Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

perf: perform matching over indices #443

Merged
merged 5 commits into from
Sep 17, 2024
Merged
Show file tree
Hide file tree
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 41 additions & 2 deletions src/N3Store.js
Original file line number Diff line number Diff line change
Expand Up @@ -1026,6 +1026,26 @@ export default class N3Store {
}
}

/**
* Performs matching over the store indexes
*/
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shoud add similar docs to #438 around falsy values

jeswr marked this conversation as resolved.
Show resolved Hide resolved
function indexMatch(index, ids, depth = 0) {
const ind = ids[depth];
if (ind && !(ind in index))
return false;

let target = false;
for (const key in (ind ? { [ind]: index[ind] } : index)) {
const result = depth === 2 ? null : indexMatch(index[key], ids, depth + 1);

if (result !== false) {
target = target || Object.create(null);
target[key] = result;
}
}
return target;
}
jeswr marked this conversation as resolved.
Show resolved Hide resolved

/**
* A class that implements both DatasetCore and Readable.
*/
Expand All @@ -1039,8 +1059,27 @@ class DatasetCoreAndReadableStream extends Readable {
if (!this._filtered) {
const { n3Store, graph, object, predicate, subject } = this;
const newStore = this._filtered = new N3Store({ factory: n3Store._factory, entityIndex: this.options.entityIndex });
for (const quad of n3Store.readQuads(subject, predicate, object, graph))
newStore.addQuad(quad);
const graphs = n3Store._getGraphs(graph);

let subjectId, predicateId, objectId;

// Translate IRIs to internal index keys.
if (subject && !(subjectId = newStore._termToNumericId(subject)) ||
predicate && !(predicateId = newStore._termToNumericId(predicate)) ||
object && !(objectId = newStore._termToNumericId(object)))
return newStore;

for (const graph in graphs) {
const subjects = indexMatch(graphs[graph].subjects, [subjectId, predicateId, objectId]);
if (subjects) {
newStore._graphs[graph] = {
subjects,
predicates: indexMatch(graphs[graph].predicates, [predicateId, objectId, subjectId]),
objects: indexMatch(graphs[graph].objects, [objectId, subjectId, predicateId]),
Comment on lines +1078 to +1083
Copy link
Collaborator Author

@jeswr jeswr Sep 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We could optimise slightly further by using the same technique as #readQuads to determine which index to match over first as this will be more efficient for identifying if there are no matches. But it would be sufficient to leave that work for a follow up PR.

};
}
}
newStore._size = null;
}
return this._filtered;
}
Expand Down
6 changes: 6 additions & 0 deletions test/N3Store-test.js
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@ import {
Store,
termFromId, termToId,
EntityIndex,
DataFactory,
} from '../src';
import {
NamedNode,
Expand All @@ -13,6 +14,8 @@ import namespaces from '../src/IRIs';
import { Readable } from 'readable-stream';
import { arrayifyStream } from 'arrayify-stream';

const { namedNode } = DataFactory;

describe('Store', () => {
describe('The Store export', () => {
it('should be a function', () => {
Expand Down Expand Up @@ -695,6 +698,9 @@ describe('Store', () => {

expect(nextDataset.has(new Quad('s2', 'p1', 'o1'))).toBe(true);
expect(nextDataset.has(new Quad('s2', 'p2', 'o2'))).toBe(true);
expect(nextDataset.match(namedNode('s2'), null, null).has(new Quad('s2', 'p2', 'o2'))).toBe(true);
expect(nextDataset.match(null, namedNode('p2'), null).has(new Quad('s2', 'p2', 'o2'))).toBe(true);
expect(nextDataset.match(null, null, namedNode('o2')).has(new Quad('s2', 'p2', 'o2'))).toBe(true);

nextDataset.delete(new Quad('s2', 'p1', 'o1'));
nextDataset.delete(new Quad('s2', 'p2', 'o2'));
Expand Down