <<> " does not exist"
|
2022-04-21 10:19:37 +03:00
|
|
|
|
|
|
|
getTabInfoFromSchemaCache ::
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
Backend b =>
|
2022-04-21 10:19:37 +03:00
|
|
|
SchemaCache ->
|
|
|
|
SourceName ->
|
|
|
|
TriggerName ->
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
Maybe (TableInfo b)
|
2022-04-21 10:19:37 +03:00
|
|
|
getTabInfoFromSchemaCache schemaCache sourceName triggerName = do
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
tableCache <- unsafeTableCache sourceName $ scSources schemaCache
|
|
|
|
find (isJust . HM.lookup triggerName . _tiEventTriggerInfoMap) (HM.elems tableCache)
|
2021-01-09 02:09:15 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
askEventTriggerInfo ::
|
|
|
|
forall b m.
|
|
|
|
(QErrM m, CacheRM m, Backend b) =>
|
|
|
|
SourceName ->
|
|
|
|
TriggerName ->
|
|
|
|
m (EventTriggerInfo b)
|
2021-09-06 14:15:36 +03:00
|
|
|
askEventTriggerInfo sourceName triggerName = do
|
|
|
|
triggerInfo <- askTabInfoFromTrigger @b sourceName triggerName
|
|
|
|
let eventTriggerInfoMap = _tiEventTriggerInfoMap triggerInfo
|
|
|
|
HM.lookup triggerName eventTriggerInfoMap `onNothing` throw400 NotExists errMsg
|
2021-01-09 02:09:15 +03:00
|
|
|
where
|
2021-09-06 14:15:36 +03:00
|
|
|
errMsg = "event trigger " <> triggerName <<> " does not exist"
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
|
2023-04-25 14:22:27 +03:00
|
|
|
checkIfTriggerNameExists ::
|
|
|
|
forall b m.
|
|
|
|
(Backend b, CacheRM m) =>
|
|
|
|
SourceName ->
|
|
|
|
TriggerName ->
|
|
|
|
m (Bool)
|
|
|
|
checkIfTriggerNameExists sourceName triggerName = do
|
|
|
|
schemaCache <- askSchemaCache
|
|
|
|
-- TODO: The name getTabInfoFromSchemaCache is misleading here.
|
|
|
|
-- There is a JIRA ticket that addresses this (https://hasurahq.atlassian.net/browse/GS-535)
|
|
|
|
let tableInfoMaybe = getTabInfoFromSchemaCache @b schemaCache sourceName triggerName
|
|
|
|
case tableInfoMaybe of
|
|
|
|
Nothing -> pure False
|
|
|
|
_ -> pure True
|
|
|
|
|
Clean metadata arguments
## Description
Thanks to #1664, the Metadata API types no longer require a `ToJSON` instance. This PR follows up with a cleanup of the types of the arguments to the metadata API:
- whenever possible, it moves those argument types to where they're used (RQL.DDL.*)
- it removes all unrequired instances (mostly `ToJSON`)
This PR does not attempt to do it for _all_ such argument types. For some of the metadata operations, the type used to describe the argument to the API and used to represent the value in the metadata are one and the same (like for `CreateEndpoint`). Sometimes, the two types are intertwined in complex ways (`RemoteRelationship` and `RemoteRelationshipDef`). In the spirit of only doing uncontroversial cleaning work, this PR only moves types that are not used outside of RQL.DDL.
Furthermore, this is a small step towards separating the different types all jumbled together in RQL.Types.
## Notes
This PR also improves several `FromJSON` instances to make use of `withObject`, and to use a human readable string instead of a type name in error messages whenever possible. For instance:
- before: `expected Object for Object, but encountered X`
after: `expected Object for add computed field, but encountered X`
- before: `Expecting an object for update query`
after: `expected Object for update query, but encountered X`
This PR also renames `CreateFunctionPermission` to `FunctionPermissionArgument`, to remove the quite surprising `type DropFunctionPermission = CreateFunctionPermission`.
This PR also deletes some dead code, mostly in RQL.DML.
