[Tools] 辞書データ解析用のサブコマンドをancoに追加 (#85)

* base value should be updated here

* feature: implement cli tool for reading louds data

* Update document

Docs/cli.md

@@ -4,13 +4,13 @@
 
 `anco`を利用するには、最初にinstallが必要です。
 
-```
+```bash
-sh install_cli.sh
+sudo sh install_cli.sh
 ```
 
 例えば以下のように利用できます。
 
-```
+```bash
 your@pc Desktop % anco にほんごにゅうりょく --disable_prediction -n 10
 日本語入力
 にほんご入力
@@ -23,3 +23,40 @@ your@pc Desktop % anco にほんごにゅうりょく --disable_prediction -n 10
 にほんご
 2本後
 ```
+
+## 変換API
+
+`anco run`コマンドを利用して変換を行うことが出来ます。
+
+## 辞書リーダ
+
+`anco dict`コマンドを利用して辞書データを解析することが出来ます。
+
+```bash
+your@pc Desktop % anco dict read ア -d ./Sources/KanaKanjiConverterModuleWithDefaultDictionary/azooKey_dictionary_storage/Dictionary/                       
+=== Summary for target ア ===
+- directory: ./Sources/KanaKanjiConverterModuleWithDefaultDictionary/azooKey_dictionary_storage/Dictionary/
+- target: ア
+- memory?: false
+- count of entry: 24189
+- time for execute: 0.0378040075302124
+```
+
+`--ruby`および`--word`オプションを利用して、正規表現でフィルターをかけることが出来ます。
+
+```bash
+your@pc Desktop % anco dict read ア -d ./Sources/KanaKanjiConverterModuleWithDefaultDictionary/azooKey_dictionary_storage/Dictionary/ --word ".*全"
+=== Summary for target ア ===
+- directory: ./Sources/KanaKanjiConverterModuleWithDefaultDictionary/azooKey_dictionary_storage/Dictionary/
+- target: ア
+- memory?: false
+- count of entry: 24189
+- time for execute: 0.07062792778015137
+=== Found Entries ===
+- count of found entry: 3
+Ruby: アキラ Word: 全 Value: -11.7107 CID: (1291, 1291) MID: 424
+Ruby: アンゼン Word: 安全 Value: -7.241 CID: (1287, 1287) MID: 169
+Ruby: アンシンアンゼン Word: 安心安全 Value: -11.7638 CID: (1283, 1287) MID: 17
+```
+
+`--sort`オプションを使うとエントリーの並び替えが可能です。

Sources/CliTool/Anco.swift

@@ -5,7 +5,7 @@ import ArgumentParser
 public struct Anco: ParsableCommand {
     public static var configuration = CommandConfiguration(
         abstract: "Anco is A(zooKey) Kana-Ka(n)ji (co)nverter",
-        subcommands: [Subcommands.Run.self],
+        subcommands: [Subcommands.Run.self, Subcommands.Dict.self],
         defaultSubcommand: Subcommands.Run.self
     )
 

Sources/CliTool/DefaultStringInterpolation+CommandLineUtils.swift

@@ -0,0 +1,14 @@
+//
+//  DefaultStringInterpolation+CommandLineUtils.swift
+//
+//
+//  Created by miwa on 2024/04/29.
+//
+
+import Foundation
+
+extension DefaultStringInterpolation {
+    mutating func appendInterpolation(bold value: String){
+        self.appendInterpolation("\u{1B}[1m" + value + "\u{1B}[m")
+    }
+}

Sources/CliTool/Subcommands/DictCommands/DictCommand.swift

@@ -0,0 +1,13 @@
+import Foundation
+import KanaKanjiConverterModuleWithDefaultDictionary
+import ArgumentParser
+
+extension Subcommands {
+    struct Dict: ParsableCommand {
+        static var configuration = CommandConfiguration(
+            commandName: "dict",
+            abstract: "Show dict information", 
+            subcommands: [Self.Read.self]
+        )
+    }
+}

