Skip to content

Overhaul quantification optimizations #716

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

Merged
merged 16 commits into from
Feb 7, 2024
Merged

Overhaul quantification optimizations #716

Changes from 1 commit
Commits
Show all changes
16 commits
Select commit Hold shift + click to select a range
c2b93ab
Speed up quantification fast paths by unswitching the loop
milseman Dec 11, 2023
b48f09e
squash me: cleanup
milseman Dec 11, 2023
efd4a1d
comments
milseman Dec 11, 2023
45b4da3
Refactor off of mutating methods
milseman Dec 12, 2023
eb6e0e0
wip: trying to generalize and make fast enough
milseman Dec 12, 2023
3fe6fe2
wip: on the way to total un-switching while converting to non-mutating
milseman Dec 12, 2023
1687315
wip: getting closer
milseman Dec 12, 2023
15706a4
wip
milseman Dec 13, 2023
01177fa
wip
milseman Dec 13, 2023
e5ee845
all converted to string methods
milseman Dec 14, 2023
2d7a391
wip
milseman Dec 14, 2023
4c3ebf0
fix findo/replaco
milseman Dec 14, 2023
3c5dea0
better names and comments
milseman Dec 15, 2023
50b2f91
wip: typo
milseman Dec 15, 2023
93a220b
Address review comments
Catfish-Man Feb 1, 2024
7ece5c3
Fix assertion
Catfish-Man Feb 6, 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
Prev Previous commit
Next Next commit
wip
  • Loading branch information
@milseman
milseman committed Dec 14, 2023
commit 2d7a3914acd744e64ed65929a6b6fba985b815a4
339 changes: 302 additions & 37 deletions Sources/_StringProcessing/Engine/MEQuantify.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,49 +9,120 @@ extension Processor {
let produceSavePointRange = payload.quantKind == .eager
let isScalarSemantics = payload.isScalarSemantics

let isZeroOrMore = payload.minTrips == 0 && payload.maxExtraTrips == nil
let isOneOrMore = payload.minTrips == 1 && payload.maxExtraTrips == nil

let matchResult: (next: String.Index, savePointRange: Range<Position>?)?
switch payload.type {
case .asciiBitset:
matchResult = input.matchQuantifiedASCIIBitset(
registers[payload.bitset],
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
if isZeroOrMore {
matchResult = input.matchZeroOrMoreASCIIBitset(
registers[payload.bitset],
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
} else if isOneOrMore {
matchResult = input.matchOneOrMoreASCIIBitset(
registers[payload.bitset],
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
} else {
matchResult = input.matchQuantifiedASCIIBitset(
registers[payload.bitset],
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
}

case .asciiChar:
matchResult = input.matchQuantifiedScalar(
Unicode.Scalar(payload.asciiChar),
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
if isZeroOrMore {
matchResult = input.matchZeroOrMoreScalar(
Unicode.Scalar(payload.asciiChar),
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
} else if isOneOrMore {
matchResult = input.matchOneOrMoreScalar(
Unicode.Scalar(payload.asciiChar),
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
} else {
matchResult = input.matchQuantifiedScalar(
Unicode.Scalar(payload.asciiChar),
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics)
}

case .any:
matchResult = input.matchQuantifiedRegexDot(
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics,
anyMatchesNewline: payload.anyMatchesNewline)
if isZeroOrMore {
matchResult = input.matchZeroOrMoreRegexDot(
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
anyMatchesNewline: payload.anyMatchesNewline,
isScalarSemantics: isScalarSemantics)
} else if isOneOrMore {
matchResult = input.matchOneOrMoreRegexDot(
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
anyMatchesNewline: payload.anyMatchesNewline,
isScalarSemantics: isScalarSemantics)
} else {
matchResult = input.matchQuantifiedRegexDot(
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
anyMatchesNewline: payload.anyMatchesNewline,
isScalarSemantics: isScalarSemantics)
}

case .builtin:
matchResult = input.matchQuantifiedBuiltinCC(
payload.builtin,
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isInverted: payload.builtinIsInverted,
isStrictASCII: payload.builtinIsStrict,
isScalarSemantics: isScalarSemantics)
if isZeroOrMore {
matchResult = input.matchZeroOrMoreBuiltinCC(
payload.builtin,
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isInverted: payload.builtinIsInverted,
isStrictASCII: payload.builtinIsStrict,
isScalarSemantics: isScalarSemantics)
} else if isOneOrMore {
matchResult = input.matchOneOrMoreBuiltinCC(
payload.builtin,
at: currentPosition,
limitedBy: end,
produceSavePointRange: produceSavePointRange,
isInverted: payload.builtinIsInverted,
isStrictASCII: payload.builtinIsStrict,
isScalarSemantics: isScalarSemantics)
} else {
matchResult = input.matchQuantifiedBuiltinCC(
payload.builtin,
at: currentPosition,
limitedBy: end,
minMatches: minMatches,
maxMatches: maxMatches,
produceSavePointRange: produceSavePointRange,
isInverted: payload.builtinIsInverted,
isStrictASCII: payload.builtinIsStrict,
isScalarSemantics: isScalarSemantics)
}
}

