@lokascript/semantic 1.0.0

package/dist/languages/ja.js

@@ -0,0 +1,1430 @@
+// src/registry.ts
+var tokenizers = /* @__PURE__ */ new Map();
+var profiles = /* @__PURE__ */ new Map();
+var patternCache = /* @__PURE__ */ new Map();
+function registerLanguage(code, tokenizer, profile) {
+  tokenizers.set(code, tokenizer);
+  profiles.set(code, profile);
+  patternCache.delete(code);
+}
+
+// src/tokenizers/base.ts
+var TokenStreamImpl = class {
+  constructor(tokens, language) {
+    this.pos = 0;
+    this.tokens = tokens;
+    this.language = language;
+  }
+  peek(offset = 0) {
+    const index = this.pos + offset;
+    if (index < 0 || index >= this.tokens.length) {
+      return null;
+    }
+    return this.tokens[index];
+  }
+  advance() {
+    if (this.isAtEnd()) {
+      throw new Error("Unexpected end of token stream");
+    }
+    return this.tokens[this.pos++];
+  }
+  isAtEnd() {
+    return this.pos >= this.tokens.length;
+  }
+  mark() {
+    return { position: this.pos };
+  }
+  reset(mark) {
+    this.pos = mark.position;
+  }
+  position() {
+    return this.pos;
+  }
+  /**
+   * Get remaining tokens as an array.
+   */
+  remaining() {
+    return this.tokens.slice(this.pos);
+  }
+  /**
+   * Consume tokens while predicate is true.
+   */
+  takeWhile(predicate) {
+    const result = [];
+    while (!this.isAtEnd() && predicate(this.peek())) {
+      result.push(this.advance());
+    }
+    return result;
+  }
+  /**
+   * Skip tokens while predicate is true.
+   */
+  skipWhile(predicate) {
+    while (!this.isAtEnd() && predicate(this.peek())) {
+      this.advance();
+    }
+  }
+};
+function createPosition(start, end) {
+  return { start, end };
+}
+function createToken(value, kind, position, normalizedOrOptions) {
+  if (typeof normalizedOrOptions === "string") {
+    return { value, kind, position, normalized: normalizedOrOptions };
+  }
+  if (normalizedOrOptions) {
+    const { normalized: normalized2, stem, stemConfidence } = normalizedOrOptions;
+    const token = { value, kind, position };
+    if (normalized2 !== void 0) {
+      token.normalized = normalized2;
+    }
+    if (stem !== void 0) {
+      token.stem = stem;
+      if (stemConfidence !== void 0) {
+        token.stemConfidence = stemConfidence;
+      }
+    }
+    return token;
+  }
+  return { value, kind, position };
+}
+function isWhitespace(char) {
+  return /\s/.test(char);
+}
+function isSelectorStart(char) {
+  return char === "#" || char === "." || char === "[" || char === "@" || char === "*" || char === "<";
+}
+function isQuote(char) {
+  return char === '"' || char === "'" || char === "`" || char === "\u300C" || char === "\u300D";
+}
+function isDigit(char) {
+  return /\d/.test(char);
+}
+function isAsciiLetter(char) {
+  return /[a-zA-Z]/.test(char);
+}
+function isAsciiIdentifierChar(char) {
+  return /[a-zA-Z0-9_-]/.test(char);
+}
+function createUnicodeRangeClassifier(ranges) {
+  return (char) => {
+    const code = char.charCodeAt(0);
+    return ranges.some(([start, end]) => code >= start && code <= end);
+  };
+}
+function combineClassifiers(...classifiers) {
+  return (char) => classifiers.some((fn) => fn(char));
+}
+function extractCssSelector(input, startPos) {
+  if (startPos >= input.length) return null;
+  const char = input[startPos];
+  if (!isSelectorStart(char)) return null;
+  let pos = startPos;
+  let selector = "";
+  if (char === "#" || char === ".") {
+    selector += input[pos++];
+    while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+      selector += input[pos++];
+    }
+    if (selector.length <= 1) return null;
+    if (pos < input.length && input[pos] === "." && char === "#") {
+      const methodStart = pos + 1;
+      let methodEnd = methodStart;
+      while (methodEnd < input.length && isAsciiIdentifierChar(input[methodEnd])) {
+        methodEnd++;
+      }
+      if (methodEnd < input.length && input[methodEnd] === "(") {
+        return selector;
+      }
+    }
+  } else if (char === "[") {
+    let depth = 1;
+    let inQuote = false;
+    let quoteChar = null;
+    let escaped = false;
+    selector += input[pos++];
+    while (pos < input.length && depth > 0) {
+      const c = input[pos];
+      selector += c;
+      if (escaped) {
+        escaped = false;
+      } else if (c === "\\") {
+        escaped = true;
+      } else if (inQuote) {
+        if (c === quoteChar) {
+          inQuote = false;
+          quoteChar = null;
+        }
+      } else {
+        if (c === '"' || c === "'" || c === "`") {
+          inQuote = true;
+          quoteChar = c;
+        } else if (c === "[") {
+          depth++;
+        } else if (c === "]") {
+          depth--;
+        }
+      }
+      pos++;
+    }
+    if (depth !== 0) return null;
+  } else if (char === "@") {
+    selector += input[pos++];
+    while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+      selector += input[pos++];
+    }
+    if (selector.length <= 1) return null;
+  } else if (char === "*") {
+    selector += input[pos++];
+    while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+      selector += input[pos++];
+    }
+    if (selector.length <= 1) return null;
+  } else if (char === "<") {
+    selector += input[pos++];
+    if (pos >= input.length || !isAsciiLetter(input[pos])) return null;
+    while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+      selector += input[pos++];
+    }
+    while (pos < input.length) {
+      const modChar = input[pos];
+      if (modChar === ".") {
+        selector += input[pos++];
+        if (pos >= input.length || !isAsciiIdentifierChar(input[pos])) {
+          return null;
+        }
+        while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+          selector += input[pos++];
+        }
+      } else if (modChar === "#") {
+        selector += input[pos++];
+        if (pos >= input.length || !isAsciiIdentifierChar(input[pos])) {
+          return null;
+        }
+        while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+          selector += input[pos++];
+        }
+      } else if (modChar === "[") {
+        let depth = 1;
+        let inQuote = false;
+        let quoteChar = null;
+        let escaped = false;
+        selector += input[pos++];
+        while (pos < input.length && depth > 0) {
+          const c = input[pos];
+          selector += c;
+          if (escaped) {
+            escaped = false;
+          } else if (c === "\\") {
+            escaped = true;
+          } else if (inQuote) {
+            if (c === quoteChar) {
+              inQuote = false;
+              quoteChar = null;
+            }
+          } else {
+            if (c === '"' || c === "'" || c === "`") {
+              inQuote = true;
+              quoteChar = c;
+            } else if (c === "[") {
+              depth++;
+            } else if (c === "]") {
+              depth--;
+            }
+          }
+          pos++;
+        }
+        if (depth !== 0) return null;
+      } else {
+        break;
+      }
+    }
+    while (pos < input.length && isWhitespace(input[pos])) {
+      selector += input[pos++];
+    }
+    if (pos < input.length && input[pos] === "/") {
+      selector += input[pos++];
+      while (pos < input.length && isWhitespace(input[pos])) {
+        selector += input[pos++];
+      }
+    }
+    if (pos >= input.length || input[pos] !== ">") return null;
+    selector += input[pos++];
+  }
+  return selector || null;
+}
+function isPossessiveMarker(input, pos) {
+  if (pos >= input.length || input[pos] !== "'") return false;
+  if (pos + 1 >= input.length) return false;
+  const nextChar = input[pos + 1].toLowerCase();