This PR also moves a PG-specific source resolving function from DDL.Schema.Source to the only place where it is used: App.hs.
https://github.com/hasura/graphql-engine-mono/pull/1844
GitOrigin-RevId: a594521194bb7fe6a111b02a9e099896f9fed59c
2021-07-27 13:41:42 +03:00
|
|
|
-- This change helps us create functions for the event triggers
|
|
|
|
-- without the function name being truncated by PG, since PG allows
|
|
|
|
-- for only 63 chars for identifiers.
|
|
|
|
-- Reasoning for the 42 characters:
|
|
|
|
-- 63 - (notify_hasura_) - (_INSERT | _UPDATE | _DELETE)
|
|
|
|
maxTriggerNameLength :: Int
|
|
|
|
maxTriggerNameLength = 42
|
2021-09-09 14:54:19 +03:00
|
|
|
|
2023-03-14 15:27:17 +03:00
|
|
|
-- Consists of a list of environment variables with invalid/missing values
|
|
|
|
newtype ResolveHeaderError = ResolveHeaderError {unResolveHeaderError :: [Text]} deriving (Show)
|
|
|
|
|
|
|
|
instance ToTxt ResolveHeaderError where
|
|
|
|
toTxt = commaSeparated . unResolveHeaderError
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
getHeaderInfosFromConf ::
|
|
|
|
QErrM m =>
|
|
|
|
Env.Environment ->
|
|
|
|
[HeaderConf] ->
|
|
|
|
m [EventHeaderInfo]
|
2021-09-09 14:54:19 +03:00
|
|
|
getHeaderInfosFromConf env = mapM getHeader
|
|
|
|
where
|
|
|
|
getHeader :: QErrM m => HeaderConf -> m EventHeaderInfo
|
|
|
|
getHeader hconf = case hconf of
|
|
|
|
(HeaderConf _ (HVValue val)) -> return $ EventHeaderInfo hconf val
|
2021-09-24 01:56:37 +03:00
|
|
|
(HeaderConf _ (HVEnv val)) -> do
|
2021-09-09 14:54:19 +03:00
|
|
|
envVal <- getEnv env val
|
|
|
|
return $ EventHeaderInfo hconf envVal
|
|
|
|
|
2023-03-14 15:27:17 +03:00
|
|
|
-- This is similar to `getHeaderInfosFromConf` but it doesn't fail when an env var is invalid
|
|
|
|
getHeaderInfosFromConfEither ::
|
|
|
|
Env.Environment ->
|
|
|
|
[HeaderConf] ->
|
|
|
|
Either ResolveHeaderError [EventHeaderInfo]
|
|
|
|
getHeaderInfosFromConfEither env hConfList =
|
|
|
|
if isHeaderError
|
|
|
|
then Left (ResolveHeaderError $ lefts headerInfoList)
|
|
|
|
else Right (rights headerInfoList)
|
|
|
|
where
|
|
|
|
isHeaderError = any isLeft headerInfoList
|
|
|
|
headerInfoList = map getHeader hConfList
|
|
|
|
getHeader :: HeaderConf -> Either Text EventHeaderInfo
|
|
|
|
getHeader hconf = case hconf of
|
|
|
|
(HeaderConf _ (HVValue val)) -> Right $ EventHeaderInfo hconf val
|
|
|
|
(HeaderConf _ (HVEnv val)) ->
|
|
|
|
(Right . EventHeaderInfo hconf) =<< getEnvEither env val
|
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
getWebhookInfoFromConf ::
|
|
|
|
QErrM m =>
|
|
|
|
Env.Environment ->
|
|
|
|
WebhookConf ->
|
|
|
|
m WebhookConfInfo
|
2021-09-20 16:14:28 +03:00
|
|
|
getWebhookInfoFromConf env webhookConf = case webhookConf of
|
2021-09-09 14:54:19 +03:00
|
|
|
WCValue w -> do
|
|
|
|
resolvedWebhook <- resolveWebhook env w
|
2022-06-05 23:27:09 +03:00
|
|
|
let urlTemplate = printURLTemplate $ unInputWebhook w
|
|
|
|
-- `urlTemplate` can either be the template value({{TEST}}) or a plain text.
|
|
|
|
-- When `urlTemplate` is a template value then '_envVarName' of the 'EnvRecord'
|
|
|
|
-- will be the template value i.e '{{TEST}}'
|
|
|
|
-- When `urlTemplate` is a plain text then '_envVarName' of the 'EnvRecord' will be the plain text value.
