After login macOS starts the input method on first use, and keys typed while it loads its data were delayed or lost. Startup now: - parses lexicon.txt as raw bytes (no per-line String splitting or Double parsing) and builds the tone-ignoring index lazily on a background queue; - loads the precomputed English dictionary and bigram table (shared with the Windows build) instead of retraining the bigram model on every launch, and uses binary search instead of building a 230k-entry Set. Measured (release, warm cache): lexicon 0.65 -> 0.10 s, English 0.30 -> 0.04 s. The Windows (Python) core also builds its toneless index lazily: 0.56 -> 0.37 s. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
153 lines
6.7 KiB
Swift
153 lines
6.7 KiB
Swift
import Foundation
|
|
|
|
/// Scores how much a typed string looks like English.
|
|
///
|
|
/// Three sources of evidence:
|
|
/// - a list of common words (strong evidence, includes modern words like "ok", "email"),
|
|
/// - a large dictionary (e.g. /usr/share/dict/words, weaker evidence),
|
|
/// - a letter-bigram model trained on those words, for words not in either list.
|
|
public final class EnglishModel {
|
|
private var common = Set<String>()
|
|
private var sortedCommon: [String] = []
|
|
/// Dictionary words (lowercase, > 2 letters), sorted: membership and prefix checks are
|
|
/// binary searches, so no 230k-entry Set has to be built at startup.
|
|
private var sortedDict: [String] = []
|
|
private var bigram: [String: Double] = [:] // "ab" -> log2 P(b|a); "^a", "a$" for boundaries
|
|
private var unseenBigram = -12.0
|
|
|
|
/// - Parameter bigram: precomputed table (english-bigram.json); trained from the words if nil.
|
|
public init(commonWords: [String], dictionaryWords: [String] = [],
|
|
bigram: (table: [String: Double], unseen: Double)? = nil) {
|
|
common = Set(commonWords.map { $0.lowercased() }.filter { !$0.isEmpty })
|
|
sortedCommon = common.sorted()
|
|
var dict = dictionaryWords.map { $0.lowercased() }.filter { $0.count > 2 }
|
|
if !zip(dict, dict.dropFirst()).allSatisfy({ $0 < $1 }) {
|
|
dict = Array(Set(dict)).sorted()
|
|
}
|
|
sortedDict = dict
|
|
if let bigram {
|
|
self.bigram = bigram.table
|
|
unseenBigram = bigram.unseen
|
|
} else {
|
|
trainBigram()
|
|
}
|
|
}
|
|
|
|
/// Loads the bundled data files. Falls back to the system word list and to training
|
|
/// the bigram table when the precomputed files are missing.
|
|
public convenience init(commonWordsURL: URL?, dictionaryURL: URL? = nil, bigramURL: URL? = nil) {
|
|
func lines(_ url: URL?) -> [String] {
|
|
guard let url, let text = try? String(contentsOf: url, encoding: .utf8) else { return [] }
|
|
return text.split(whereSeparator: \.isNewline)
|
|
.map { $0.trimmingCharacters(in: .whitespaces) }
|
|
.filter { !$0.isEmpty && !$0.hasPrefix("#") }
|
|
}
|
|
var bigram: (table: [String: Double], unseen: Double)?
|
|
if let url = bigramURL, let data = try? Data(contentsOf: url),
|
|
let obj = try? JSONSerialization.jsonObject(with: data) as? [String: Any],
|
|
let table = obj["table"] as? [String: Double], let unseen = obj["unseen"] as? Double {
|
|
bigram = (table, unseen)
|
|
}
|
|
let dictURL = dictionaryURL.flatMap { FileManager.default.fileExists(atPath: $0.path) ? $0 : nil }
|
|
?? URL(fileURLWithPath: "/usr/share/dict/words")
|
|
self.init(commonWords: lines(commonWordsURL), dictionaryWords: lines(dictURL), bigram: bigram)
|
|
}
|
|
|
|
public func isCommon(_ w: String) -> Bool { common.contains(w.lowercased()) }
|
|
public func isWord(_ w: String) -> Bool { let l = w.lowercased(); return common.contains(l) || inDictionary(l) }
|
|
|
|
private func inDictionary(_ w: String) -> Bool {
|
|
let i = lowerBound(w, in: sortedDict)
|
|
return i < sortedDict.count && sortedDict[i] == w
|
|
}
|
|
|
|
/// Some word strictly longer than `p` starts with it (common words or dictionary).
