GraphOps refactor #1544
Merged
GraphOps refactor #1544
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
|
|
||
| fn base(&self) -> &(dyn BoxableGraphView<'static> + Send + Sync + 'static) { | ||
| fn base(&self) -> &(dyn BoxableGraphView + Send + Sync + 'static) { |
There was a problem hiding this comment.
meaning?
fn base(&self) -> &Self::BaseThere was a problem hiding this comment.
The lifetime for BoxableGraphView is no longer needed because it is not dealing with the iterators any more.
There was a problem hiding this comment.
what I meant is you can just use Self::Base instead of copying the type that is right above
| fn edge_filter(&self) -> Option<&EdgeFilter> { | ||
| None | ||
| /// | ||
| fn edges_filtered(&self) -> bool { |
There was a problem hiding this comment.
empty comment and we don't seem to inline these but we do NodeFilterOps
| e.time_t() | ||
| .or_else(|| self.edge_additions(e, layer_ids).first_t()) | ||
| .or_else(|| self.edge_additions(e, layer_ids.clone()).first_t()) |
There was a problem hiding this comment.
should edge_additions also take &LayerIds?
| match e.time().map(|ti| ti.t()) { | ||
| Some(t) => Some(min( | ||
| self.edge_additions(e, layer_ids.clone()) | ||
| .range(t.saturating_add(1)..i64::MAX) | ||
| .first_t() | ||
| .unwrap_or(i64::MAX), | ||
| self.edge_deletions(e, layer_ids) | ||
| self.edge_deletions(e, layer_ids.clone()) |
There was a problem hiding this comment.
edge_deletions didn't get the &LayerIds treatement
fabianmurariu
pushed a commit
that referenced
this pull request
May 21, 2024
* make sure nodes are added in correct order when a node dataframe is provided * wip paralllel * major refactor * fix SCC test (this is a set so order does not matter) * delete the graph_ops internal trait as it is no longer needed * eliminate no-longer needed lifetime * some more cleaning up * actually pass in the locked graph to the map operations and add par_iter for nodes * bring back some windowing optimisation * change the node count iterator * fix the performance regressions
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.
What changes were proposed in this pull request?
This changes the way the internal graph operations are implemented. The internal view api now acts as a filter that is passed down to the underlying storage for execution. This simplifies the underlying execution and allows us to expose parallel iterators (wip)
Why are the changes needed?
Prepare for more parallel apis and general simplification
Does this PR introduce any user-facing change? If yes is this documented?
the public api is unaffected
How was this patch tested?
all the tests still pass after refactor
Are there any further changes required?
A lot more cleanup and refactor needed to expose parallel execution to python