+  if (nextChar !== "s") return false;
+  if (pos + 2 >= input.length) return true;
+  const afterS = input[pos + 2];
+  return isWhitespace(afterS) || afterS === "*" || !isAsciiIdentifierChar(afterS);
+}
+function extractStringLiteral(input, startPos) {
+  if (startPos >= input.length) return null;
+  const openQuote = input[startPos];
+  if (!isQuote(openQuote)) return null;
+  if (openQuote === "'" && isPossessiveMarker(input, startPos)) {
+    return null;
+  }
+  const closeQuoteMap = {
+    '"': '"',
+    "'": "'",
+    "`": "`",
+    "\u300C": "\u300D"
+  };
+  const closeQuote = closeQuoteMap[openQuote];
+  if (!closeQuote) return null;
+  let pos = startPos + 1;
+  let literal = openQuote;
+  let escaped = false;
+  while (pos < input.length) {
+    const char = input[pos];
+    literal += char;
+    if (escaped) {
+      escaped = false;
+    } else if (char === "\\") {
+      escaped = true;
+    } else if (char === closeQuote) {
+      return literal;
+    }
+    pos++;
+  }
+  return literal;
+}
+function isUrlStart(input, pos) {
+  if (pos >= input.length) return false;
+  const char = input[pos];
+  const next = input[pos + 1] || "";
+  const third = input[pos + 2] || "";
+  if (char === "/" && next !== "/" && /[a-zA-Z0-9._-]/.test(next)) {
+    return true;
+  }
+  if (char === "/" && next === "/" && /[a-zA-Z]/.test(third)) {
+    return true;
+  }
+  if (char === "." && (next === "/" || next === "." && third === "/")) {
+    return true;
+  }
+  const slice = input.slice(pos, pos + 8).toLowerCase();
+  if (slice.startsWith("http://") || slice.startsWith("https://")) {
+    return true;
+  }
+  return false;
+}
+function extractUrl(input, startPos) {
+  if (!isUrlStart(input, startPos)) return null;
+  let pos = startPos;
+  let url = "";
+  const urlChars = /[a-zA-Z0-9/:._\-?&=%@+~!$'()*,;[\]]/;
+  while (pos < input.length) {
+    const char = input[pos];
+    if (char === "#") {
+      if (url.length > 0 && /[a-zA-Z0-9/.]$/.test(url)) {
+        url += char;
+        pos++;
+        while (pos < input.length && /[a-zA-Z0-9_-]/.test(input[pos])) {
+          url += input[pos++];
+        }
+      }
+      break;
+    }
+    if (urlChars.test(char)) {
+      url += char;
+      pos++;
+    } else {
+      break;
+    }
+  }
+  if (url.length < 2) return null;
+  return url;
+}
+function extractNumber(input, startPos) {
+  if (startPos >= input.length) return null;
+  const char = input[startPos];
+  if (!isDigit(char) && char !== "-" && char !== "+") return null;
+  let pos = startPos;
+  let number = "";
+  if (input[pos] === "-" || input[pos] === "+") {
+    number += input[pos++];
+  }
+  if (pos >= input.length || !isDigit(input[pos])) {
+    return null;
+  }
+  while (pos < input.length && isDigit(input[pos])) {
+    number += input[pos++];
+  }
+  if (pos < input.length && input[pos] === ".") {
+    number += input[pos++];
+    while (pos < input.length && isDigit(input[pos])) {
+      number += input[pos++];
+    }
+  }
+  if (pos < input.length) {
+    const suffix = input.slice(pos, pos + 2);
+    if (suffix === "ms") {
+      number += "ms";
+    } else if (input[pos] === "s" || input[pos] === "m" || input[pos] === "h") {
+      number += input[pos];
+    }
+  }
+  return number;
+}
+var _BaseTokenizer = class _BaseTokenizer {
+  constructor() {
+    /** Keywords derived from profile, sorted longest-first for greedy matching */
+    this.profileKeywords = [];
+    /** Map for O(1) keyword lookups by lowercase native word */
+    this.profileKeywordMap = /* @__PURE__ */ new Map();
+  }
+  /**
+   * Initialize keyword mappings from a language profile.
+   * Builds a list of native→english mappings from:
+   * - profile.keywords (primary + alternatives)
+   * - profile.references (me, it, you, etc.)
+   * - profile.roleMarkers (into, from, with, etc.)
+   *
+   * Results are sorted longest-first for greedy matching (important for non-space languages).
+   * Extras take precedence over profile entries when there are duplicates.
+   *
+   * @param profile - Language profile containing keyword translations
+   * @param extras - Additional keyword entries to include (literals, positional, events)
+   */
+  initializeKeywordsFromProfile(profile, extras = []) {
+    const keywordMap = /* @__PURE__ */ new Map();
+    if (profile.keywords) {
+      for (const [normalized2, translation] of Object.entries(profile.keywords)) {
+        keywordMap.set(translation.primary, {
+          native: translation.primary,
+          normalized: translation.normalized || normalized2
+        });
+        if (translation.alternatives) {
+          for (const alt of translation.alternatives) {
+            keywordMap.set(alt, {
+              native: alt,
+              normalized: translation.normalized || normalized2
+            });
+          }
+        }
+      }
+    }
+    if (profile.references) {
+      for (const [normalized2, native] of Object.entries(profile.references)) {
+        keywordMap.set(native, { native, normalized: normalized2 });
+      }
+    }
+    if (profile.roleMarkers) {
+      for (const [role, marker] of Object.entries(profile.roleMarkers)) {
+        if (marker.primary) {
+          keywordMap.set(marker.primary, { native: marker.primary, normalized: role });
+        }
+        if (marker.alternatives) {
+          for (const alt of marker.alternatives) {
+            keywordMap.set(alt, { native: alt, normalized: role });
+          }
+        }
+      }
+    }
+    for (const extra of extras) {
+      keywordMap.set(extra.native, extra);
+    }
+    this.profileKeywords = Array.from(keywordMap.values()).sort(
+      (a, b) => b.native.length - a.native.length
+    );
+    this.profileKeywordMap = /* @__PURE__ */ new Map();
+    for (const keyword of this.profileKeywords) {
+      this.profileKeywordMap.set(keyword.native.toLowerCase(), keyword);
+      const normalized2 = this.removeDiacritics(keyword.native);
+      if (normalized2 !== keyword.native && !this.profileKeywordMap.has(normalized2.toLowerCase())) {
+        this.profileKeywordMap.set(normalized2.toLowerCase(), keyword);
+      }
+    }
+  }
+  /**
+   * Remove diacritical marks from a word for normalization.
+   * Primarily for Arabic (shadda, fatha, kasra, damma, sukun, etc.)
+   * but could be extended for other languages.
+   *
+   * @param word - Word to normalize
+   * @returns Word without diacritics
+   */
+  removeDiacritics(word) {
+    return word.replace(/[\u064B-\u0652\u0670]/g, "");
+  }
+  /**
+   * Try to match a keyword from profile at the current position.
+   * Uses longest-first greedy matching (important for non-space languages).
+   *
+   * @param input - Input string
+   * @param pos - Current position
+   * @returns Token if matched, null otherwise
+   */
+  tryProfileKeyword(input, pos) {
+    for (const entry of this.profileKeywords) {
+      if (input.slice(pos).startsWith(entry.native)) {
+        return createToken(
+          entry.native,
+          "keyword",
+          createPosition(pos, pos + entry.native.length),
+          entry.normalized
+        );
+      }
+    }
+    return null;
+  }
+  /**
+   * Check if the remaining input starts with any known keyword.
+   * Useful for non-space languages to detect word boundaries.
+   *
+   * @param input - Input string
+   * @param pos - Current position
+   * @returns true if a keyword starts at this position
+   */
+  isKeywordStart(input, pos) {
+    const remaining = input.slice(pos);
+    return this.profileKeywords.some((entry) => remaining.startsWith(entry.native));
+  }
+  /**
+   * Look up a keyword by native word (case-insensitive).