|
|
|
|
return $ WebhookConfInfo webhookConf (EnvRecord urlTemplate resolvedWebhook)
|
2021-09-20 16:14:28 +03:00
|
|
|
WCEnv webhookEnvVar -> do
|
|
|
|
envVal <- getEnv env webhookEnvVar
|
2022-06-05 23:27:09 +03:00
|
|
|
return $ WebhookConfInfo webhookConf (EnvRecord webhookEnvVar (ResolvedWebhook envVal))
|
2021-09-09 14:54:19 +03:00
|
|
|
|
2021-09-24 01:56:37 +03:00
|
|
|
buildEventTriggerInfo ::
|
|
|
|
forall b m.
|
|
|
|
(Backend b, QErrM m) =>
|
|
|
|
Env.Environment ->
|
|
|
|
SourceName ->
|
|
|
|
TableName b ->
|
|
|
|
EventTriggerConf b ->
|
2022-11-30 21:12:14 +03:00
|
|
|
m (EventTriggerInfo b, Seq SchemaDependency)
|
2022-11-29 20:41:41 +03:00
|
|
|
buildEventTriggerInfo
|
|
|
|
env
|
|
|
|
source
|
|
|
|
tableName
|
|
|
|
( EventTriggerConf
|
|
|
|
name
|
|
|
|
def
|
|
|
|
webhook
|
|
|
|
webhookFromEnv
|
|
|
|
rconf
|
|
|
|
mheaders
|
|
|
|
reqTransform
|
|
|
|
respTransform
|
|
|
|
cleanupConfig
|
|
|
|
triggerOnReplication
|
|
|
|
) = do
|
|
|
|
webhookConf <- case (webhook, webhookFromEnv) of
|
|
|
|
(Just w, Nothing) -> return $ WCValue w
|
|
|
|
(Nothing, Just wEnv) -> return $ WCEnv wEnv
|
|
|
|
_ -> throw500 "expected webhook or webhook_from_env"
|
|
|
|
let headerConfs = fromMaybe [] mheaders
|
|
|
|
webhookInfo <- getWebhookInfoFromConf env webhookConf
|
|
|
|
headerInfos <- getHeaderInfosFromConf env headerConfs
|
|
|
|
let eTrigInfo =
|
|
|
|
EventTriggerInfo
|
|
|
|
name
|
|
|
|
def
|
|
|
|
rconf
|
|
|
|
webhookInfo
|
|
|
|
headerInfos
|
|
|
|
reqTransform
|
|
|
|
respTransform
|
|
|
|
cleanupConfig
|
|
|
|
triggerOnReplication
|
|
|
|
tabDep =
|
|
|
|
SchemaDependency
|
|
|
|
( SOSourceObj source $
|
|
|
|
AB.mkAnyBackend $
|
|
|
|
SOITable @b tableName
|
|
|
|
)
|
|
|
|
DRParent
|
2022-11-30 21:12:14 +03:00
|
|
|
pure (eTrigInfo, tabDep Seq.:<| getTrigDefDeps @b source tableName def)
|
2021-09-24 01:56:37 +03:00
|
|
|
|
|
|
|
getTrigDefDeps ::
|
|
|
|
forall b.
