-
Notifications
You must be signed in to change notification settings - Fork 150
Symbol search: improve performance for anchored queries #682
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
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -239,6 +239,7 @@ func (t *symbolRegexpMatchTree) matches(cp *contentProvider, cost int, known map | |
type symbolSubstrMatchTree struct { | ||
*substrMatchTree | ||
|
||
exact bool | ||
patternSize uint32 | ||
fileEndRunes []uint32 | ||
fileEndSymbol []uint32 | ||
|
@@ -292,12 +293,19 @@ func (t *symbolSubstrMatchTree) prepare(doc uint32) { | |
continue | ||
} | ||
|
||
if end <= sections[secIdx].End { | ||
t.current[0].symbol = true | ||
t.current[0].symbolIdx = uint32(secIdx) | ||
trimmed = append(trimmed, t.current[0]) | ||
if end > sections[secIdx].End { | ||
t.current = t.current[1:] | ||
continue | ||
} | ||
|
||
if t.exact && !(start == sections[secIdx].Start && end == sections[secIdx].End) { | ||
t.current = t.current[1:] | ||
continue | ||
} | ||
|
||
t.current[0].symbol = true | ||
t.current[0].symbolIdx = uint32(secIdx) | ||
trimmed = append(trimmed, t.current[0]) | ||
t.current = t.current[1:] | ||
} | ||
t.current = trimmed | ||
|
@@ -983,14 +991,20 @@ func (d *indexData) newMatchTree(q query.Q, opt matchTreeOpt) (matchTree, error) | |
optCopy := opt | ||
optCopy.DisableWordMatchOptimization = true | ||
|
||
subMT, err := d.newMatchTree(s.Expr, optCopy) | ||
expr, wasAnchored := s.Expr, false | ||
if regexpExpr, ok := expr.(*query.Regexp); ok { | ||
expr, wasAnchored = stripAnchors(regexpExpr) | ||
} | ||
|
||
subMT, err := d.newMatchTree(expr, optCopy) | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I looked at putting this logic into |
||
if err != nil { | ||
return nil, err | ||
} | ||
|
||
if substr, ok := subMT.(*substrMatchTree); ok { | ||
return &symbolSubstrMatchTree{ | ||
substrMatchTree: substr, | ||
exact: wasAnchored, | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. one last bug here, I think the regexp we pass to symbolRegexpMatchTree (if we get to that path) will now exclude anchors. |
||
patternSize: uint32(utf8.RuneCountInString(substr.query.Pattern)), | ||
fileEndRunes: d.fileEndRunes, | ||
fileEndSymbol: d.fileEndSymbol, | ||
|
@@ -1256,3 +1270,37 @@ func pruneMatchTree(mt matchTree) (matchTree, error) { | |
} | ||
return mt, err | ||
} | ||
|
||
func stripAnchors(in *query.Regexp) (out *query.Regexp, stripped bool) { | ||
stripRegexpAnchors := func(in *syntax.Regexp) (out *syntax.Regexp, stripped bool) { | ||
if in.Op != syntax.OpConcat { | ||
return out, false | ||
} | ||
|
||
if len(in.Sub) < 3 { | ||
return out, false | ||
} | ||
|
||
firstOp, lastOp := in.Sub[0].Op, in.Sub[len(in.Sub)-1].Op | ||
|
||
if firstOp != syntax.OpBeginLine && firstOp != syntax.OpBeginText { | ||
return out, false | ||
} | ||
if lastOp != syntax.OpEndLine && lastOp != syntax.OpEndText { | ||
return out, false | ||
} | ||
|
||
inCopy := *in | ||
inCopy.Sub = in.Sub[1 : len(in.Sub)-1] // remove the first and last ops, which are the anchors | ||
return &inCopy, true | ||
} | ||
|
||
newRegexp, stripped := stripRegexpAnchors(in.Regexp) | ||
if !stripped { | ||
return in, false | ||
} | ||
|
||
inCopy := *in | ||
inCopy.Regexp = newRegexp | ||
return &inCopy, true | ||
} |
Uh oh!
There was an error while loading. Please reload this page.