+   * O(1) lookup using the keyword map.
+   *
+   * @param native - Native word to look up
+   * @returns KeywordEntry if found, undefined otherwise
+   */
+  lookupKeyword(native) {
+    return this.profileKeywordMap.get(native.toLowerCase());
+  }
+  /**
+   * Check if a word is a known keyword (case-insensitive).
+   * O(1) lookup using the keyword map.
+   *
+   * @param native - Native word to check
+   * @returns true if the word is a keyword
+   */
+  isKeyword(native) {
+    return this.profileKeywordMap.has(native.toLowerCase());
+  }
+  /**
+   * Set the morphological normalizer for this tokenizer.
+   */
+  setNormalizer(normalizer) {
+    this.normalizer = normalizer;
+  }
+  /**
+   * Try to normalize a word using the morphological normalizer.
+   * Returns null if no normalizer is set or normalization fails.
+   *
+   * Note: We don't check isNormalizable() here because the individual tokenizers
+   * historically called normalize() directly without that check. The normalize()
+   * method itself handles returning noChange() for words that can't be normalized.
+   */
+  tryNormalize(word) {
+    if (!this.normalizer) return null;
+    const result = this.normalizer.normalize(word);
+    if (result.stem !== word && result.confidence >= 0.7) {
+      return result;
+    }
+    return null;
+  }
+  /**
+   * Try morphological normalization and keyword lookup.
+   *
+   * If the word can be normalized to a stem that matches a known keyword,
+   * returns a keyword token with morphological metadata (stem, stemConfidence).
+   *
+   * This is the common pattern for handling conjugated verbs across languages:
+   * 1. Normalize the word (e.g., "toggled" → "toggle")
+   * 2. Look up the stem in the keyword map
+   * 3. Create a token with both the original form and stem metadata
+   *
+   * @param word - The word to normalize and look up
+   * @param startPos - Start position for the token
+   * @param endPos - End position for the token
+   * @returns Token if stem matches a keyword, null otherwise
+   */
+  tryMorphKeywordMatch(word, startPos, endPos) {
+    const result = this.tryNormalize(word);
+    if (!result) return null;
+    const stemEntry = this.lookupKeyword(result.stem);
+    if (!stemEntry) return null;
+    const tokenOptions = {
+      normalized: stemEntry.normalized,
+      stem: result.stem,
+      stemConfidence: result.confidence
+    };
+    return createToken(word, "keyword", createPosition(startPos, endPos), tokenOptions);
+  }
+  /**
+   * Try to extract a CSS selector at the current position.
+   */
+  trySelector(input, pos) {
+    const selector = extractCssSelector(input, pos);
+    if (selector) {
+      return createToken(selector, "selector", createPosition(pos, pos + selector.length));
+    }
+    return null;
+  }
+  /**
+   * Try to extract an event modifier at the current position.
+   * Event modifiers are .once, .debounce(N), .throttle(N), .queue(strategy)
+   */
+  tryEventModifier(input, pos) {
+    if (input[pos] !== ".") {
+      return null;
+    }
+    const match = input.slice(pos).match(/^\.(?:once|debounce|throttle|queue)(?:\(([^)]+)\))?(?:\s|$|\.)/);
+    if (!match) {
+      return null;
+    }
+    const fullMatch = match[0].replace(/(\s|\.)$/, "");
+    const modifierName = fullMatch.slice(1).split("(")[0];
+    const value = match[1];
+    const token = createToken(
+      fullMatch,
+      "event-modifier",
+      createPosition(pos, pos + fullMatch.length)
+    );
+    return {
+      ...token,
+      metadata: {
+        modifierName,
+        value: value ? modifierName === "queue" ? value : parseInt(value, 10) : void 0
+      }
+    };
+  }
+  /**
+   * Try to extract a string literal at the current position.
+   */
+  tryString(input, pos) {
+    const literal = extractStringLiteral(input, pos);
+    if (literal) {
+      return createToken(literal, "literal", createPosition(pos, pos + literal.length));
+    }
+    return null;
+  }
+  /**
+   * Try to extract a number at the current position.
+   */
+  tryNumber(input, pos) {
+    const number = extractNumber(input, pos);
+    if (number) {
+      return createToken(number, "literal", createPosition(pos, pos + number.length));
+    }
+    return null;
+  }
+  /**
+   * Try to match a time unit from a list of patterns.
+   *
+   * @param input - Input string
+   * @param pos - Position after the number
+   * @param timeUnits - Array of time unit mappings (native pattern → standard suffix)
+   * @param skipWhitespace - Whether to skip whitespace before time unit (default: false)
+   * @returns Object with matched suffix and new position, or null if no match
+   */
+  tryMatchTimeUnit(input, pos, timeUnits, skipWhitespace = false) {
+    let unitPos = pos;
+    if (skipWhitespace) {
+      while (unitPos < input.length && isWhitespace(input[unitPos])) {
+        unitPos++;
+      }
+    }
+    const remaining = input.slice(unitPos);
+    for (const unit of timeUnits) {
+      const candidate = remaining.slice(0, unit.length);
+      const matches = unit.caseInsensitive ? candidate.toLowerCase() === unit.pattern.toLowerCase() : candidate === unit.pattern;
+      if (matches) {
+        if (unit.notFollowedBy) {
+          const nextChar = remaining[unit.length] || "";
+          if (nextChar === unit.notFollowedBy) continue;
+        }
+        if (unit.checkBoundary) {
+          const nextChar = remaining[unit.length] || "";
+          if (isAsciiIdentifierChar(nextChar)) continue;
+        }
+        return { suffix: unit.suffix, endPos: unitPos + unit.length };
+      }
+    }
+    return null;
+  }
+  /**
+   * Parse a base number (sign, integer, decimal) without time units.
+   * Returns the number string and end position.
+   *
+   * @param input - Input string
+   * @param startPos - Start position
+   * @param allowSign - Whether to allow +/- sign (default: true)
+   * @returns Object with number string and end position, or null
+   */
+  parseBaseNumber(input, startPos, allowSign = true) {
+    let pos = startPos;
+    let number = "";
+    if (allowSign && (input[pos] === "-" || input[pos] === "+")) {
+      number += input[pos++];
+    }
+    if (pos >= input.length || !isDigit(input[pos])) {
+      return null;
+    }
+    while (pos < input.length && isDigit(input[pos])) {
+      number += input[pos++];
+    }
+    if (pos < input.length && input[pos] === ".") {
+      number += input[pos++];
+      while (pos < input.length && isDigit(input[pos])) {
+        number += input[pos++];
+      }
+    }
+    if (!number || number === "-" || number === "+") return null;
+    return { number, endPos: pos };
+  }
+  /**
+   * Try to extract a number with native language time units.
+   *
+   * This is a template method that handles the common pattern:
+   * 1. Parse the base number (sign, integer, decimal)
+   * 2. Try to match native language time units
+   * 3. Fall back to standard time units (ms, s, m, h)
+   *
+   * @param input - Input string
+   * @param pos - Start position
+   * @param nativeTimeUnits - Language-specific time unit mappings
+   * @param options - Configuration options
+   * @returns Token if number found, null otherwise
+   */
+  tryNumberWithTimeUnits(input, pos, nativeTimeUnits, options = {}) {
+    const { allowSign = true, skipWhitespace = false } = options;
+    const baseResult = this.parseBaseNumber(input, pos, allowSign);
+    if (!baseResult) return null;
+    let { number, endPos } = baseResult;
+    const allUnits = [...nativeTimeUnits, ..._BaseTokenizer.STANDARD_TIME_UNITS];
+    const timeMatch = this.tryMatchTimeUnit(input, endPos, allUnits, skipWhitespace);
+    if (timeMatch) {
+      number += timeMatch.suffix;
+      endPos = timeMatch.endPos;
+    }
+    return createToken(number, "literal", createPosition(pos, endPos));
+  }
+  /**
+   * Try to extract a URL at the current position.
