@lokascript/semantic 1.0.0

package/dist/languages/zh.js

@@ -0,0 +1,1257 @@
+// 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, stem, stemConfidence } = normalizedOrOptions;
+    const token = { value, kind, position };
+    if (normalized !== void 0) {
+      token.normalized = normalized;
+    }
+    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 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 [normalized, translation] of Object.entries(profile.keywords)) {
+        keywordMap.set(translation.primary, {
+          native: translation.primary,
+          normalized: translation.normalized || normalized
+        });
+        if (translation.alternatives) {
+          for (const alt of translation.alternatives) {
+            keywordMap.set(alt, {
+              native: alt,
+              normalized: translation.normalized || normalized
+            });
+          }
+        }
+      }
+    }
+    if (profile.references) {
+      for (const [normalized, native] of Object.entries(profile.references)) {
+        keywordMap.set(native, { native, normalized });
+      }
+    }
+    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 normalized = this.removeDiacritics(keyword.native);
+      if (normalized !== keyword.native && !this.profileKeywordMap.has(normalized.toLowerCase())) {
+        this.profileKeywordMap.set(normalized.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/generators/profiles/chinese.ts
+var chineseProfile = {
+  code: "zh",
+  name: "Chinese",
+  nativeName: "\u4E2D\u6587",
+  direction: "ltr",
+  wordOrder: "SVO",
+  markingStrategy: "preposition",
+  // Uses prepositions but often implicit
+  usesSpaces: false,
+  verb: {
+    position: "second",
+    subjectDrop: true
+  },
+  references: {
+    me: "\u6211",
+    // "I/me"
+    it: "\u5B83",
+    // "it"
+    you: "\u4F60",
+    // "you"
+    result: "\u7ED3\u679C",
+    event: "\u4E8B\u4EF6",
+    target: "\u76EE\u6807",
+    body: "\u4E3B\u4F53"
+  },
+  possessive: {
+    marker: "\u7684",
+    // Possessive particle (de)
+    markerPosition: "between",
+    // Chinese: 我的 value (wǒ de value) = "my value"
+    keywords: {
+      // Compound possessive forms (pronoun + 的)
+      \u6211\u7684: "me",
+      // wǒ de (my)
+      \u4F60\u7684: "you",
+      // nǐ de (your)
+      \u5B83\u7684: "it",
+      // tā de (its)
+      \u4ED6\u7684: "it",
+      // tā de (his)
+      \u5979\u7684: "it"
+      // tā de (her)
+    }
+  },
+  roleMarkers: {
+    destination: { primary: "\u5728", alternatives: ["\u5230", "\u4E8E"], position: "before" },
+    source: { primary: "\u4ECE", alternatives: ["\u7531"], position: "before" },
+    patient: { primary: "\u628A", position: "before" },
+    // BA construction
+    style: { primary: "\u7528", alternatives: ["\u4EE5"], position: "before" }
+  },
+  keywords: {
+    // Class/Attribute operations
+    toggle: { primary: "\u5207\u6362", normalized: "toggle" },
+    add: { primary: "\u6DFB\u52A0", alternatives: ["\u52A0"], normalized: "add" },
+    remove: { primary: "\u79FB\u9664", alternatives: ["\u5220\u9664", "\u53BB\u6389"], normalized: "remove" },
+    // Content operations
+    put: { primary: "\u653E\u7F6E", alternatives: ["\u653E", "\u653E\u5165"], normalized: "put" },
+    append: { primary: "\u8FFD\u52A0", alternatives: ["\u9644\u52A0"], normalized: "append" },
+    prepend: { primary: "\u524D\u7F6E", alternatives: ["\u9884\u7F6E"], normalized: "prepend" },
+    take: { primary: "\u83B7\u53D6", normalized: "take" },
+    make: { primary: "\u5236\u4F5C", normalized: "make" },