|
|
|
|
Backend b =>
|
|
|
|
SourceName ->
|
|
|
|
TableName b ->
|
|
|
|
TriggerOpsDef b ->
|
2022-11-30 21:12:14 +03:00
|
|
|
Seq SchemaDependency
|
2021-09-09 14:54:19 +03:00
|
|
|
getTrigDefDeps source tableName (TriggerOpsDef mIns mUpd mDel _) =
|
2021-09-24 01:56:37 +03:00
|
|
|
mconcat $
|
2022-11-30 21:12:14 +03:00
|
|
|
Seq.fromList
|
|
|
|
<$> catMaybes
|
|
|
|
[ subsOpSpecDeps <$> mIns,
|
|
|
|
subsOpSpecDeps <$> mUpd,
|
|
|
|
subsOpSpecDeps <$> mDel
|
|
|
|
]
|
2021-09-09 14:54:19 +03:00
|
|
|
where
|
|
|
|
subsOpSpecDeps :: SubscribeOpSpec b -> [SchemaDependency]
|
|
|
|
subsOpSpecDeps os =
|
|
|
|
let cols = getColsFromSub $ sosColumns os
|
|
|
|
mkColDependency dependencyReason col =
|
|
|
|
SchemaDependency
|
2021-09-24 01:56:37 +03:00
|
|
|
( SOSourceObj source $
|
|
|
|
AB.mkAnyBackend $
|
|
|
|
SOITableObj @b tableName (TOCol @b col)
|
|
|
|
)
|
2021-09-09 14:54:19 +03:00
|
|
|
dependencyReason
|
|
|
|
colDeps = map (mkColDependency DRColumn) cols
|
|
|
|
payload = maybe [] getColsFromSub (sosPayload os)
|
|
|
|
payloadDeps = map (mkColDependency DRPayload) payload
|
2021-09-24 01:56:37 +03:00
|
|
|
in colDeps <> payloadDeps
|
2021-09-09 14:54:19 +03:00
|
|
|
getColsFromSub sc = case sc of
|
2021-09-24 01:56:37 +03:00
|
|
|
SubCStar -> []
|
2021-09-09 14:54:19 +03:00
|
|
|
SubCArray cols -> cols
|
2022-04-11 14:24:11 +03:00
|
|
|
|
|
|
|
getTriggersMap ::
|
|
|
|
SourceMetadata b ->
|
|
|
|
InsOrdHashMap TriggerName (EventTriggerConf b)
|
|
|
|
getTriggersMap = OMap.unions . map _tmEventTriggers . OMap.elems . _smTables
|
|
|
|
|
2023-02-20 17:19:14 +03:00
|
|
|
getSourceTableAndTriggers ::
|
|
|
|
SourceMetadata b ->
|
|
|
|
[(TableName b, TriggerName)]
|
|
|
|
getSourceTableAndTriggers =
|
|
|
|
(concatMap mkKeyValue) . OMap.toList . _smTables
|
|
|
|
where
|
|
|
|
mkKeyValue (tableName, tableMetadata) = map (tableName,) $ OMap.keys (_tmEventTriggers tableMetadata)
|
|
|
|
|
2022-04-11 14:24:11 +03:00
|
|
|
getTriggerNames ::
|
|
|
|
SourceMetadata b ->
|
|
|
|
Set.HashSet TriggerName
|
|
|
|
getTriggerNames = Set.fromList . OMap.keys . getTriggersMap
|
2022-04-21 10:19:37 +03:00
|
|
|
|
|
|
|
getTableNameFromTrigger ::
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
forall b.
|
|
|
|
Backend b =>
|
2022-04-21 10:19:37 +03:00
|
|
|
SchemaCache ->
|
|
|
|
SourceName ->
|
|
|
|
TriggerName ->
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
Maybe (TableName b)
|
2022-12-08 19:03:20 +03:00
|
|
|
getTableNameFromTrigger schemaCache sourceName triggerName = do
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
tableInfo <- getTabInfoFromSchemaCache @b schemaCache sourceName triggerName
|
|
|
|
pure $ _tciName $ _tiCoreInfo tableInfo
|
2022-09-09 11:26:44 +03:00
|
|
|
|
|
|
|
runCleanupEventTriggerLog ::
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
(MonadEventLogCleanup m, MonadError QErr m, CacheRWM m) =>
|
2022-09-09 11:26:44 +03:00
|
|
|
TriggerLogCleanupConfig ->
|
|
|
|
m EncJSON
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
runCleanupEventTriggerLog conf = do
|
|
|
|
sourceCache <- scSources <$> askSchemaCache
|
|
|
|
runLogCleaner sourceCache conf `onLeftM` throwError
|
2022-09-13 11:33:44 +03:00
|
|
|
|
|
|
|
-- | Updates the cleanup switch in metadata given the source, table and trigger name
|
|
|
|
-- The Bool value represents the status of the cleaner, whether to start or pause it
|
|
|
|
updateCleanupStatusInMetadata ::
|
|
|
|
forall b m.