+   * Handles /path, ./path, ../path, //domain.com, http://, https://
+   */
+  tryUrl(input, pos) {
+    const url = extractUrl(input, pos);
+    if (url) {
+      return createToken(url, "url", createPosition(pos, pos + url.length));
+    }
+    return null;
+  }
+  /**
+   * Try to extract a variable reference (:varname) at the current position.
+   * In hyperscript, :x refers to a local variable named x.
+   */
+  tryVariableRef(input, pos) {
+    if (input[pos] !== ":") return null;
+    if (pos + 1 >= input.length) return null;
+    if (!isAsciiIdentifierChar(input[pos + 1])) return null;
+    let endPos = pos + 1;
+    while (endPos < input.length && isAsciiIdentifierChar(input[endPos])) {
+      endPos++;
+    }
+    const varRef = input.slice(pos, endPos);
+    return createToken(varRef, "identifier", createPosition(pos, endPos));
+  }
+  /**
+   * Try to extract an operator or punctuation token at the current position.
+   * Handles two-character operators (==, !=, etc.) and single-character operators.
+   */
+  tryOperator(input, pos) {
+    const twoChar = input.slice(pos, pos + 2);
+    if (["==", "!=", "<=", ">=", "&&", "||", "->"].includes(twoChar)) {
+      return createToken(twoChar, "operator", createPosition(pos, pos + 2));
+    }
+    const oneChar = input[pos];
+    if (["<", ">", "!", "+", "-", "*", "/", "="].includes(oneChar)) {
+      return createToken(oneChar, "operator", createPosition(pos, pos + 1));
+    }
+    if (["(", ")", "{", "}", ",", ";", ":"].includes(oneChar)) {
+      return createToken(oneChar, "punctuation", createPosition(pos, pos + 1));
+    }
+    return null;
+  }
+  /**
+   * Try to match a multi-character particle from a list.
+   *
+   * Used by languages like Japanese, Korean, and Chinese that have
+   * multi-character particles (e.g., Japanese から, まで, より).
+   *
+   * @param input - Input string
+   * @param pos - Current position
+   * @param particles - Array of multi-character particles to match
+   * @returns Token if matched, null otherwise
+   */
+  tryMultiCharParticle(input, pos, particles) {
+    for (const particle of particles) {
+      if (input.slice(pos, pos + particle.length) === particle) {
+        return createToken(particle, "particle", createPosition(pos, pos + particle.length));
+      }
+    }
+    return null;
+  }
+};
+/**
+ * Configuration for native language time units.
+ * Maps patterns to their standard suffix (ms, s, m, h).
+ */
+_BaseTokenizer.STANDARD_TIME_UNITS = [
+  { pattern: "ms", suffix: "ms", length: 2 },
+  { pattern: "s", suffix: "s", length: 1, checkBoundary: true },
+  { pattern: "m", suffix: "m", length: 1, checkBoundary: true, notFollowedBy: "s" },
+  { pattern: "h", suffix: "h", length: 1, checkBoundary: true }
+];
+var BaseTokenizer = _BaseTokenizer;
+
+// src/tokenizers/morphology/types.ts
+function noChange(word) {
+  return { stem: word, confidence: 1 };
+}
+function normalized(stem, confidence, metadata) {
+  if (metadata) {
+    return { stem, confidence, metadata };
+  }
+  return { stem, confidence };
+}
+
+// src/tokenizers/morphology/japanese-normalizer.ts
+var JAPANESE_SUFFIX_RULES = [
+  // Conditional forms - very common for event handlers (longest first)
+  // したら/すると/すれば are する verb conditionals
+  { pattern: "\u3057\u305F\u3089", confidence: 0.88, conjugationType: "conditional-tara", minStemLength: 2 },
+  { pattern: "\u3059\u308B\u3068", confidence: 0.88, conjugationType: "conditional-to", minStemLength: 2 },
+  { pattern: "\u3059\u308C\u3070", confidence: 0.85, conjugationType: "conditional-ba", minStemLength: 2 },
+  // たら/れば are regular verb conditionals
+  { pattern: "\u305F\u3089", confidence: 0.85, conjugationType: "conditional-tara", minStemLength: 2 },
+  { pattern: "\u308C\u3070", confidence: 0.82, conjugationType: "conditional-ba", minStemLength: 2 },
+  // Compound forms (longest first)
+  { pattern: "\u3066\u3044\u307E\u3057\u305F", confidence: 0.82, conjugationType: "past", minStemLength: 2 },
+  { pattern: "\u3066\u3044\u307E\u3059", confidence: 0.85, conjugationType: "progressive", minStemLength: 2 },
+  { pattern: "\u3066\u304F\u3060\u3055\u3044", confidence: 0.85, conjugationType: "request", minStemLength: 2 },
+  { pattern: "\u3067\u304F\u3060\u3055\u3044", confidence: 0.85, conjugationType: "request", minStemLength: 2 },
+  { pattern: "\u3066\u3044\u308B", confidence: 0.85, conjugationType: "progressive", minStemLength: 2 },
+  { pattern: "\u3066\u304A\u304F", confidence: 0.82, conjugationType: "progressive", minStemLength: 2 },
+  { pattern: "\u3066\u307F\u308B", confidence: 0.82, conjugationType: "progressive", minStemLength: 2 },
+  { pattern: "\u3066\u3042\u308B", confidence: 0.82, conjugationType: "progressive", minStemLength: 2 },
+  // Casual request forms
+  { pattern: "\u3066\u304F\u308C", confidence: 0.8, conjugationType: "casual-request", minStemLength: 2 },
+  { pattern: "\u3067\u304F\u308C", confidence: 0.8, conjugationType: "casual-request", minStemLength: 2 },
+  // Contracted/colloquial forms (ちゃう/じゃう = てしまう/でしまう)
+  { pattern: "\u3061\u3083\u3063\u305F", confidence: 0.82, conjugationType: "contracted-past", minStemLength: 2 },
+  { pattern: "\u3058\u3083\u3063\u305F", confidence: 0.82, conjugationType: "contracted-past", minStemLength: 2 },
+  { pattern: "\u3061\u3083\u3046", confidence: 0.82, conjugationType: "contracted", minStemLength: 2 },
+  { pattern: "\u3058\u3083\u3046", confidence: 0.82, conjugationType: "contracted", minStemLength: 2 },
+  // Polite forms
+  { pattern: "\u307E\u3057\u305F", confidence: 0.85, conjugationType: "past", minStemLength: 2 },
+  { pattern: "\u307E\u305B\u3093", confidence: 0.85, conjugationType: "negative", minStemLength: 2 },
+  { pattern: "\u307E\u3059", confidence: 0.85, conjugationType: "polite", minStemLength: 2 },
+  // て/た forms (very common)
+  { pattern: "\u3066", confidence: 0.85, conjugationType: "te-form", minStemLength: 2 },
+  { pattern: "\u305F", confidence: 0.85, conjugationType: "past", minStemLength: 2 },
+  // Negative forms
+  { pattern: "\u306A\u3044", confidence: 0.82, conjugationType: "negative", minStemLength: 2 },
+  { pattern: "\u306A\u304B\u3063\u305F", confidence: 0.82, conjugationType: "past", minStemLength: 2 },
+  // Potential forms