|
|
private func isPrefixOfWord(_ p: String) -> Bool {
|
|
hasPrefix(p, in: sortedCommon) || hasPrefix(p, in: sortedDict)
|
|
}
|
|
|
|
/// - Parameter complete: true when the user ended the word (space / punctuation),
|
|
/// false while still typing (prefixes of words get partial credit).
|
|
public func score(_ word: String, complete: Bool) -> Double {
|
|
let w = word.lowercased()
|
|
let n = Double(w.count)
|
|
guard !w.isEmpty else { return 0 }
|
|
if common.contains(w) { return 3 + n }
|
|
if inDictionary(w) { return 1 + 0.6 * n }
|
|
if !complete {
|
|
if hasPrefix(w, in: sortedCommon) { return 0.4 * n }
|
|
if isPrefixOfWord(w) { return 0.2 * n }
|
|
}
|
|
// Unknown word: judge by how English-like its letter sequence is.
|
|
let plaus = plausibility(w)
|
|
return min(0.5 * n, max(-2 * n, n * (plaus + 4.2)))
|
|
}
|
|
|
|
/// Feature dict for the classifier (see Features.swift); `score` is their dot
|
|
/// product with the default weights.
|
|
public func wordFeatures(_ word: String, complete: Bool) -> FeatureVector {
|
|
let w = word.lowercased()
|
|
let n = Double(w.count)
|
|
var f: FeatureVector = [:]
|
|
if n <= 2 { f["lit.short"] = 1 }
|
|
if common.contains(w) {
|
|
f["lit.common"] = 1
|
|
f["lit.common_len"] = n
|
|
} else if inDictionary(w) {
|
|
f["lit.dict"] = 1
|
|
f["lit.dict_len"] = n
|
|
} else if !complete && hasPrefix(w, in: sortedCommon) {
|
|
f["lit.prefix_common_len"] = n
|
|
} else if !complete && isPrefixOfWord(w) {
|
|
f["lit.prefix_dict_len"] = n
|
|
} else {
|
|
let plaus = plausibility(w)
|
|
f["lit.unknown_score"] = min(0.5 * n, max(-2 * n, n * (plaus + 4.2)))
|
|
}
|
|
return f
|
|
}
|
|
|
|
/// Mean log2 probability per letter transition (≈ -3.5 for English, lower for gibberish).
|
|
public func plausibility(_ w: String) -> Double {
|
|
let chars = ["^"] + w.map(String.init) + ["$"]
|
|
var total = 0.0
|
|
for i in 1..<chars.count { total += bigram[chars[i - 1] + chars[i]] ?? unseenBigram }
|
|
return total / Double(chars.count - 1)
|
|
}
|
|
|
|
private func lowerBound(_ p: String, in words: [String]) -> Int {
|
|
var lo = 0, hi = words.count
|
|
while lo < hi {
|
|
let mid = (lo + hi) / 2
|
|
if words[mid] < p { lo = mid + 1 } else { hi = mid }
|
|
}
|
|
return lo
|
|
}
|
|
|
|
private func hasPrefix(_ p: String, in words: [String]) -> Bool {
|
|
var lo = lowerBound(p, in: words)
|
|
// words[lo] is the first word >= p; skip an exact match, then test the next one.
|
|
if lo < words.count && words[lo] == p { lo += 1 }
|
|
return lo < words.count && words[lo].hasPrefix(p)
|
|
}
|
|
|
|
private func trainBigram() {
|
|
var pair: [String: Double] = [:]
|
|
var first: [String: Double] = [:]
|
|
for w in Set(sortedCommon + sortedDict).sorted() where w.allSatisfy({ $0.isLetter && $0.isASCII }) {
|
|
let weight = common.contains(w) ? 5.0 : 1.0
|
|
let chars = ["^"] + w.map(String.init) + ["$"]
|
|
for i in 1..<chars.count {
|
|
pair[chars[i - 1] + chars[i], default: 0] += weight
|
|
first[chars[i - 1], default: 0] += weight
|
|
}
|
|
}
|
|
// Add-one smoothing over 27 possible successors.
|
|
for (k, c) in pair {
|
|
let a = String(k.prefix(1))
|
|
bigram[k] = log2((c + 1) / ((first[a] ?? 0) + 27))
|
|
}
|
|
unseenBigram = log2(1.0 / ((first.values.max() ?? 1000) + 27))
|
|
}
|
|
}
|