+    clone: { primary: "\u590D\u5236", normalized: "clone" },
+    swap: { primary: "\u4EA4\u6362", normalized: "swap" },
+    morph: { primary: "\u53D8\u5F62", alternatives: ["\u8F6C\u6362"], normalized: "morph" },
+    // Variable operations
+    set: { primary: "\u8BBE\u7F6E", alternatives: ["\u8BBE\u5B9A"], normalized: "set" },
+    get: { primary: "\u83B7\u5F97", alternatives: ["\u83B7\u53D6", "\u53D6\u5F97"], normalized: "get" },
+    increment: { primary: "\u589E\u52A0", normalized: "increment" },
+    decrement: { primary: "\u51CF\u5C11", normalized: "decrement" },
+    log: { primary: "\u65E5\u5FD7", normalized: "log" },
+    // Visibility
+    show: { primary: "\u663E\u793A", alternatives: ["\u5C55\u793A"], normalized: "show" },
+    hide: { primary: "\u9690\u85CF", normalized: "hide" },
+    transition: { primary: "\u8FC7\u6E21", normalized: "transition" },
+    // Events
+    on: { primary: "\u5F53", alternatives: ["\u5728...\u65F6"], normalized: "on" },
+    trigger: { primary: "\u89E6\u53D1", normalized: "trigger" },
+    send: { primary: "\u53D1\u9001", normalized: "send" },
+    // DOM focus
+    focus: { primary: "\u805A\u7126", normalized: "focus" },
+    blur: { primary: "\u5931\u7126", normalized: "blur" },
+    // Navigation
+    go: { primary: "\u524D\u5F80", normalized: "go" },
+    // Async
+    wait: { primary: "\u7B49\u5F85", normalized: "wait" },
+    fetch: { primary: "\u83B7\u53D6", normalized: "fetch" },
+    settle: { primary: "\u7A33\u5B9A", normalized: "settle" },
+    // Control flow
+    if: { primary: "\u5982\u679C", normalized: "if" },
+    when: { primary: "\u5F53", normalized: "when" },
+    where: { primary: "\u54EA\u91CC", normalized: "where" },
+    else: { primary: "\u5426\u5219", normalized: "else" },
+    repeat: { primary: "\u91CD\u590D", normalized: "repeat" },
+    for: { primary: "\u4E3A", normalized: "for" },
+    while: { primary: "\u5F53", normalized: "while" },
+    continue: { primary: "\u7EE7\u7EED", normalized: "continue" },
+    halt: { primary: "\u505C\u6B62", normalized: "halt" },
+    throw: { primary: "\u629B\u51FA", normalized: "throw" },
+    call: { primary: "\u8C03\u7528", normalized: "call" },
+    return: { primary: "\u8FD4\u56DE", normalized: "return" },
+    then: { primary: "\u7136\u540E", alternatives: ["\u63A5\u7740", "\u4E4B\u540E"], normalized: "then" },
+    and: { primary: "\u5E76\u4E14", alternatives: ["\u548C", "\u800C\u4E14"], normalized: "and" },
+    end: { primary: "\u7ED3\u675F", alternatives: ["\u7EC8\u6B62", "\u5B8C"], normalized: "end" },
+    // Advanced
+    js: { primary: "JS\u6267\u884C", alternatives: ["js"], normalized: "js" },
+    async: { primary: "\u5F02\u6B65", normalized: "async" },
+    tell: { primary: "\u544A\u8BC9", normalized: "tell" },
+    default: { primary: "\u9ED8\u8BA4", normalized: "default" },
+    init: { primary: "\u521D\u59CB\u5316", normalized: "init" },
+    behavior: { primary: "\u884C\u4E3A", normalized: "behavior" },
+    install: { primary: "\u5B89\u88C5", normalized: "install" },
+    measure: { primary: "\u6D4B\u91CF", normalized: "measure" },
+    // Modifiers
+    into: { primary: "\u8FDB\u5165", normalized: "into" },
+    before: { primary: "\u4E4B\u524D", normalized: "before" },
+    after: { primary: "\u4E4B\u540E", normalized: "after" },
+    // Event modifiers (for repeat until event)
+    until: { primary: "\u76F4\u5230", normalized: "until" },
+    event: { primary: "\u4E8B\u4EF6", normalized: "event" },
+    from: { primary: "\u4ECE", normalized: "from" }
