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.

Sign up for GitHub

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

Jump to bottom

Autocomplete: Improve jaccard similiarty retriever #2662

Merged
merged 12 commits into from
Jan 11, 2024
Merged

Autocomplete: Improve jaccard similiarty retriever #2662

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
925ed70
Autocomplete: Improve jaccard similiarty retriever and allow multiple…
philipp-spiess Jan 10, 2024
a8fc260
Optimiz around empty lines and fix off-by-one-errror
philipp-spiess Jan 10, 2024
eba0d0f
Fix tests intermediately
philipp-spiess Jan 10, 2024
fa19b1a
Use helpers to create document and docContext
philipp-spiess Jan 10, 2024
0d69c77
Implement jaccard snippet merging
philipp-spiess Jan 10, 2024
928dc94
Fix linter issues
philipp-spiess Jan 10, 2024
3cfab48
Move new implementation to its own folder
philipp-spiess Jan 11, 2024
2cddd91
Bring back old implementation for reference
philipp-spiess Jan 11, 2024
7098062
Add flags for new impl
philipp-spiess Jan 11, 2024
26fdec3
Final tweaks
philipp-spiess Jan 11, 2024
bdb8807
It's the similarity, yo!
philipp-spiess Jan 11, 2024
b5a293f
add change log
philipp-spiess Jan 11, 2024
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
161 changes: 128 additions & 33 deletions vscode/src/completions/context/retrievers/jaccard-similarity/bestJaccardMatch.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
import dedent from 'dedent'
import { describe, expect, it } from 'vitest'

import { bestJaccardMatch, getWords } from './bestJaccardMatch'
import { bestJaccardMatches, getWordOccurrences } from './bestJaccardMatch'

const targetSnippet = `
import { bestJaccardMatch, getWords } from './context'
Expand All @@ -25,30 +26,18 @@ describe('getWords', () => {
})
`

const matchSnippet = `
describe('bestJaccardMatch', () => {
it('should return the best match', () => {
const matchText = [
'foo',
'bar',
'baz',
'qux',
'quux',
].join('\n')
})
})
`
const MAX_MATCHES = 50

describe('getWords', () => {
it('works with regular text', () => {
expect(getWords('foo bar baz')).toEqual(
expect(getWordOccurrences('foo bar baz')).toEqual(
new Map<string, number>([
['foo', 1],
['bar', 1],
['baz', 1],
])
)
expect(getWords('running rocks slipped over')).toEqual(
expect(getWordOccurrences('running rocks slipped over')).toEqual(
new Map<string, number>([
['run', 1],
['rock', 1],
Expand All @@ -58,7 +47,7 @@ describe('getWords', () => {
})

it('works with code snippets', () => {
expect(getWords(targetSnippet)).toEqual(
expect(getWordOccurrences(targetSnippet)).toEqual(
new Map<string, number>([
['import', 1],
['bestjaccardmatch', 1],
Expand Down Expand Up @@ -87,7 +76,7 @@ describe('getWords', () => {
})

describe('bestJaccardMatch', () => {
it('should return the best match in 5 line windows', () => {
it('should return the best match', () => {
const matchText = [
'foo',
'bar',
Expand All @@ -104,36 +93,142 @@ describe('bestJaccardMatch', () => {
'xyzzy',
'thud',
].join('\n')
expect(bestJaccardMatch('foo\nbar\nbaz', matchText, 3)).toEqual({
expect(bestJaccardMatches('foo\nbar\nbaz', matchText, 3, MAX_MATCHES)[0]).toEqual({
score: 1,
content: 'foo\nbar\nbaz',
endLine: 3,
startLine: 0,
})
expect(bestJaccardMatch('bar\nquux', matchText, 4)).toEqual({
expect(bestJaccardMatches('bar\nquux', matchText, 4, MAX_MATCHES)[0]).toEqual({
score: 0.5,
content: 'bar\nbaz\nqux\nquux',
endLine: 5,
startLine: 1,
})
expect(
bestJaccardMatch(
bestJaccardMatches(
['grault', 'notexist', 'garply', 'notexist', 'waldo', 'notexist', 'notexist'].join('\n'),
matchText,
6
)
6,
MAX_MATCHES
)[0]
).toEqual({
score: 0.3,
content: ['corge', 'grault', 'garply', 'waldo', 'fred', 'plugh'].join('\n'),
startLine: 4,
endLine: 10,
content: ['quux', 'quuz', 'corge', 'grault', 'garply', 'waldo'].join('\n'),
})
})

it('returns more than one match', () => {
const matchText = dedent`
foo
bar
baz
qux
foo
quuz
corge
grault
garply
waldo
fred
plugh
xyzzy
thud`

const matches = bestJaccardMatches('foo\nbar\nbaz', matchText, 3, MAX_MATCHES)

// Since we slide over the target text line-by-line, we expect matchText.lines - 2 windows
// to be returned
expect(matches).toHaveLength(matchText.split('\n').length - 2)
expect(matches.map(match => match.content.split('\n'))).toEqual([
['foo', 'bar', 'baz'],
['bar', 'baz', 'qux'],
['baz', 'qux', 'foo'],
['qux', 'foo', 'quuz'],
['foo', 'quuz', 'corge'],
['quuz', 'corge', 'grault'],
['corge', 'grault', 'garply'],
['grault', 'garply', 'waldo'],
['garply', 'waldo', 'fred'],
['waldo', 'fred', 'plugh'],
['fred', 'plugh', 'xyzzy'],
['plugh', 'xyzzy', 'thud'],
])
})

it('works with code snippets', () => {
expect(bestJaccardMatch(targetSnippet, matchSnippet, 5)).toMatchInlineSnapshot(`
{
"content": "describe('bestJaccardMatch', () => {
it('should return the best match', () => {
const matchText = [
'foo',
'bar',",
"score": 0.08695652173913043,
}
expect(
bestJaccardMatches(
targetSnippet,
dedent`
describe('bestJaccardMatch', () => {
it('should return the best match', () => {
const matchText = [
'foo',
'bar',
'baz',
'qux',
'quux',
].join('\n')
})
})
`,
5,
MAX_MATCHES
)[0]
).toMatchInlineSnapshot(`
{
"content": "describe('bestJaccardMatch', () => {
it('should return the best match', () => {
const matchText = [
'foo',
'bar',",
"endLine": 5,
"score": 0.08695652173913043,
"startLine": 0,
}
`)
})

it('skips over windows with empty start lines', () => {
const matches = bestJaccardMatches(
'foo',
dedent`
// foo
// unrelated 1
// unrelated 2


// foo
// unrelated 3
// unrelated 4
`,
3,
MAX_MATCHES
)

expect(matches[0].content).toBe('// foo\n// unrelated 1\n// unrelated 2')
expect(matches[1].content).toBe('// foo\n// unrelated 3\n// unrelated 4')
})

it("does not skips over windows with empty start lines if we're at the en", () => {
Copy link
Member

Choose a reason for hiding this comment

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

Suggested change
it("does not skips over windows with empty start lines if we're at the en", () => {
it("does not skip over windows with empty start lines if we're at the end", () => {

const matches = bestJaccardMatches(
targetSnippet,
dedent`
// foo
// unrelated
// unrelated


// foo
`,
3,
MAX_MATCHES
)

expect(matches[0].content).toBe('\n\n// foo')
expect(matches[1].content).toBe('// foo\n// unrelated\n// unrelated')
})
})
Loading
Loading