OSGKeyboard/OSGKeyboardShared/Typing/RimeSchemaGenerator.swift

// RimeSchemaGenerator.swift
// OSGKeyboard · Shared
//
// Generates OSG-owned schemas from public Microsoft/Sogou key maps.
// No GPL Rime schema files are copied or distributed.

import Foundation

public enum RimeSchemaGenerator {
    public static func defaultConfiguration() -> String {
        """
        # Generated by OSGKeyboard.
        config_version: "1.0"
        schema_list:
          - schema: \(TypingInputSchema.fullPinyin.rawValue)
          - schema: \(TypingInputSchema.microsoftDoublePinyin.rawValue)
          - schema: \(TypingInputSchema.sogouDoublePinyin.rawValue)
        switcher:
          caption: 输入方案
          hotkeys: []
        menu:
          # Large page so librime prepares enough candidates for the expand panel.
          # The bridge also iterates beyond the current page via candidate_list_*.
          page_size: 100
        ascii_composer:
          good_old_caps_lock: true
          switch_key:
            Shift_L: noop
            Shift_R: noop
            Control_L: noop
            Control_R: noop
        key_binder:
          bindings: []
        recognizer:
          patterns: {}
        """
    }

    public static func schema(
        for inputSchema: TypingInputSchema,
        fuzzyPairs: Set<PinyinFuzzyPair>
    ) -> String {
        let alphabet = inputSchema == .fullPinyin
            ? "zyxwvutsrqponmlkjihgfedcba"
            : "zyxwvutsrqponmlkjihgfedcba;"
        // Dialect single-letter syllables must be erased before fuzzy/abbrev
        // so `n/l` cannot revive 嗯 as `l`, and `wom` cannot exact-match 我呒.
        let algebra = dialectEraseRules + fuzzyRules(fuzzyPairs) + algebraRules(for: inputSchema)
        let algebraYAML = algebra.map { "    - '\($0)'" }.joined(separator: "\n")

        return """
        # Generated by OSGKeyboard. Do not edit; change settings in the host app.
        schema:
          schema_id: \(inputSchema.rawValue)
          name: \(inputSchema.displayName)
          version: "1.0"
          author:
            - OSGKeyboard contributors
          description: |
            Commercially permissive OSG schema backed by osg_pinyin.

        switches:
          - name: ascii_mode
            reset: 0
            states: [中, 英]

        engine:
          processors:
            - ascii_composer
            - recognizer
            - key_binder
            - speller
            - punctuator
            - selector
            - navigator
            - express_editor
          segmentors:
            - ascii_segmentor
            - matcher
            - abc_segmentor
            - punct_segmentor
            - fallback_segmentor
          translators:
            - punct_translator
            - script_translator

        speller:
          alphabet: "\(alphabet)"
          initials: "\(alphabet.replacingOccurrences(of: ";", with: ""))"
          delimiter: " '"
          algebra:
        \(algebraYAML)

        translator:
          dictionary: osg_pinyin
          prism: \(inputSchema.rawValue)
          enable_sentence: true
          enable_completion: true
          enable_user_dict: true
          initial_quality: 1.2

        punctuator:
          half_shape:
            ",": "，"
            ".": "。"
            "?": "？"
            "!": "！"

        key_binder:
          import_preset: default

        recognizer:
          import_preset: default
        """
    }

    /// Drop Wu/dialect exact spellings (`呒 m`, `嗯 n/ng`, `噷 hm`) before
    /// first-letter abbrev, matching rime-pinyin-simp. Keep the dictionary
    /// rows; 嗯 remains reachable as `en`, 呒 as `mu`.
    public static let dialectEraseRules: [String] = [
        "erase/^hm$/",
        "erase/^m$/",
        "erase/^n$/",
        "erase/^ng$/"
    ]

    /// Rules run against full-pinyin dictionary codes before double-pinyin
    /// transforms, so fuzzy pairs work consistently in all three schemas.
    public static func fuzzyRules(_ enabled: Set<PinyinFuzzyPair>) -> [String] {
        var rules: [String] = []
        for pair in PinyinFuzzyPair.allCases where enabled.contains(pair) {
            switch pair {
            case .zhZ:
                rules += ["derive/^zh/z/", "derive/^z([^h])/zh$1/"]
            case .chC:
                rules += ["derive/^ch/c/", "derive/^c([^h])/ch$1/"]
            case .shS:
                rules += ["derive/^sh/s/", "derive/^s([^h])/sh$1/"]
            case .nL:
                rules += ["derive/^n/l/", "derive/^l/n/"]
            case .fH:
                rules += ["derive/^f/h/", "derive/^h/f/"]
            case .anAng:
                rules += ["derive/ang$/an/", "derive/an$/ang/"]
            case .enEng:
                rules += ["derive/eng$/en/", "derive/en$/eng/"]
            case .inIng:
                rules += ["derive/ing$/in/", "derive/in$/ing/"]
            }
        }
        return rules
    }

    private static func algebraRules(for schema: TypingInputSchema) -> [String] {
        switch schema {
        case .fullPinyin:
            return [
                "derive/^([jqxy])u$/$1v/",
                "abbrev/^([a-z]).+$/$1/",
                // Two-letter initials; do not add this to double pinyin.
                "abbrev/^([zcs]h).+$/$1/"
            ]

        case .microsoftDoublePinyin, .sogouDoublePinyin:
            // Microsoft and Sogou's commonly shipped layouts are key-compatible.
            // Uppercase markers prevent transforms from matching each other.
            return [
                "erase/^xx$/",
                "derive/^([jqxy])u$/$1v/",
                "derive/^([aoe].*)$/o$1/",
                "xform/^([ae])(.*)$/$1$1$2/",
                "xform/iu$/Q/",
                "xform/[iu]a$/W/",
                "xform/er$|[uv]an$/R/",
                "xform/[uv]e$/T/",
                "xform/v$|uai$/Y/",
                "xform/^sh/U/",
                "xform/^ch/I/",
                "xform/^zh/V/",
                "xform/uo$/O/",
                "xform/[uv]n$/P/",
                "xform/(.)i?ong$/$1S/",
                "xform/[iu]ang$/D/",
                "xform/(.)en$/$1F/",
                "xform/(.)eng$/$1G/",
                "xform/(.)ang$/$1H/",
                "xform/ian$/M/",
                "xform/(.)an$/$1J/",
                "xform/iao$/C/",
                "xform/(.)ao$/$1K/",
                "xform/(.)ai$/$1L/",
                "xform/(.)ei$/$1Z/",
                "xform/ie$/X/",
                "xform/ui$/V/",
                "derive/T$/V/",
                "xform/(.)ou$/$1B/",
                "xform/in$/N/",
                "xform/ing$/;/",
                "xlit/QWRTYUIOPSDFGHMJCKLZXVBN/qwrtyuiopsdfghmjcklzxvbn/"
            ]
        }
    }
}