guard let (next, savePointRange) = matchResult else {
Expand Down Expand Up @@ -121,7 +192,55 @@ extension String {
// position, because newline-sequence in scalar semantic mode still
// matches two scalars

return (currentPosition, rangeStart..<rangeEnd)
return (currentPosition, rangeStart..<rangeEnd)
}

/// NOTE: [Zero|One]OrMore overloads are to specialize the inlined run loop,
/// which has a substantive perf impact (especially for zero-or-more)

fileprivate func matchZeroOrMoreASCIIBitset(
_ asciiBitset: ASCIIBitset,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 0,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchASCIIBitset(
asciiBitset,
at: currentPosition,
limitedBy: end,
isScalarSemantics: isScalarSemantics)
}
}
fileprivate func matchOneOrMoreASCIIBitset(
_ asciiBitset: ASCIIBitset,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 1,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchASCIIBitset(
asciiBitset,
at: currentPosition,
limitedBy: end,
isScalarSemantics: isScalarSemantics)
}
}

fileprivate func matchQuantifiedASCIIBitset(
Expand Down Expand Up @@ -149,6 +268,54 @@ extension String {
}
}

fileprivate func matchZeroOrMoreScalar(
_ scalar: Unicode.Scalar,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 0,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchScalar(
scalar,
at: currentPosition,
limitedBy: end,
boundaryCheck: !isScalarSemantics,
isCaseInsensitive: false)
}
}
fileprivate func matchOneOrMoreScalar(
_ scalar: Unicode.Scalar,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 1,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchScalar(
scalar,
at: currentPosition,
limitedBy: end,
boundaryCheck: !isScalarSemantics,
isCaseInsensitive: false)

}
}

fileprivate func matchQuantifiedScalar(
_ scalar: Unicode.Scalar,
at currentPosition: Index,
Expand Down Expand Up @@ -176,6 +343,59 @@ extension String {
}
}

fileprivate func matchZeroOrMoreBuiltinCC(
_ builtinCC: _CharacterClassModel.Representation,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isInverted: Bool,
isStrictASCII: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 0,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchBuiltinCC(
builtinCC,
at: currentPosition,
limitedBy: end,
isInverted: isInverted,
isStrictASCII: isStrictASCII,
isScalarSemantics: isScalarSemantics)
}
}
fileprivate func matchOneOrMoreBuiltinCC(
_ builtinCC: _CharacterClassModel.Representation,
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
isInverted: Bool,
isStrictASCII: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 1,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchBuiltinCC(
builtinCC,
at: currentPosition,
limitedBy: end,
isInverted: isInverted,
isStrictASCII: isStrictASCII,
isScalarSemantics: isScalarSemantics)
}
}

fileprivate func matchQuantifiedBuiltinCC(
_ builtinCC: _CharacterClassModel.Representation,
at currentPosition: Index,
Expand Down Expand Up @@ -205,14 +425,59 @@ extension String {
}
}

fileprivate func matchZeroOrMoreRegexDot(
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
anyMatchesNewline: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 0,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchRegexDot(
at: currentPosition,
limitedBy: end,
anyMatchesNewline: anyMatchesNewline,
isScalarSemantics: isScalarSemantics)
}
}
fileprivate func matchOneOrMoreRegexDot(
at currentPosition: Index,
limitedBy end: Index,
produceSavePointRange: Bool,
anyMatchesNewline: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
limitedBy: end,
minMatches: 1,
maxMatches: UInt64.max,
produceSavePointRange: produceSavePointRange,
isScalarSemantics: isScalarSemantics
) { currentPosition, end, isScalarSemantics in
matchRegexDot(
at: currentPosition,
limitedBy: end,
anyMatchesNewline: anyMatchesNewline,
isScalarSemantics: isScalarSemantics)
}
}

fileprivate func matchQuantifiedRegexDot(
at currentPosition: Index,
limitedBy end: Index,
minMatches: UInt64,
maxMatches: UInt64,
produceSavePointRange: Bool,
isScalarSemantics: Bool,
anyMatchesNewline: Bool
anyMatchesNewline: Bool,
isScalarSemantics: Bool
) -> (next: Index, savePointRange: Range<Index>?)? {
_runQuantLoop(
at: currentPosition,
Expand Down