+  },
+  tokenization: {
+    boundaryStrategy: "character"
+  }
+};
+
+// src/tokenizers/chinese.ts
+var isChinese = createUnicodeRangeClassifier([
+  [19968, 40959],
+  // CJK Unified Ideographs
+  [13312, 19903],
+  // CJK Unified Ideographs Extension A
+  [131072, 173791],
+  // CJK Unified Ideographs Extension B
+  [63744, 64255],
+  // CJK Compatibility Ideographs
+  [194560, 195103]
+  // CJK Compatibility Ideographs Supplement
+]);
+var PARTICLES = /* @__PURE__ */ new Set([
+  "\u628A",
+  // ba - marks direct object (BA construction)
+  "\u5728",
+  // zai - at, in, on (location)
+  "\u4ECE",
+  // cong - from
+  "\u5230",
+  // dao - to, until
+  "\u5411",
+  // xiang - towards
+  "\u7ED9",
+  // gei - to, for (recipient)
+  "\u5BF9",
+  // dui - to, towards
+  "\u7528",
+  // yong - with, using
+  "\u88AB",
+  // bei - by (passive)
+  "\u8BA9",
+  // rang - let, allow
+  "\u7684",
+  // de - possessive/attributive
+  "\u5730",
+  // de - adverbial marker
+  "\u5F97",
+  // de - complement marker
+  "\u4E86",
+  // le - completion marker
+  "\u7740",
+  // zhe - progressive marker
+  "\u8FC7",
+  // guo - experiential marker
+  "\u5417",
+  // ma - question particle
+  "\u5462",
+  // ne - question/emphasis particle
+  "\u5427"
+  // ba - suggestion particle
+]);
+var MULTI_CHAR_PARTICLES = ["\u7136\u540E", "\u63A5\u7740", "\u5E76\u4E14", "\u6216\u8005", "\u5982\u679C", "\u90A3\u4E48", "\u5426\u5219"];
+var CHINESE_EXTRAS = [
+  // Values/Literals
+  { native: "\u771F", normalized: "true" },
+  { native: "\u5047", normalized: "false" },
+  { native: "\u7A7A", normalized: "null" },
+  { native: "\u672A\u5B9A\u4E49", normalized: "undefined" },
+  // Positional
+  { native: "\u7B2C\u4E00\u4E2A", normalized: "first" },
+  { native: "\u9996\u4E2A", normalized: "first" },
+  { native: "\u6700\u540E\u4E00\u4E2A", normalized: "last" },
+  { native: "\u672B\u4E2A", normalized: "last" },
+  { native: "\u4E0B\u4E00\u4E2A", normalized: "next" },
+  { native: "\u4E0A\u4E00\u4E2A", normalized: "previous" },
+  { native: "\u6700\u8FD1\u7684", normalized: "closest" },
+  { native: "\u7236\u7EA7", normalized: "parent" },
+  // Events
+  { native: "\u70B9\u51FB", normalized: "click" },
+  { native: "\u53CC\u51FB", normalized: "dblclick" },
+  { native: "\u8F93\u5165", normalized: "input" },
+  { native: "\u53D8\u66F4", normalized: "change" },
+  { native: "\u6539\u53D8", normalized: "change" },
+  { native: "\u63D0\u4EA4", normalized: "submit" },
+  { native: "\u6309\u952E", normalized: "keydown" },
+  { native: "\u91CA\u653E\u952E", normalized: "keyup" },
+  { native: "\u9F20\u6807\u79FB\u5165", normalized: "mouseover" },
+  { native: "\u9F20\u6807\u79FB\u51FA", normalized: "mouseout" },
+  { native: "\u83B7\u5F97\u7126\u70B9", normalized: "focus" },
+  { native: "\u5931\u53BB\u7126\u70B9", normalized: "blur" },
+  { native: "\u52A0\u8F7D", normalized: "load" },
+  { native: "\u6EDA\u52A8", normalized: "scroll" },
+  // Additional references
+  { native: "\u6211\u7684", normalized: "my" },
+  { native: "\u5B83\u7684", normalized: "its" },
+  // Time units
+  { native: "\u79D2", normalized: "s" },
+  { native: "\u6BEB\u79D2", normalized: "ms" },
+  { native: "\u5206\u949F", normalized: "m" },
+  { native: "\u5C0F\u65F6", normalized: "h" },
+  // Logical operators
+  { native: "\u548C", normalized: "and" },
+  { native: "\u6216\u8005", normalized: "or" },
+  { native: "\u6216", normalized: "or" },
+  { native: "\u4E0D", normalized: "not" },
+  { native: "\u975E", normalized: "not" },