|
|
|
|
(Backend b, QErrM m, CacheRWM m, MetadataM m) =>
|
|
|
|
AutoTriggerLogCleanupConfig ->
|
|
|
|
EventTriggerCleanupStatus ->
|
|
|
|
SourceName ->
|
|
|
|
TableName b ->
|
|
|
|
TriggerName ->
|
|
|
|
m ()
|
|
|
|
updateCleanupStatusInMetadata cleanupConfig cleanupSwitch sourceName tableName triggerName = do
|
|
|
|
let newCleanupConfig = Just $ cleanupConfig {_atlccPaused = cleanupSwitch}
|
|
|
|
metadataObj =
|
|
|
|
MOSourceObjId sourceName $
|
|
|
|
AB.mkAnyBackend $
|
|
|
|
SMOTableObj @b tableName $
|
|
|
|
MTOTrigger triggerName
|
|
|
|
|
|
|
|
buildSchemaCacheFor metadataObj $
|
|
|
|
MetadataModifier $
|
|
|
|
tableMetadataSetter @b sourceName tableName . tmEventTriggers . ix triggerName %~ updateCleanupConfig newCleanupConfig
|
|
|
|
|
|
|
|
-- | Function to start/stop the cleanup action based on the event triggers supplied in
|
|
|
|
-- TriggerLogCleanupToggleConfig conf
|
|
|
|
toggleEventTriggerCleanupAction ::
|
|
|
|
forall m.
|
|
|
|
(MonadIO m, QErrM m, CacheRWM m, MetadataM m) =>
|
|
|
|
TriggerLogCleanupToggleConfig ->
|
|
|
|
EventTriggerCleanupStatus ->
|
|
|
|
m EncJSON
|
|
|
|
toggleEventTriggerCleanupAction conf cleanupSwitch = do
|
|
|
|
schemaCache <- askSchemaCache
|
|
|
|
case conf of
|
|
|
|
TriggerLogCleanupSources tlcs -> do
|
|
|
|
case tlcs of
|
|
|
|
TriggerAllSource -> do
|
|
|
|
ifor_ (scSources schemaCache) $ \sourceName backendSourceInfo -> do
|
2023-04-18 08:36:02 +03:00
|
|
|
AB.dispatchAnyBackend @BackendEventTrigger backendSourceInfo \(SourceInfo {..} :: SourceInfo b) -> do
|
|
|
|
traverseTableHelper _siTables cleanupSwitch sourceName
|
2022-09-13 11:33:44 +03:00
|
|
|
TriggerSource sourceNameLst -> do
|
|
|
|
forM_ sourceNameLst $ \sourceName -> do
|
|
|
|
backendSourceInfo <-
|
|
|
|
HM.lookup sourceName (scSources schemaCache)
|
|
|
|
`onNothing` throw400 NotExists ("source with name " <> sourceNameToText sourceName <> " does not exists")
|
|
|
|
|
2023-04-18 08:36:02 +03:00
|
|
|
AB.dispatchAnyBackend @BackendEventTrigger backendSourceInfo \(SourceInfo {..} :: SourceInfo b) -> do
|
|
|
|
traverseTableHelper _siTables cleanupSwitch sourceName
|
2022-09-13 11:33:44 +03:00
|
|
|
TriggerQualifier qualifierLst -> do
|
|
|
|
forM_ qualifierLst $ \qualifier -> do
|
|
|
|
let sourceName = _etqSourceName qualifier
|
|
|
|
triggerNames = _etqEventTriggers qualifier
|
|
|
|
|
|
|
|
backendSourceInfo <-
|
|
|
|
HM.lookup sourceName (scSources schemaCache)
|
|
|
|
`onNothing` throw400 NotExists ("source with name " <> sourceNameToText sourceName <> " does not exists")
|
|
|
|
|
|
|
|
AB.dispatchAnyBackend @BackendEventTrigger backendSourceInfo \(SourceInfo {} :: SourceInfo b) -> do
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
for_ triggerNames $ \triggerName -> do
|
2022-09-13 11:33:44 +03:00
|
|
|
eventTriggerInfo <- askEventTriggerInfo @b sourceName triggerName
|
small cleanups of pro's init
### Description
(This PR is better reviewed commit by commit.)