Sources/CliTool/Subcommands/DictCommands/ReadCommand.swift

@@ -0,0 +1,123 @@
+import Foundation
+import KanaKanjiConverterModule
+import ArgumentParser
+
+extension Subcommands.Dict {
+    struct Read: ParsableCommand {
+        enum SortOrder: String, Codable, ExpressibleByArgument {
+            case value
+            case ruby
+            case word
+
+            init?(argument: String) {
+                self.init(rawValue: argument)
+            }
+        }
+
+        @Argument(help: "辞書データのfilename")
+        var target: String = ""
+
+        @Option(name: [.customLong("dictionary_dir"), .customShort("d")], help: "The directory for dictionary data.")
+        var dictionaryDirectory: String = "./"
+
+        @Option(name: [.customLong("ruby")], help: "Regex for entry ruby filter")
+        var rubyFilter: String = ""
+
+        @Option(name: [.customLong("word")], help: "Regex for entry word filter")
+        var wordFilter: String = ""
+
+        @Option(name: [.customLong("sort")], help: "Sort order")
+        var sortOrder: SortOrder = .ruby
+
+        static var configuration = CommandConfiguration(
+            commandName: "read",
+            abstract: "Read dictionary data and extract informations"
+        )
+
+        @MainActor mutating func run() throws {
+            guard #available(macOS 13, *) else {
+                return
+            }
+            let start = Date()
+            let isMemory = self.target == "memory"
+            guard let louds = LOUDS.load(self.target, option: self.requestOptions()) else {
+                print(
+                    """
+                    \(bold: "=== Summary for target \(self.target) ===")
+                    - directory: \(self.dictionaryDirectory)
+                    - target: \(self.target)
+                    - memory?: \(isMemory)
+                    - result: LOUDS data was not found
+                    - time for execute: \(Date().timeIntervalSince(start))
+                    """
+                )
+                return
+            }
+            // ありったけ取り出す
+            let nodeIndices = louds.prefixNodeIndices(chars: [], maxDepth: .max)
+            let store = DicdataStore(convertRequestOptions: self.requestOptions())
+            let result = store.getDicdataFromLoudstxt3(identifier: self.target, indices: nodeIndices)
+            var filteredResult = result
+            var hasFilter = false
+            if !rubyFilter.isEmpty {
+                let filter = try Regex(rubyFilter)
+                hasFilter = true
+                filteredResult = filteredResult.filter {
+                    $0.ruby.wholeMatch(of: filter) != nil
+                }
+            }
+            if !wordFilter.isEmpty {
+                let filter = try Regex(wordFilter)
+                hasFilter = true
+                filteredResult = filteredResult.filter {
+                    $0.word.wholeMatch(of: filter) != nil
+                }
+            }
+
+            print(
+                """
+                \(bold: "=== Summary for target \(self.target) ===")
+                - directory: \(self.dictionaryDirectory)
+                - target: \(self.target)
+                - memory?: \(isMemory)
+                - count of entry: \(result.count)
+                - time for execute: \(Date().timeIntervalSince(start))
+                """
+            )
+
+            if hasFilter {
+                let sortFunction: (DicdataElement, DicdataElement) -> Bool = switch self.sortOrder {
+                case .ruby: { $0.ruby < $1.ruby || $0.ruby.count < $1.ruby.count}
+                case .value: { $0.value() < $1.value() }
+                case .word: { $0.word < $1.word }
+                }
+
+                print("\(bold: "=== Found Entries ===")")
+                print("- count of found entry: \(filteredResult.count)")
+                for entry in filteredResult.sorted(by: sortFunction) {
+                    print("\(bold: "Ruby:") \(entry.ruby) \(bold: "Word:") \(entry.word) \(bold: "Value:") \(entry.value()) \(bold: "CID:") \((entry.lcid, entry.rcid)) \(bold: "MID:") \(entry.mid)")
+                }
+            }
+        }
+
+        func requestOptions() -> ConvertRequestOptions {
+            .init(
+                N_best: 0,
+                requireJapanesePrediction: false,
+                requireEnglishPrediction: false,
+                keyboardLanguage: .ja_JP,
+                typographyLetterCandidate: false,
+                unicodeCandidate: true,
+                englishCandidateInRoman2KanaInput: true,
+                fullWidthRomanCandidate: false,
+                halfWidthKanaCandidate: false,
+                learningType: .nothing,
+                maxMemoryCount: 0,
+                dictionaryResourceURL: URL(fileURLWithPath: self.dictionaryDirectory),
+                memoryDirectoryURL: URL(fileURLWithPath: self.dictionaryDirectory),
+                sharedContainerURL: URL(fileURLWithPath: self.dictionaryDirectory),
+                metadata: .init(appVersionString: "anco")
+            )
+        }
+    }
+}

Sources/KanaKanjiConverterModule/DicdataStore/DicdataStore.swift

@@ -189,7 +189,7 @@ public final class DicdataStore {
         return louds.prefixNodeIndices(chars: charIDs, maxDepth: depth)
     }
 
-    func getDicdataFromLoudstxt3(identifier: String, indices: some Sequence<Int>) -> [DicdataElement] {
+    package func getDicdataFromLoudstxt3(identifier: String, indices: some Sequence<Int>) -> [DicdataElement] {
         debug("getDicdataFromLoudstxt3", identifier, indices)
         // split = 2048
         let dict = [Int: [Int]].init(grouping: indices, by: {$0 >> 11})

Sources/KanaKanjiConverterModule/LOUDS/LOUDS.swift

@@ -9,7 +9,7 @@
 import Foundation
 
 /// LOUDS
-struct LOUDS: Sendable {
+package struct LOUDS: Sendable {
     private typealias Unit = UInt64
     private static let unit = 64
     private static let uExp = 6
@@ -182,7 +182,7 @@ struct LOUDS: Sendable {
     /// - Parameter chars: CharIDに変換した文字列
     /// - Parameter maxDepth: 先に進む深さの最大値
     /// - Returns: 対応するloudstxt3ファイル内のインデックスのリスト
-    @inlinable func prefixNodeIndices(chars: [UInt8], maxDepth: Int) -> [Int] {
+    @inlinable package func prefixNodeIndices(chars: [UInt8], maxDepth: Int) -> [Int] {
         guard let nodeIndex = self.searchNodeIndex(chars: chars) else {
             return []
         }

Sources/KanaKanjiConverterModule/LOUDS/extension LOUDS.swift

@@ -54,7 +54,7 @@ extension LOUDS {
     /// LOUDSをファイルから読み込む関数
     /// - Parameter identifier: ファイル名
     /// - Returns: 存在すればLOUDSデータを返し、存在しなければ`nil`を返す。
-    static func load(_ identifier: String, option: ConvertRequestOptions) -> LOUDS? {
+    package static func load(_ identifier: String, option: ConvertRequestOptions) -> LOUDS? {
         let (charsURL, loudsURL) = getLOUDSURL(identifier, option: option)
         let nodeIndex2ID: [UInt8]
         do {