+  { pattern: "\u3089\u308C\u308B", confidence: 0.8, conjugationType: "potential", minStemLength: 2 },
+  { pattern: "\u308C\u308B", confidence: 0.78, conjugationType: "potential", minStemLength: 2 },
+  // Passive forms
+  { pattern: "\u3089\u308C\u305F", confidence: 0.8, conjugationType: "passive", minStemLength: 2 },
+  // Causative forms
+  { pattern: "\u3055\u305B\u308B", confidence: 0.8, conjugationType: "causative", minStemLength: 2 },
+  { pattern: "\u305B\u308B", confidence: 0.78, conjugationType: "causative", minStemLength: 2 },
+  // Volitional forms
+  { pattern: "\u3088\u3046", confidence: 0.8, conjugationType: "volitional", minStemLength: 2 },
+  // Dictionary form ending (る-verbs) - lower confidence due to ambiguity
+  { pattern: "\u308B", confidence: 0.75, conjugationType: "dictionary", minStemLength: 3 }
+];
+var SURU_PATTERNS = [
+  // Conditional forms (most important for native idioms)
+  { pattern: "\u3057\u305F\u3089", confidence: 0.88, conjugationType: "conditional-tara" },
+  { pattern: "\u3059\u308B\u3068", confidence: 0.88, conjugationType: "conditional-to" },
+  { pattern: "\u3059\u308C\u3070", confidence: 0.85, conjugationType: "conditional-ba" },
+  // Progressive forms
+  { pattern: "\u3057\u3066\u3044\u307E\u3059", confidence: 0.85, conjugationType: "progressive" },
+  { pattern: "\u3057\u3066\u3044\u308B", confidence: 0.85, conjugationType: "progressive" },
+  // Other forms
+  { pattern: "\u3057\u307E\u3057\u305F", confidence: 0.85, conjugationType: "past" },
+  { pattern: "\u3057\u307E\u3059", confidence: 0.85, conjugationType: "polite" },
+  { pattern: "\u3057\u306A\u3044", confidence: 0.82, conjugationType: "negative" },
+  { pattern: "\u3057\u3066", confidence: 0.85, conjugationType: "te-form" },
+  { pattern: "\u3057\u305F", confidence: 0.85, conjugationType: "past" },
+  { pattern: "\u3059\u308B", confidence: 0.88, conjugationType: "dictionary" }
+];
+function isHiragana(char) {
+  const code = char.charCodeAt(0);
+  return code >= 12352 && code <= 12447;
+}
+function isKatakana(char) {
+  const code = char.charCodeAt(0);
+  return code >= 12448 && code <= 12543;
+}
+function isKanji(char) {
+  const code = char.charCodeAt(0);
+  return code >= 19968 && code <= 40959 || code >= 13312 && code <= 19903;
+}
+function containsJapanese(word) {
+  for (const char of word) {
+    if (isHiragana(char) || isKatakana(char) || isKanji(char)) {
+      return true;
+    }
+  }
+  return false;
+}
+var JapaneseMorphologicalNormalizer = class {
+  constructor() {
+    this.language = "ja";
+  }
+  /**
+   * Check if a word might be a Japanese verb that can be normalized.
+   */
+  isNormalizable(word) {
+    if (!containsJapanese(word)) return false;
+    if (word.length < 2) return false;
+    const lastChar = word[word.length - 1];
+    return isHiragana(lastChar);
+  }
+  /**
+   * Normalize a Japanese word to its stem form.
+   */
+  normalize(word) {
+    const compoundResult = this.normalizeCompound(word);
+    if (compoundResult) return compoundResult;
+    const suruResult = this.trySuruNormalization(word);
+    if (suruResult) return suruResult;
+    for (const rule of JAPANESE_SUFFIX_RULES) {
+      if (word.endsWith(rule.pattern)) {
+        const stem = word.slice(0, -rule.pattern.length);
+        const minLength = rule.minStemLength ?? 2;
+        if (stem.length < minLength) continue;
+        const metadata = {
+          removedSuffixes: [rule.pattern]
+        };
+        if (rule.conjugationType) {
+          metadata.conjugationType = rule.conjugationType;
+        }
+        return normalized(stem, rule.confidence, metadata);
+      }
+    }
+    return noChange(word);
+  }
+  /**
+   * Try to normalize a する verb.
+   */
+  trySuruNormalization(word) {
+    for (const pattern of SURU_PATTERNS) {
+      if (word.endsWith(pattern.pattern)) {
+        const stem = word.slice(0, -pattern.pattern.length);
+        if (stem.length < 1) continue;
+        return normalized(stem, pattern.confidence, {
+          removedSuffixes: [pattern.pattern],
+          conjugationType: pattern.conjugationType,
+          originalForm: "suru-verb"
+        });
+      }
+    }
+    return null;
+  }
+  /**
+   * Normalize compound conjugations (multi-layer suffixes).
+   * These are combinations like ていなかった (was not doing), でいない (is not doing).
+   * Handles cases that single-suffix rules miss.
+   */
+  normalizeCompound(word) {
+    const compoundPatterns = [
+      // Progressive negative past forms
+      {
+        pattern: "\u3066\u3044\u306A\u304B\u3063\u305F",
+        suffixes: ["\u3066", "\u3044", "\u306A\u304B\u3063\u305F"],
+        confidence: 0.8,
+        minStemLength: 2
+      },
+      {
+        pattern: "\u3067\u3044\u306A\u304B\u3063\u305F",
+        suffixes: ["\u3067", "\u3044", "\u306A\u304B\u3063\u305F"],
+        confidence: 0.8,
+        minStemLength: 2
+      },
+      // Progressive negative forms
+      { pattern: "\u3066\u3044\u306A\u3044", suffixes: ["\u3066", "\u3044", "\u306A\u3044"], confidence: 0.82, minStemLength: 2 },
+      { pattern: "\u3067\u3044\u306A\u3044", suffixes: ["\u3067", "\u3044", "\u306A\u3044"], confidence: 0.82, minStemLength: 2 },
+      // Progressive past forms
+      { pattern: "\u3066\u3044\u305F", suffixes: ["\u3066", "\u3044", "\u305F"], confidence: 0.85, minStemLength: 2 },
+      { pattern: "\u3067\u3044\u305F", suffixes: ["\u3067", "\u3044", "\u305F"], confidence: 0.85, minStemLength: 2 }
+    ];
+    for (const { pattern, suffixes, confidence, minStemLength } of compoundPatterns) {
+      if (word.endsWith(pattern)) {
+        const stem = word.slice(0, -pattern.length);
+        if (stem.length < minStemLength) continue;
+        return normalized(stem, confidence, {
+          removedSuffixes: suffixes,
+          conjugationType: "compound"
+        });
+      }
+    }
+    return null;
+  }
+};
+var japaneseMorphologicalNormalizer = new JapaneseMorphologicalNormalizer();
+
+// src/generators/profiles/japanese.ts
+var japaneseProfile = {
+  code: "ja",
+  name: "Japanese",
+  nativeName: "\u65E5\u672C\u8A9E",
+  direction: "ltr",
+  wordOrder: "SOV",
+  markingStrategy: "particle",
+  usesSpaces: false,
+  // Japanese uses verb stem/masu-stem form, no clear infinitive/imperative distinction
