Experiment/dependencies graph #364
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Removed dependency query from DatasetRepository. Integrated dependencies graph into GraphQL queries for upstream/downstream links. Integrated dependencies graph into dataset deletion. Reacting on `DatasetCreated` events.
zaychenko-sergei
changed the base branch from
master
to
experiment/dataset-update-flow
sergiimk
reviewed
Dec 14, 2023
zaychenko-sergei
merged commit 56cfbd6
into
experiment/dataset-update-flow
4 of 5 checks passed
zaychenko-sergei
added a commit
that referenced
this pull request
Dec 14, 2023
* Simplest startup job to initialize dependencies graph. * Removed dependency query from DatasetRepository. * Integrated dependencies graph into GraphQL queries for upstream/downstream links. * Integrated dependencies graph into dataset deletion. * Reacting on `DatasetCreated` events. * Implemented reaction of dependencies graph on changes in dataset inputs * Implemented lazy vs eager dependencies initialization
zaychenko-sergei
added a commit
that referenced
this pull request
Dec 18, 2023
DatasetUpdateFlow => UpdateSchedule. Added Update aggregate, representing a single instance of update process for a given dataset. Formatter fix Review: renamings Review: externalized time source for event-sourcing aggregates Review: compacted task events in update flow Review: added update cancellation (before tasks scheduled) Review: accepted suggestions to name delay reasons as start conditions, and secondary triggers as simply adding triggers Merge corrections In-memory implementation of repositories Drafted update scheduler service. ES: support optional aggregate loads. Sketched `UpdateService` without tasks scheduling yet Scheduler steps: - schedule update task on manual trigger - read all active auto-schedules at the beginning of the run process - react on schedule change events: update table of active schedules Separate set in each in-memory event repository: quick return of query objects Drafted Time Wheel concept Connected time wheel and update service: initial scheduling and run loop Drafted in-memory dependency graph service (based on petgraph library). Scheduling downstream datasets when dataset update completes, respecting throttling period logic Enqueue next auto-polling root dataset update when current update succeeds Prototyped EventBus + added 1st demo link for schedule modification event Minor event structure fixes Simplified UpdateSchedule events Connected task finish and dataset removal events. Large DI changes in existing tests to support EventBus dependency Concurrent execution of event handlers in dispatcher Converted event bus handlers to traits. Registering handlers in the catalog. Merge corrections Review: renamed DatasetDeleted event Review: avoid excessive events cloning Shifted down 'get_queries' to update schedule's event store only Async event handler combiner now collects all handlers results, before reducing the error for reporting Resolved basic code review notes Added `get_last_update` by dataset operation Formatter fixed dill 0.8 - replaced `builder_for` on `Component::builder()` Integrated `dill::interface` feature and removed many explicit binds Review: renamed task event classes, enum for dataset events Review: reworked relevance of update schedule, using statuses instead (active, paused, stopped) Review: allow update schedules to be re-added after dataset reincarnation with the same ID Review: a few TODOs on performance improvements Review: reimplemented TimeWheel using binary heap Review: removed `pause` and `resume` methods in `UpdateSchedule` aggregate, use `set_schedule` only Renamed update schedules => update configurations Separated schedules and start conditions in update configurations Generalized dataset flow configurations System vs Dataset flow configurations Not very smart, but a model of System and Dataset flows. Scheduling service largely not implemented yet. Generic-based flow events, state Refactored flow configurations aggregate to generic events/state similarly Code reuse approach for flow/flow-config aggregates based on trait extensions Attempts to generalize flow configuration services (at least traits). Folder reorganization in interface and in-mem crate. Implemented generic in-memory event stores and integrated them into all current aggregates Implemented SystemFlow in-memory repository Implemented in-memory Flow service for all kinds of flows Decomposing Flow service: extracted ActiveConfigsState Decomposing Flow service: extracted PendingFlowsState Compacted DatasetFlow & SystemFlow into Flow Similarly compacted FlowConfiguration aggregate Simplifications in FlowService Review: 'flow-system' and 'flow-system-inmem' are final names Review: 'flow-system' and 'flow-system-inmem' are final names Review: improved enum all-value iteration methods in flow types Review: specific => of_type Review: removed duplicate OwnedDatasetFlowKey Review: killed redundand feature flags Review: tracing without formatting Moved `DependencyGraphService` to core domain Review: removed reundand field Experiment/dependencies graph (#364) * Simplest startup job to initialize dependencies graph. * Removed dependency query from DatasetRepository. * Integrated dependencies graph into GraphQL queries for upstream/downstream links. * Integrated dependencies graph into dataset deletion. * Reacting on `DatasetCreated` events. * Implemented reaction of dependencies graph on changes in dataset inputs * Implemented lazy vs eager dependencies initialization Merge corrections Test fix Review: minor renamings