+  { native: "\u662F", normalized: "is" },
+  // Additional synonyms not in profile
+  { native: "\u82E5", normalized: "if" },
+  { native: "\u4E0D\u7136", normalized: "else" },
+  { native: "\u5FAA\u73AF", normalized: "repeat" },
+  { native: "\u904D\u5386", normalized: "for" },
+  { native: "\u6BCF\u4E2A", normalized: "for" },
+  { native: "\u4E3A\u6BCF", normalized: "for" },
+  { native: "\u4E2D\u6B62", normalized: "halt" },
+  { native: "\u629B", normalized: "throw" },
+  { native: "\u547C\u53EB", normalized: "call" },
+  { native: "\u56DE", normalized: "return" },
+  { native: "\u811A\u672C", normalized: "js" },
+  { native: "\u901A\u77E5", normalized: "tell" },
+  { native: "\u7F3A\u7701", normalized: "default" },
+  { native: "\u521D\u59CB", normalized: "init" },
+  { native: "\u52A8\u4F5C", normalized: "behavior" },
+  { native: "\u6FC0\u53D1", normalized: "trigger" },
+  { native: "\u5BF9\u7126", normalized: "focus" },
+  { native: "\u6A21\u7CCA", normalized: "blur" },
+  { native: "\u8DF3\u8F6C", normalized: "go" },
+  { native: "\u5BFC\u822A", normalized: "go" },
+  { native: "\u6293\u53D6", normalized: "fetch" },
+  { native: "\u83B7\u53D6\u6570\u636E", normalized: "fetch" },
+  { native: "\u5B89\u5B9A", normalized: "settle" },
+  { native: "\u62FF\u53D6", normalized: "take" },
+  { native: "\u53D6", normalized: "take" },
+  { native: "\u521B\u5EFA", normalized: "make" },
+  { native: "\u514B\u9686", normalized: "clone" },
+  { native: "\u8BB0\u5F55", normalized: "log" },
+  { native: "\u6253\u5370", normalized: "log" },
+  { native: "\u52A8\u753B", normalized: "transition" },
+  // Modifiers
+  { native: "\u5230\u91CC\u9762", normalized: "into" },
+  { native: "\u91CC", normalized: "into" },
+  { native: "\u524D", normalized: "before" },
+  { native: "\u540E", normalized: "after" },
+  { native: "\u90A3\u4E48", normalized: "then" },
+  { native: "\u5B8C", normalized: "end" }
+];
+var CHINESE_TIME_UNITS = [
+  { pattern: "\u6BEB\u79D2", suffix: "ms", length: 2 },
+  { pattern: "\u5206\u949F", suffix: "m", length: 2 },
+  { pattern: "\u5C0F\u65F6", suffix: "h", length: 2 },
+  { pattern: "\u79D2", suffix: "s", length: 1 },
+  { pattern: "\u5206", suffix: "m", length: 1 }
+];
+var ChineseTokenizer = class extends BaseTokenizer {
+  constructor() {
+    super();
+    this.language = "zh";
+    this.direction = "ltr";
+    this.initializeKeywordsFromProfile(chineseProfile, CHINESE_EXTRAS);
+  }
+  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]) || input[pos] === "\u201C" || input[pos] === "\u201D" || input[pos] === "\u2018" || input[pos] === "\u2019") {
+        const stringToken = this.tryChineseString(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.extractChineseNumber(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) {
+        tokens.push(multiParticle);
+        pos = multiParticle.position.end;
+        continue;
+      }
+      if (isChinese(input[pos])) {
+        const wordToken = this.extractChineseWord(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, "zh");
+  }
+  classifyToken(token) {
+    if (PARTICLES.has(token)) return "particle";
+    if (this.isKeyword(token)) return "keyword";
+    if (token.startsWith("#") || token.startsWith(".") || token.startsWith("[") || token.startsWith("<"))
+      return "selector";
+    if (token.startsWith('"') || token.startsWith("'") || token.startsWith("\u201C") || token.startsWith("\u2018"))
+      return "literal";
+    if (/^\d/.test(token)) return "literal";
+    return "identifier";
+  }
+  /**
+   * Extract a Chinese word.