This PR is an aggregation of small incremental changes to Pro's init:
- it deletes some dead code,
- it starts reorganizing the code of that file by sections, similar to OSS' init,
- it extracts and cleans up license key cache init (groups several blocks of code in one separate function)
- makes some changes to a service class to reduce the dependency on `_acAppStateRef`
This PR is a first step: our goal is to move the schema cache build _in_ the app monad, in order to achieve #8344. To do so, we will need to remove `_acAppStateRef` from Pro's `AppContext`. There are two different paths we can take from here, which is why i cut this PR here:
- the first would be to change the different instances we implement on `AppM` to take as an argument the parts of the schema cache they depend on, rather than reading them from `_acAppStateRef`; as of this PR, `MetricsConfig` is the only such field;
- the second would be to apply the same strategy we already used for the TLSAllowList, and use a `IORef` that can be updated after the schema cache is built; this change would have a smaller footprint, but introduces one new `IORef` per such field, which feels something we don't want to generalize
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/8655
GitOrigin-RevId: 809697d460bdb5c83ef7d30a2e835f589bcd80a6
2023-04-06 18:35:43 +03:00
|
|
|
tableName <-
|
|
|
|
getTableNameFromTrigger @b schemaCache sourceName triggerName
|
|
|
|
`onNothing` throw400 NotExists ("event trigger " <> triggerName <<> " does not exist")
|
|
|
|
cleanupConfig <-
|
|
|
|
etiCleanupConfig eventTriggerInfo
|
|
|
|
`onNothing` throw400 NotExists ("cleanup config does not exist for " <> triggerNameToTxt triggerName)
|
|
|
|
updateCleanupStatusInMetadata @b cleanupConfig cleanupSwitch sourceName tableName triggerName
|
2022-09-13 11:33:44 +03:00
|
|
|
pure successMsg
|
|
|
|
where
|
|
|
|
traverseTableHelper ::
|
|
|
|
forall b.
|
|
|
|
(Backend b) =>
|
|
|
|
TableCache b ->
|
|
|
|
EventTriggerCleanupStatus ->
|
|
|
|
SourceName ->
|
|
|
|
m ()
|
|
|
|
traverseTableHelper tableCache switch sourceName = forM_ tableCache $ \tableInfo -> do
|
|
|
|
let tableName = (_tciName . _tiCoreInfo) tableInfo
|
|
|
|
eventTriggerInfoMap = _tiEventTriggerInfoMap tableInfo
|
|
|
|
ifor_ eventTriggerInfoMap $ \triggerName eventTriggerInfo -> do
|
2022-10-04 00:49:32 +03:00
|
|
|
for_ (etiCleanupConfig eventTriggerInfo) $ \cleanupConfig ->
|
2022-09-13 11:33:44 +03:00
|
|
|
updateCleanupStatusInMetadata @b cleanupConfig switch sourceName tableName triggerName
|
|
|
|
|
2022-09-21 08:59:14 +03:00
|
|
|
runEventTriggerResumeCleanup ::
|
2022-09-13 11:33:44 +03:00
|
|
|
forall m.