+  // for UI commands. Uses katakana loanwords (トグル) or native stems (切り替え)
+  defaultVerbForm: "base",
+  verb: {
+    position: "end",
+    suffixes: ["\u308B", "\u3066", "\u305F", "\u307E\u3059", "\u306A\u3044"],
+    subjectDrop: true
+  },
+  references: {
+    me: "\u81EA\u5206",
+    // "self" - in hyperscript context, refers to current element
+    it: "\u305D\u308C",
+    // "it"
+    you: "\u3042\u306A\u305F",
+    // "you"
+    result: "\u7D50\u679C",
+    event: "\u30A4\u30D9\u30F3\u30C8",
+    target: "\u30BF\u30FC\u30B2\u30C3\u30C8",
+    body: "\u30DC\u30C7\u30A3"
+  },
+  possessive: {
+    marker: "\u306E",
+    markerPosition: "between",
+    // In Japanese: 自分の value (jibun no value) = "my value"
+    keywords: {
+      \u79C1\u306E: "me",
+      // watashi no (my)
+      \u3042\u306A\u305F\u306E: "you",
+      // anata no (your)
+      \u305D\u306E: "it"
+      // sono (its)
+    }
+  },
+  roleMarkers: {
+    patient: { primary: "\u3092", position: "after" },
+    destination: { primary: "\u306B", alternatives: ["\u3078", "\u3067"], position: "after" },
+    source: { primary: "\u304B\u3089", position: "after" },
+    style: { primary: "\u3067", position: "after" },
+    event: { primary: "\u3092", position: "after" }
+    // Event as object marker
+    // Possession marker for "X's Y" patterns
+    // Note: の is used between target and patient: #button の .active
+  },
+  keywords: {
+    // Class/Attribute operations
+    toggle: {
+      primary: "\u5207\u308A\u66FF\u3048",
+      alternatives: ["\u5207\u308A\u66FF\u3048\u308B", "\u30C8\u30B0\u30EB", "\u30C8\u30B0\u30EB\u3059\u308B"],
+      normalized: "toggle"
+    },
+    add: { primary: "\u8FFD\u52A0", alternatives: ["\u8FFD\u52A0\u3059\u308B", "\u52A0\u3048\u308B"], normalized: "add" },
+    remove: { primary: "\u524A\u9664", alternatives: ["\u524A\u9664\u3059\u308B", "\u53D6\u308A\u9664\u304F"], normalized: "remove" },
+    // Content operations
+    put: { primary: "\u7F6E\u304F", alternatives: ["\u5165\u308C\u308B", "\u30BB\u30C3\u30C8"], normalized: "put" },
+    append: { primary: "\u672B\u5C3E\u8FFD\u52A0", alternatives: ["\u672B\u5C3E\u306B\u8FFD\u52A0", "\u30A2\u30DA\u30F3\u30C9"], normalized: "append" },
+    prepend: {
+      primary: "\u5148\u982D\u8FFD\u52A0",
+      alternatives: ["\u5148\u982D\u306B\u8FFD\u52A0", "\u30D7\u30EA\u30DA\u30F3\u30C9"],
+      normalized: "prepend"
+    },
+    take: { primary: "\u53D6\u308B", alternatives: ["\u53D6\u5F97"], normalized: "take" },
+    make: { primary: "\u4F5C\u308B", alternatives: ["\u4F5C\u6210"], normalized: "make" },
+    clone: { primary: "\u8907\u88FD", alternatives: ["\u30AF\u30ED\u30FC\u30F3"], normalized: "clone" },
+    swap: { primary: "\u4EA4\u63DB", alternatives: ["\u30B9\u30EF\u30C3\u30D7", "\u5165\u308C\u66FF\u3048"], normalized: "swap" },
+    morph: { primary: "\u5909\u5F62", alternatives: ["\u30E2\u30FC\u30D5", "\u5909\u63DB"], normalized: "morph" },
+    // Variable operations
+    set: { primary: "\u8A2D\u5B9A", alternatives: ["\u8A2D\u5B9A\u3059\u308B", "\u30BB\u30C3\u30C8"], normalized: "set" },
+    get: { primary: "\u53D6\u5F97", alternatives: ["\u53D6\u5F97\u3059\u308B", "\u30B2\u30C3\u30C8"], normalized: "get" },
+    increment: {
+      primary: "\u5897\u52A0",
+      alternatives: ["\u5897\u3084\u3059", "\u30A4\u30F3\u30AF\u30EA\u30E1\u30F3\u30C8"],
+      normalized: "increment"
+    },
+    decrement: {
+      primary: "\u6E1B\u5C11",
+      alternatives: ["\u6E1B\u3089\u3059", "\u30C7\u30AF\u30EA\u30E1\u30F3\u30C8"],
+      normalized: "decrement"
+    },
+    log: { primary: "\u8A18\u9332", alternatives: ["\u30ED\u30B0", "\u51FA\u529B"], normalized: "log" },
+    // Visibility
+    show: { primary: "\u8868\u793A", alternatives: ["\u8868\u793A\u3059\u308B", "\u898B\u305B\u308B"], normalized: "show" },
+    hide: { primary: "\u96A0\u3059", alternatives: ["\u975E\u8868\u793A", "\u975E\u8868\u793A\u306B\u3059\u308B"], normalized: "hide" },
+    transition: {
+      primary: "\u9077\u79FB",
+      alternatives: ["\u30C8\u30E9\u30F3\u30B8\u30B7\u30E7\u30F3", "\u30A2\u30CB\u30E1\u30FC\u30B7\u30E7\u30F3"],
+      normalized: "transition"
+    },
+    // Events
+    on: { primary: "\u3067", alternatives: ["\u6642", "\u3068\u304D"], normalized: "on" },
+    trigger: { primary: "\u5F15\u304D\u91D1", alternatives: ["\u767A\u706B", "\u30C8\u30EA\u30AC\u30FC"], normalized: "trigger" },
+    send: { primary: "\u9001\u308B", alternatives: ["\u9001\u4FE1"], normalized: "send" },
+    // DOM focus
+    focus: { primary: "\u30D5\u30A9\u30FC\u30AB\u30B9", alternatives: ["\u96C6\u4E2D"], normalized: "focus" },
+    blur: { primary: "\u307C\u304B\u3057", alternatives: ["\u30D5\u30A9\u30FC\u30AB\u30B9\u89E3\u9664"], normalized: "blur" },
+    // Navigation
+    go: { primary: "\u79FB\u52D5", alternatives: ["\u884C\u304F", "\u30CA\u30D3\u30B2\u30FC\u30C8"], normalized: "go" },
+    // Async
+    wait: { primary: "\u5F85\u3064", alternatives: ["\u5F85\u6A5F"], normalized: "wait" },
+    fetch: { primary: "\u53D6\u5F97", alternatives: ["\u30D5\u30A7\u30C3\u30C1"], normalized: "fetch" },
+    settle: { primary: "\u5B89\u5B9A", alternatives: ["\u843D\u3061\u7740\u304F"], normalized: "settle" },
+    // Control flow
+    if: { primary: "\u3082\u3057", alternatives: ["\u6761\u4EF6"], normalized: "if" },
+    when: { primary: "\u3068\u304D", normalized: "when" },
+    where: { primary: "\u3069\u3053", normalized: "where" },
+    else: { primary: "\u305D\u3046\u3067\u306A\u3051\u308C\u3070", alternatives: ["\u305D\u308C\u4EE5\u5916"], normalized: "else" },
+    repeat: { primary: "\u7E70\u308A\u8FD4\u3057", alternatives: ["\u7E70\u308A\u8FD4\u3059", "\u30EA\u30D4\u30FC\u30C8"], normalized: "repeat" },
+    for: { primary: "\u305F\u3081\u306B", alternatives: ["\u5404"], normalized: "for" },
+    while: { primary: "\u306E\u9593", alternatives: ["\u9593"], normalized: "while" },
+    continue: { primary: "\u7D9A\u3051\u308B", alternatives: ["\u7D99\u7D9A"], normalized: "continue" },
+    halt: { primary: "\u505C\u6B62", alternatives: ["\u6B62\u3081\u308B", "\u30CF\u30EB\u30C8"], normalized: "halt" },
+    throw: { primary: "\u6295\u3052\u308B", alternatives: ["\u30B9\u30ED\u30FC"], normalized: "throw" },