+   * Uses greedy matching to find the longest known keyword.
+   * Chinese doesn't have inflection, so we don't need morphological normalization.
+   * profileKeywords is already sorted longest-first, enabling greedy matching.
+   */
+  extractChineseWord(input, startPos) {
+    for (const entry of this.profileKeywords) {
+      const keyword = entry.native;
+      const candidate = input.slice(startPos, startPos + keyword.length);
+      if (candidate === keyword) {
+        let allChinese = true;
+        for (let i = 0; i < keyword.length; i++) {
+          if (!isChinese(keyword[i])) {
+            allChinese = false;
+            break;
+          }
+        }
+        if (allChinese) {
+          return createToken(
+            candidate,
+            "keyword",
+            createPosition(startPos, startPos + keyword.length),
+            entry.normalized
+          );
+        }
+      }
+    }
+    let pos = startPos;
+    let word = "";
+    while (pos < input.length) {
+      const char = input[pos];
+      if (PARTICLES.has(char) && word.length > 0) {
+        break;
+      }
+      if (isChinese(char)) {
+        word += char;
+        pos++;
+      } else {
+        break;
+      }
+    }
+    if (!word) return null;
+    if (PARTICLES.has(word)) {
+      return createToken(word, "particle", createPosition(startPos, pos));
+    }
+    return createToken(word, "identifier", createPosition(startPos, pos));
+  }
+  /**
+   * Extract an ASCII word (for mixed Chinese/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));
+  }
+  /**
+   * Try to extract a string literal, including Chinese quotes.
+   * Chinese quotes: \u201C " (open) \u201D " (close) \u2018 ' (open) \u2019 ' (close)
+   */
+  tryChineseString(input, pos) {
+    const char = input[pos];
+    if (char === '"' || char === "'" || char === "`") {
+      return this.tryString(input, pos);
+    }
+    if (char === "\u201C") {
+      let endPos = pos + 1;
+      while (endPos < input.length && input[endPos] !== "\u201D") {
+        endPos++;
+      }
+      if (endPos >= input.length) return null;
+      const value = input.slice(pos, endPos + 1);
+      return createToken(value, "literal", createPosition(pos, endPos + 1));
+    }
+    if (char === "\u2018") {
+      let endPos = pos + 1;
+      while (endPos < input.length && input[endPos] !== "\u2019") {
+        endPos++;
+      }
+      if (endPos >= input.length) return null;
+      const value = input.slice(pos, endPos + 1);
+      return createToken(value, "literal", createPosition(pos, endPos + 1));
+    }
+    return null;
+  }
+  /**
+   * Extract a number, including Chinese time unit suffixes.
+   * Chinese time units attach directly without whitespace.
+   */
+  extractChineseNumber(input, startPos) {
+    return this.tryNumberWithTimeUnits(input, startPos, CHINESE_TIME_UNITS, {
+      allowSign: false,
+      skipWhitespace: false
+    });
+  }
+};
+var chineseTokenizer = new ChineseTokenizer();
+
+// src/languages/zh.ts
+registerLanguage("zh", chineseTokenizer, chineseProfile);
+export {
+  chineseProfile,
+  chineseTokenizer
+};
+//# sourceMappingURL=zh.js.map