|
|
|
|
(MonadIO m, QErrM m, CacheRWM m, MetadataM m) =>
|
|
|
|
TriggerLogCleanupToggleConfig ->
|
|
|
|
m EncJSON
|
2022-09-21 08:59:14 +03:00
|
|
|
runEventTriggerResumeCleanup conf = toggleEventTriggerCleanupAction conf ETCSUnpaused
|
2022-09-13 11:33:44 +03:00
|
|
|
|
|
|
|
runEventTriggerPauseCleanup ::
|
|
|
|
(MonadError QErr m, CacheRWM m, MonadIO m, MetadataM m) =>
|
|
|
|
TriggerLogCleanupToggleConfig ->
|
|
|
|
m EncJSON
|
|
|
|
runEventTriggerPauseCleanup conf = toggleEventTriggerCleanupAction conf ETCSPaused
|
|
|
|
|
|
|
|
-- | Collects and returns all the event triggers with cleanup config
|
|
|
|
getAllEventTriggersWithCleanupConfig :: TableInfo b -> [(TriggerName, AutoTriggerLogCleanupConfig)]
|
|
|
|
getAllEventTriggersWithCleanupConfig tInfo = mapMaybe (\(triggerName, triggerInfo) -> (triggerName,) <$> etiCleanupConfig triggerInfo) $ Map.toList $ _tiEventTriggerInfoMap tInfo
|
2022-09-15 14:45:14 +03:00
|
|
|
|
|
|
|
hasCleanupCronScheduleUpdated :: Maybe AutoTriggerLogCleanupConfig -> Maybe AutoTriggerLogCleanupConfig -> Bool
|
|
|
|
hasCleanupCronScheduleUpdated Nothing _ = False
|
|
|
|
hasCleanupCronScheduleUpdated _ Nothing = True
|
|
|
|
hasCleanupCronScheduleUpdated (Just oldConfig) (Just newConfig) =
|
|
|
|
_atlccSchedule oldConfig /= _atlccSchedule newConfig
|
|
|
|
|
|
|
|
getAllETWithCleanupConfigInTableMetadata :: TableMetadata b -> [(TriggerName, AutoTriggerLogCleanupConfig)]
|
|
|
|
getAllETWithCleanupConfigInTableMetadata tMetadata =
|
|
|
|
mapMaybe
|
|
|
|
( \(triggerName, triggerConf) ->
|
|
|
|
(triggerName,)
|
|
|
|
<$> etcCleanupConfig triggerConf
|
|
|
|
)
|
2022-11-02 23:53:23 +03:00
|
|
|
$ OMap.toList
|
|
|
|
$ _tmEventTriggers tMetadata
|
2023-04-25 14:22:27 +03:00
|
|
|
|
|
|
|
runGetEventLogs ::
|
|
|
|
forall b m.
|
|
|
|
(MonadIO m, CacheRM m, MonadError QErr m, BackendEventTrigger b, MetadataM m) =>
|
|
|
|
GetEventLogs b ->
|
|
|
|
m EncJSON
|
|
|
|
runGetEventLogs getEventLogs = do
|
|
|
|
sourceConfig <- askSourceConfig @b sourceName
|
|
|
|
doesTriggerExists <- checkIfTriggerNameExists @b sourceName triggerName
|
|
|
|
if not doesTriggerExists
|
|
|
|
then throw400 NotExists $ "event trigger " <> triggerName <<> " does not exist"
|
|
|
|
else encJFromJValue <$> fetchEventLogs sourceConfig getEventLogs
|
|
|
|
where
|
|
|
|
sourceName = _gelSourceName getEventLogs
|
|
|
|
triggerName = _gelName getEventLogs
|
|
|
|
|
|
|
|
runGetEventInvocationLogs ::
|
|
|
|
forall b m.
|
|
|
|
(MonadIO m, CacheRM m, MonadError QErr m, BackendEventTrigger b, MetadataM m) =>
|
|
|
|
GetEventInvocations b ->
|
|
|
|
m EncJSON
|
|
|
|
runGetEventInvocationLogs getEventInvocations = do
|
|
|
|
sourceConfig <- askSourceConfig @b sourceName
|
|
|
|
doesTriggerExists <- checkIfTriggerNameExists @b sourceName triggerName
|
|
|
|
if not doesTriggerExists
|
|
|
|
then throw400 NotExists $ "event trigger " <> triggerName <<> " does not exist"
|
|
|
|
else encJFromJValue <$> fetchEventInvocationLogs sourceConfig getEventInvocations
|
|
|
|
where
|
|
|
|
sourceName = _geiSourceName getEventInvocations
|
|
|
|
triggerName = _geiName getEventInvocations
|
|
|
|
|
|
|
|
runGetEventById ::
|
|
|
|
forall b m.
|
|
|
|
(MonadIO m, CacheRM m, MonadError QErr m, BackendEventTrigger b, MetadataM m) =>
|
|
|
|
GetEventById b ->
|
|
|
|
m EncJSON
|
|
|
|
runGetEventById getEventById = do
|
|
|
|
sourceConfig <- askSourceConfig @b sourceName
|
|
|
|
encJFromJValue <$> fetchEventById sourceConfig getEventById
|
|
|
|
where
|
|
|
|
sourceName = _gebiSourceName getEventById
|