+    call: { primary: "\u547C\u3073\u51FA\u3057", alternatives: ["\u30B3\u30FC\u30EB", "\u547C\u3076"], normalized: "call" },
+    return: { primary: "\u623B\u308B", alternatives: ["\u8FD4\u3059", "\u30EA\u30BF\u30FC\u30F3"], normalized: "return" },
+    then: { primary: "\u305D\u308C\u304B\u3089", alternatives: ["\u6B21\u306B", "\u305D\u3057\u3066"], normalized: "then" },
+    and: { primary: "\u305D\u3057\u3066", alternatives: ["\u3068", "\u307E\u305F"], normalized: "and" },
+    end: { primary: "\u7D42\u308F\u308A", alternatives: ["\u7D42\u4E86", "\u304A\u308F\u308A"], normalized: "end" },
+    // Advanced
+    js: { primary: "JS\u5B9F\u884C", alternatives: ["js"], normalized: "js" },
+    async: { primary: "\u975E\u540C\u671F", alternatives: ["\u30A2\u30B7\u30F3\u30AF"], normalized: "async" },
+    tell: { primary: "\u4F1D\u3048\u308B", alternatives: ["\u30C6\u30EB"], normalized: "tell" },
+    default: { primary: "\u65E2\u5B9A", alternatives: ["\u30C7\u30D5\u30A9\u30EB\u30C8"], normalized: "default" },
+    init: { primary: "\u521D\u671F\u5316", alternatives: ["\u30A4\u30CB\u30C3\u30C8"], normalized: "init" },
+    behavior: { primary: "\u632F\u308B\u821E\u3044", alternatives: ["\u30D3\u30D8\u30A4\u30D3\u30A2"], normalized: "behavior" },
+    install: { primary: "\u30A4\u30F3\u30B9\u30C8\u30FC\u30EB", alternatives: ["\u5C0E\u5165"], normalized: "install" },
+    measure: { primary: "\u6E2C\u5B9A", alternatives: ["\u8A08\u6E2C", "\u30E1\u30B8\u30E3\u30FC"], normalized: "measure" },
+    // Modifiers
+    into: { primary: "\u3078", alternatives: ["\u306B"], normalized: "into" },
+    before: { primary: "\u524D\u306B", alternatives: ["\u524D"], normalized: "before" },
+    after: { primary: "\u5F8C\u306B", alternatives: ["\u5F8C"], normalized: "after" },
+    // Event modifiers (for repeat until event)
+    until: { primary: "\u307E\u3067", alternatives: ["\u8FC4"], normalized: "until" },
+    event: { primary: "\u30A4\u30D9\u30F3\u30C8", alternatives: ["\u4E8B\u8C61"], normalized: "event" },
+    from: { primary: "\u304B\u3089", normalized: "from" }
+  },
+  tokenization: {
+    particles: ["\u3092", "\u306B", "\u3067", "\u304B\u3089", "\u306E", "\u304C", "\u306F", "\u3082", "\u3078", "\u3068"],
+    boundaryStrategy: "particle"
+  }
+};
+
+// src/tokenizers/japanese.ts
+var isHiragana2 = createUnicodeRangeClassifier([[12352, 12447]]);
+var isKatakana2 = createUnicodeRangeClassifier([[12448, 12543]]);
+var isKanji2 = createUnicodeRangeClassifier([
+  [19968, 40959],
+  // CJK Unified Ideographs
+  [13312, 19903]
+  // CJK Unified Ideographs Extension A
+]);
+var isJapanese = combineClassifiers(isHiragana2, isKatakana2, isKanji2);
+var PARTICLES = /* @__PURE__ */ new Set([
+  "\u3092",
+  // wo - object marker
+  "\u306B",
+  // ni - destination, time
+  "\u3067",
+  // de - location of action, means
+  "\u304B\u3089",
+  // kara - from
+  "\u307E\u3067",
+  // made - until
+  "\u3078",
+  // e - direction
+  "\u3068",
+  // to - and, with
+  "\u306E",
+  // no - possessive
+  "\u304C",
+  // ga - subject marker
+  "\u306F",
+  // wa - topic marker
+  "\u3082",
+  // mo - also
+  "\u3088\u308A"
+  // yori - than, from
+]);
+var SINGLE_CHAR_PARTICLES = /* @__PURE__ */ new Set(["\u3092", "\u306B", "\u3067", "\u3078", "\u3068", "\u306E", "\u304C", "\u306F", "\u3082"]);
+var MULTI_CHAR_PARTICLES = ["\u304B\u3089", "\u307E\u3067", "\u3088\u308A"];
+var PARTICLE_ROLES = /* @__PURE__ */ new Map([
+  ["\u3092", { role: "patient", confidence: 0.95, description: "object marker" }],
+  ["\u306B", { role: "destination", confidence: 0.85, description: "destination/time marker" }],
+  ["\u3067", { role: "manner", confidence: 0.88, description: "means/location marker" }],
+  ["\u304B\u3089", { role: "source", confidence: 0.9, description: "from/source marker" }],
+  ["\u307E\u3067", { role: "destination", confidence: 0.75, description: "until/boundary marker" }],
+  ["\u3078", { role: "destination", confidence: 0.9, description: "direction marker" }],
+  ["\u3068", { role: "style", confidence: 0.7, description: "with/and marker" }],
+  ["\u306E", { role: "patient", confidence: 0.6, description: "possessive marker" }],
+  ["\u304C", { role: "agent", confidence: 0.85, description: "subject marker" }],
+  ["\u306F", { role: "agent", confidence: 0.75, description: "topic marker" }],
+  ["\u3082", { role: "patient", confidence: 0.65, description: "also/too marker" }],
+  ["\u3088\u308A", { role: "source", confidence: 0.85, description: "from/than marker" }]
+]);
+var JAPANESE_EXTRAS = [
+  // Values/Literals
+  { native: "\u771F", normalized: "true" },
+  { native: "\u507D", normalized: "false" },
+  { native: "\u30CC\u30EB", normalized: "null" },
+  { native: "\u672A\u5B9A\u7FA9", normalized: "undefined" },
+  // Positional
+  { native: "\u6700\u521D", normalized: "first" },
+  { native: "\u6700\u5F8C", normalized: "last" },
+  { native: "\u6B21", normalized: "next" },
+  { native: "\u524D", normalized: "previous" },
+  { native: "\u6700\u3082\u8FD1\u3044", normalized: "closest" },
+  { native: "\u89AA", normalized: "parent" },
+  // Events
+  { native: "\u30AF\u30EA\u30C3\u30AF", normalized: "click" },
+  { native: "\u5909\u66F4", normalized: "change" },
+  { native: "\u9001\u4FE1", normalized: "submit" },
+  { native: "\u5165\u529B", normalized: "input" },
+  { native: "\u30ED\u30FC\u30C9", normalized: "load" },
+  { native: "\u30B9\u30AF\u30ED\u30FC\u30EB", normalized: "scroll" },
+  { native: "\u30AD\u30FC\u30C0\u30A6\u30F3", normalized: "keydown" },
+  { native: "\u30AD\u30FC\u30A2\u30C3\u30D7", normalized: "keyup" },
+  { native: "\u30DE\u30A6\u30B9\u30AA\u30FC\u30D0\u30FC", normalized: "mouseover" },
+  { native: "\u30DE\u30A6\u30B9\u30A2\u30A6\u30C8", normalized: "mouseout" },
+  { native: "\u30D6\u30E9\u30FC", normalized: "blur" },
+  // References (additional forms)
+  { native: "\u79C1", normalized: "me" },
+  { native: "\u79C1\u306E", normalized: "my" },
+  { native: "\u305D\u306E", normalized: "its" },
+  // Note: Attached particle forms (を切り替え, を追加, etc.) are intentionally NOT included
+  // because they would cause ambiguous parsing. The separate particle + verb pattern
+  // (を + 切り替え) is preferred for consistent semantic analysis.