zaychenko-sergei
added a commit
that referenced
this pull request
Dec 20, 2023
DatasetUpdateFlow => UpdateSchedule. Added Update aggregate, representing a single instance of update process for a given dataset. Formatter fix Review: renamings Review: externalized time source for event-sourcing aggregates Review: compacted task events in update flow Review: added update cancellation (before tasks scheduled) Review: accepted suggestions to name delay reasons as start conditions, and secondary triggers as simply adding triggers Merge corrections In-memory implementation of repositories Drafted update scheduler service. ES: support optional aggregate loads. Sketched `UpdateService` without tasks scheduling yet Scheduler steps: - schedule update task on manual trigger - read all active auto-schedules at the beginning of the run process - react on schedule change events: update table of active schedules Separate set in each in-memory event repository: quick return of query objects Drafted Time Wheel concept Connected time wheel and update service: initial scheduling and run loop Drafted in-memory dependency graph service (based on petgraph library). Scheduling downstream datasets when dataset update completes, respecting throttling period logic Enqueue next auto-polling root dataset update when current update succeeds Prototyped EventBus + added 1st demo link for schedule modification event Minor event structure fixes Simplified UpdateSchedule events Connected task finish and dataset removal events. Large DI changes in existing tests to support EventBus dependency Concurrent execution of event handlers in dispatcher Converted event bus handlers to traits. Registering handlers in the catalog. Merge corrections Review: renamed DatasetDeleted event Review: avoid excessive events cloning Shifted down 'get_queries' to update schedule's event store only Async event handler combiner now collects all handlers results, before reducing the error for reporting Resolved basic code review notes Added `get_last_update` by dataset operation Formatter fixed dill 0.8 - replaced `builder_for` on `Component::builder()` Integrated `dill::interface` feature and removed many explicit binds Review: renamed task event classes, enum for dataset events Review: reworked relevance of update schedule, using statuses instead (active, paused, stopped) Review: allow update schedules to be re-added after dataset reincarnation with the same ID Review: a few TODOs on performance improvements Review: reimplemented TimeWheel using binary heap Review: removed `pause` and `resume` methods in `UpdateSchedule` aggregate, use `set_schedule` only Renamed update schedules => update configurations Separated schedules and start conditions in update configurations Generalized dataset flow configurations System vs Dataset flow configurations Not very smart, but a model of System and Dataset flows. Scheduling service largely not implemented yet. Generic-based flow events, state Refactored flow configurations aggregate to generic events/state similarly Code reuse approach for flow/flow-config aggregates based on trait extensions Attempts to generalize flow configuration services (at least traits). Folder reorganization in interface and in-mem crate. Implemented generic in-memory event stores and integrated them into all current aggregates Implemented SystemFlow in-memory repository Implemented in-memory Flow service for all kinds of flows Decomposing Flow service: extracted ActiveConfigsState Decomposing Flow service: extracted PendingFlowsState Compacted DatasetFlow & SystemFlow into Flow Similarly compacted FlowConfiguration aggregate Simplifications in FlowService Review: 'flow-system' and 'flow-system-inmem' are final names Review: 'flow-system' and 'flow-system-inmem' are final names Review: improved enum all-value iteration methods in flow types Review: specific => of_type Review: removed duplicate OwnedDatasetFlowKey Review: killed redundand feature flags Review: tracing without formatting Moved `DependencyGraphService` to core domain Review: removed reundand field Experiment/dependencies graph (#364) * Simplest startup job to initialize dependencies graph. * Removed dependency query from DatasetRepository. * Integrated dependencies graph into GraphQL queries for upstream/downstream links. * Integrated dependencies graph into dataset deletion. * Reacting on `DatasetCreated` events. * Implemented reaction of dependencies graph on changes in dataset inputs * Implemented lazy vs eager dependencies initialization Merge corrections Test fix Review: minor renamings
