Skip to content

WIP: Add fuzzy keyword search to title for bleve #31510

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

Closed
wants to merge 1 commit into from
Closed

WIP: Add fuzzy keyword search to title for bleve #31510

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 8 additions & 0 deletions modules/indexer/internal/bleve/query.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,14 @@ func MatchPhraseQuery(matchPhrase, field, analyzer string, fuzziness int) *query
return q
}

// FuzzyQuery generates a fuzzy query for the given phrase, field, and fuzziness
func FuzzyQuery(matchPhrase, field string, fuzziness int) *query.FuzzyQuery {
q := bleve.NewFuzzyQuery(matchPhrase)
q.FieldVal = field
q.Fuzziness = fuzziness
return q
}

// BoolFieldQuery generates a bool field query for the given value and field
func BoolFieldQuery(value bool, field string) *query.BoolFieldQuery {
q := bleve.NewBoolFieldQuery(value)
Expand Down
10 changes: 6 additions & 4 deletions modules/indexer/internal/bleve/util.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,12 +50,14 @@ func openIndexer(path string, latestVersion int) (bleve.Index, int, error) {

func GuessFuzzinessByKeyword(s string) int {
// according to https://github.com/blevesearch/bleve/issues/1563, the supported max fuzziness is 2
// magic number 4 was chosen to determine the levenshtein distance per each character of a keyword
// BUT, when using CJK (eg: `갃갃갃` `啊啊啊`), it mismatches a lot.
// BUT, when using CJK (eg: `갃갃갃` `啊啊啊`), it mismatches a lot
// which we need to live with, as we need to support not just ASCII
// in case of code points >= 128 we will increase the fuzziness to 2
// the standard is 1
for _, r := range s {
if r >= 128 {
return 0
return 2
}
}
return min(2, len(s)/4)
return 1
}
2 changes: 1 addition & 1 deletion modules/indexer/issues/bleve/bleve.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -162,7 +162,7 @@ func (b *Indexer) Search(ctx context.Context, options *internal.SearchOptions) (
}

queries = append(queries, bleve.NewDisjunctionQuery([]query.Query{
inner_bleve.MatchPhraseQuery(options.Keyword, "title", issueIndexerAnalyzer, fuzziness),
inner_bleve.FuzzyQuery(options.Keyword, "title", fuzziness),
inner_bleve.MatchPhraseQuery(options.Keyword, "content", issueIndexerAnalyzer, fuzziness),
inner_bleve.MatchPhraseQuery(options.Keyword, "comments", issueIndexerAnalyzer, fuzziness),
}...))
Expand Down
16 changes: 8 additions & 8 deletions modules/indexer/issues/indexer_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -209,13 +209,13 @@ func searchIssueIsPull(t *testing.T) {
SearchOptions{
IsPull: optional.Some(false),
},
[]int64{17, 16, 15, 14, 13, 6, 5, 18, 10, 7, 4, 1},
[]int64{25, 24, 23, 17, 16, 15, 14, 13, 6, 5, 18, 10, 7, 4, 1},
},
{
SearchOptions{
IsPull: optional.Some(true),
},
[]int64{22, 21, 12, 11, 20, 19, 9, 8, 3, 2},
[]int64{22, 21, 28, 27, 26, 12, 11, 20, 19, 9, 8, 3, 2},
},
}
for _, test := range tests {
Expand All @@ -236,7 +236,7 @@ func searchIssueIsClosed(t *testing.T) {
SearchOptions{
IsClosed: optional.Some(false),
},
[]int64{22, 21, 17, 16, 15, 14, 13, 12, 11, 20, 6, 19, 18, 10, 7, 9, 8, 3, 2, 1},
[]int64{25, 24, 23, 22, 21, 28, 27, 26, 17, 16, 15, 14, 13, 12, 11, 20, 6, 19, 18, 10, 7, 9, 8, 3, 2, 1},
},
{
SearchOptions{
Expand Down Expand Up @@ -302,7 +302,7 @@ func searchIssueByLabelID(t *testing.T) {
SearchOptions{
ExcludedLabelIDs: []int64{1},
},
[]int64{22, 21, 17, 16, 15, 14, 13, 12, 11, 20, 6, 5, 19, 18, 10, 7, 4, 9, 8, 3},
[]int64{25, 24, 23, 22, 21, 28, 27, 26, 17, 16, 15, 14, 13, 12, 11, 20, 6, 5, 19, 18, 10, 7, 4, 9, 8, 3},
},
}
for _, test := range tests {
Expand All @@ -323,7 +323,7 @@ func searchIssueByTime(t *testing.T) {
SearchOptions{
UpdatedAfterUnix: optional.Some(int64(0)),
},
[]int64{22, 21, 17, 16, 15, 14, 13, 12, 11, 20, 6, 5, 19, 18, 10, 7, 4, 9, 8, 3, 2, 1},
[]int64{25, 24, 23, 22, 21, 28, 27, 26, 17, 16, 15, 14, 13, 12, 11, 20, 6, 5, 19, 18, 10, 7, 4, 9, 8, 3, 2, 1},
},
}
for _, test := range tests {
Expand All @@ -344,7 +344,7 @@ func searchIssueWithOrder(t *testing.T) {
SearchOptions{
SortBy: internal.SortByCreatedAsc,
},
[]int64{1, 2, 3, 8, 9, 4, 7, 10, 18, 19, 5, 6, 20, 11, 12, 13, 14, 15, 16, 17, 21, 22},
[]int64{1, 2, 3, 8, 9, 4, 7, 10, 18, 19, 5, 6, 20, 11, 12, 13, 14, 15, 16, 17, 26, 27, 28, 21, 22, 23, 24, 25},
},
}
for _, test := range tests {
Expand Down Expand Up @@ -401,8 +401,8 @@ func searchIssueWithPaginator(t *testing.T) {
PageSize: 5,
},
},
[]int64{22, 21, 17, 16, 15},
22,
[]int64{25, 24, 23, 22, 21},
28,
},
}
for _, test := range tests {
Expand Down
1 change: 1 addition & 0 deletions routers/web/repo/issue.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2677,6 +2677,7 @@ func SearchIssues(ctx *context.Context) {
MilestoneIDs: includedMilestones,
ProjectID: projectID,
SortBy: issue_indexer.SortByCreatedDesc,
IsFuzzyKeyword: true,
}

if since != 0 {
Expand Down
Loading