+  // Conditional event forms
+  { native: "\u3057\u305F\u3089", normalized: "on" },
+  { native: "\u3059\u308B\u3068", normalized: "on" },
+  { native: "\u6642\u306B", normalized: "on" },
+  // Control flow helpers
+  { native: "\u3082\u3057", normalized: "if" },
+  // Starts with particle も, needs explicit entry
+  { native: "\u306A\u3089\u3070", normalized: "then" },
+  { native: "\u306A\u3089", normalized: "then" },
+  // Time units
+  { native: "\u79D2", normalized: "s" },
+  { native: "\u30DF\u30EA\u79D2", normalized: "ms" },
+  { native: "\u5206", normalized: "m" },
+  { native: "\u6642\u9593", normalized: "h" }
+];
+var JAPANESE_TIME_UNITS = [
+  { pattern: "\u30DF\u30EA\u79D2", suffix: "ms", length: 3 },
+  { pattern: "\u6642\u9593", suffix: "h", length: 2 },
+  { pattern: "\u79D2", suffix: "s", length: 1 },
+  { pattern: "\u5206", suffix: "m", length: 1 }
+];
+var JapaneseTokenizer = class extends BaseTokenizer {
+  constructor() {
+    super();
+    this.language = "ja";
+    this.direction = "ltr";
+    this.initializeKeywordsFromProfile(japaneseProfile, JAPANESE_EXTRAS);
+    this.normalizer = new JapaneseMorphologicalNormalizer();
+  }
+  tokenize(input) {
+    const tokens = [];
+    let pos = 0;
+    while (pos < input.length) {
+      if (isWhitespace(input[pos])) {
+        pos++;
+        continue;
+      }
+      if (isSelectorStart(input[pos])) {
+        const modifierToken = this.tryEventModifier(input, pos);
+        if (modifierToken) {
+          tokens.push(modifierToken);
+          pos = modifierToken.position.end;
+          continue;
+        }
+        const selectorToken = this.trySelector(input, pos);
+        if (selectorToken) {
+          tokens.push(selectorToken);
+          pos = selectorToken.position.end;
+          continue;
+        }
+      }
+      if (isQuote(input[pos])) {
+        const stringToken = this.tryString(input, pos);
+        if (stringToken) {
+          tokens.push(stringToken);
+          pos = stringToken.position.end;
+          continue;
+        }
+      }
+      if (isUrlStart(input, pos)) {
+        const urlToken = this.tryUrl(input, pos);
+        if (urlToken) {
+          tokens.push(urlToken);
+          pos = urlToken.position.end;
+          continue;
+        }
+      }
+      if (isDigit(input[pos])) {
+        const numberToken = this.extractJapaneseNumber(input, pos);
+        if (numberToken) {
+          tokens.push(numberToken);
+          pos = numberToken.position.end;
+          continue;
+        }
+      }
+      const varToken = this.tryVariableRef(input, pos);
+      if (varToken) {
+        tokens.push(varToken);
+        pos = varToken.position.end;
+        continue;
+      }
+      const multiParticle = this.tryMultiCharParticle(input, pos, MULTI_CHAR_PARTICLES);
+      if (multiParticle) {
+        const metadata = PARTICLE_ROLES.get(multiParticle.value);
+        if (metadata) {
+          tokens.push({
+            ...multiParticle,
+            metadata: {
+              particleRole: metadata.role,
+              particleConfidence: metadata.confidence
+            }
+          });
+        } else {
+          tokens.push(multiParticle);
+        }
+        pos = multiParticle.position.end;
+        continue;
+      }
+      if (SINGLE_CHAR_PARTICLES.has(input[pos])) {
+        const keywordToken = this.tryProfileKeyword(input, pos);
+        if (keywordToken && keywordToken.value.length > 1) {
+          tokens.push(keywordToken);
+          pos = keywordToken.position.end;
+          continue;
+        }
+        const particle = input[pos];
+        const metadata = PARTICLE_ROLES.get(particle);
+        if (metadata) {
+          tokens.push({
+            ...createToken(particle, "particle", createPosition(pos, pos + 1)),
+            metadata: {
+              particleRole: metadata.role,
+              particleConfidence: metadata.confidence
+            }
+          });
+        } else {
+          tokens.push(createToken(particle, "particle", createPosition(pos, pos + 1)));
+        }
+        pos++;
+        continue;
+      }
+      if (isJapanese(input[pos])) {
+        const wordToken = this.extractJapaneseWord(input, pos);
+        if (wordToken) {
+          tokens.push(wordToken);
+          pos = wordToken.position.end;
+          continue;
+        }
+      }
+      if (isAsciiIdentifierChar(input[pos])) {
+        const asciiToken = this.extractAsciiWord(input, pos);
+        if (asciiToken) {
+          tokens.push(asciiToken);
+          pos = asciiToken.position.end;
+          continue;
+        }
+      }
+      pos++;
+    }
+    return new TokenStreamImpl(tokens, "ja");
+  }
+  classifyToken(token) {
+    if (PARTICLES.has(token)) return "particle";
+    if (this.isKeyword(token)) return "keyword";
+    if (token.startsWith("#") || token.startsWith(".") || token.startsWith("[")) return "selector";
+    if (token.startsWith('"') || token.startsWith("'") || token.startsWith("\u300C")) return "literal";
+    if (/^\d/.test(token)) return "literal";
+    return "identifier";
+  }
+  /**
+   * Extract a Japanese word (sequence of kanji/kana).
+   * Stops at particles, ASCII, or whitespace.
+   *
+   * Uses morphological normalization to handle verb conjugations:
+   * 1. First checks if the exact word is in the keyword map
+   * 2. If not found, tries to strip conjugation suffixes and check again
+   */
+  extractJapaneseWord(input, startPos) {
+    let pos = startPos;
+    let word = "";
+    while (pos < input.length) {
+      const char = input[pos];
+      if (SINGLE_CHAR_PARTICLES.has(char) && word.length > 0) {
+        break;
+      }
+      let foundMulti = false;
+      for (const particle of MULTI_CHAR_PARTICLES) {
+        if (input.slice(pos, pos + particle.length) === particle && word.length > 0) {
+          foundMulti = true;
+          break;
+        }
+      }
+      if (foundMulti) break;
+      if (isJapanese(char)) {
+        word += char;
+        pos++;
+      } else {
+        break;
+      }
+    }
+    if (!word) return null;
+    const keywordEntry = this.lookupKeyword(word);
+    if (keywordEntry) {
+      return createToken(word, "keyword", createPosition(startPos, pos), keywordEntry.normalized);
+    }
+    const morphToken = this.tryMorphKeywordMatch(word, startPos, pos);
+    if (morphToken) return morphToken;
+    return createToken(word, "identifier", createPosition(startPos, pos));
+  }
+  /**
+   * Extract an ASCII word (for mixed Japanese/English content).
+   */
+  extractAsciiWord(input, startPos) {
+    let pos = startPos;
+    let word = "";
+    while (pos < input.length && isAsciiIdentifierChar(input[pos])) {
+      word += input[pos++];
+    }
+    if (!word) return null;
+    return createToken(word, "identifier", createPosition(startPos, pos));
+  }
+  /**
+   * Extract a number, including Japanese time unit suffixes.
+   * Japanese time units attach directly without whitespace.
+   */
+  extractJapaneseNumber(input, startPos) {
+    return this.tryNumberWithTimeUnits(input, startPos, JAPANESE_TIME_UNITS, {
+      allowSign: false,
+      skipWhitespace: false
+    });
+  }
+};
+var japaneseTokenizer = new JapaneseTokenizer();
+
+// src/languages/ja.ts
+registerLanguage("ja", japaneseTokenizer, japaneseProfile);
+export {
+  japaneseProfile,
+  japaneseTokenizer
+};
+//# sourceMappingURL=ja.js.map