zaychenko-sergei
added a commit
that referenced
this pull request
Dec 22, 2023
DatasetUpdateFlow => UpdateSchedule. Added Update aggregate, representing a single instance of update process for a given dataset. Formatter fix Review: renamings Review: externalized time source for event-sourcing aggregates Review: compacted task events in update flow Review: added update cancellation (before tasks scheduled) Review: accepted suggestions to name delay reasons as start conditions, and secondary triggers as simply adding triggers Merge corrections In-memory implementation of repositories Drafted update scheduler service. ES: support optional aggregate loads. Sketched `UpdateService` without tasks scheduling yet Scheduler steps: - schedule update task on manual trigger - read all active auto-schedules at the beginning of the run process - react on schedule change events: update table of active schedules Separate set in each in-memory event repository: quick return of query objects Drafted Time Wheel concept Connected time wheel and update service: initial scheduling and run loop Drafted in-memory dependency graph service (based on petgraph library). Scheduling downstream datasets when dataset update completes, respecting throttling period logic Enqueue next auto-polling root dataset update when current update succeeds Prototyped EventBus + added 1st demo link for schedule modification event Minor event structure fixes Simplified UpdateSchedule events Connected task finish and dataset removal events. Large DI changes in existing tests to support EventBus dependency Concurrent execution of event handlers in dispatcher Converted event bus handlers to traits. Registering handlers in the catalog. Merge corrections Review: renamed DatasetDeleted event Review: avoid excessive events cloning Shifted down 'get_queries' to update schedule's event store only Async event handler combiner now collects all handlers results, before reducing the error for reporting Resolved basic code review notes Added `get_last_update` by dataset operation Formatter fixed dill 0.8 - replaced `builder_for` on `Component::builder()` Integrated `dill::interface` feature and removed many explicit binds Review: renamed task event classes, enum for dataset events Review: reworked relevance of update schedule, using statuses instead (active, paused, stopped) Review: allow update schedules to be re-added after dataset reincarnation with the same ID Review: a few TODOs on performance improvements Review: reimplemented TimeWheel using binary heap Review: removed `pause` and `resume` methods in `UpdateSchedule` aggregate, use `set_schedule` only Renamed update schedules => update configurations Separated schedules and start conditions in update configurations Generalized dataset flow configurations System vs Dataset flow configurations Not very smart, but a model of System and Dataset flows. Scheduling service largely not implemented yet. Generic-based flow events, state Refactored flow configurations aggregate to generic events/state similarly Code reuse approach for flow/flow-config aggregates based on trait extensions Attempts to generalize flow configuration services (at least traits). Folder reorganization in interface and in-mem crate. Implemented generic in-memory event stores and integrated them into all current aggregates Implemented SystemFlow in-memory repository Implemented in-memory Flow service for all kinds of flows Decomposing Flow service: extracted ActiveConfigsState Decomposing Flow service: extracted PendingFlowsState Compacted DatasetFlow & SystemFlow into Flow Similarly compacted FlowConfiguration aggregate Simplifications in FlowService Review: 'flow-system' and 'flow-system-inmem' are final names Review: 'flow-system' and 'flow-system-inmem' are final names Review: improved enum all-value iteration methods in flow types Review: specific => of_type Review: removed duplicate OwnedDatasetFlowKey Review: killed redundand feature flags Review: tracing without formatting Moved `DependencyGraphService` to core domain Review: removed reundand field Experiment/dependencies graph (#364) * Simplest startup job to initialize dependencies graph. * Removed dependency query from DatasetRepository. * Integrated dependencies graph into GraphQL queries for upstream/downstream links. * Integrated dependencies graph into dataset deletion. * Reacting on `DatasetCreated` events. * Implemented reaction of dependencies graph on changes in dataset inputs * Implemented lazy vs eager dependencies initialization Merge corrections Test fix Review: minor renamings
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.