@lokascript/semantic 1.0.0 → 1.1.0

package/dist/languages/tr.js

@@ -1128,10 +1128,38 @@ var turkishProfile = {
     }
   },
   roleMarkers: {
-    patient: { primary: "i", alternatives: ["\u0131", "u", "\xFC"], position: "after" },
-    // Accusative
-    destination: { primary: "e", alternatives: ["a", "de", "da", "te", "ta"], position: "after" },
-    // Dative/Locative
+    patient: {
+      primary: "i",
+      alternatives: ["\u0131", "u", "\xFC", "yi", "y\u0131", "yu", "y\xFC", "ni", "n\u0131", "nu", "n\xFC"],
+      position: "after"
+    },
+    // Accusative (with buffer consonants y/n)
+    destination: {
+      primary: "e",
+      // Include both dative (e/a) and genitive (ın/in/un/ün) for possessive patterns
+      // Genitive is used in "X's Y" patterns: #button ın .active = "#button's .active"
+      alternatives: [
+        "a",
+        "ye",
+        "ya",
+        "ne",
+        "na",
+        "de",
+        "da",
+        "te",
+        "ta",
+        "\u0131n",
+        "in",
+        "un",
+        "\xFCn",
+        "n\u0131n",
+        "nin",
+        "nun",
+        "n\xFCn"
+      ],
+      position: "after"
+    },
+    // Dative/Locative + Genitive (with buffer consonants)
     source: { primary: "den", alternatives: ["dan", "ten", "tan"], position: "after" },
     // Ablative
     style: { primary: "le", alternatives: ["la", "yle", "yla"], position: "after" },
@@ -1147,10 +1175,12 @@ var turkishProfile = {
     // Content operations
     put: { primary: "koy", normalized: "put" },
     append: { primary: "ekle", normalized: "append" },
-    take: { primary: "al", normalized: "take" },
+    take: { primary: "tut", normalized: "take" },
+    // al removed to avoid collision with get
     make: { primary: "yap", normalized: "make" },
     clone: { primary: "kopyala", normalized: "clone" },
-    swap: { primary: "de\u011Fi\u015Ftir", alternatives: ["takas"], normalized: "swap" },
+    swap: { primary: "takas", normalized: "swap" },
+    // Removed değiştir alternative to avoid collision with toggle
     morph: { primary: "d\xF6n\xFC\u015Ft\xFCr", alternatives: ["\u015Fekil de\u011Fi\u015Ftir"], normalized: "morph" },
     // Variable operations
     set: { primary: "ayarla", alternatives: ["yap", "belirle"], normalized: "set" },
@@ -1167,8 +1197,14 @@ var turkishProfile = {
     trigger: { primary: "tetikle", normalized: "trigger" },
     send: { primary: "g\xF6nder", normalized: "send" },
     // DOM focus
-    focus: { primary: "odak", normalized: "focus" },
-    blur: { primary: "bulan\u0131k", normalized: "blur" },
+    focus: { primary: "odak", alternatives: ["odaklanma"], normalized: "focus" },
+    blur: { primary: "bulan\u0131k", alternatives: ["bulan\u0131kl\u0131k"], normalized: "blur" },
+    // Common event names (for event handler patterns)
+    click: { primary: "t\u0131klama", alternatives: ["t\u0131kla"], normalized: "click" },
+    hover: { primary: "\xFCzerine gelme", alternatives: ["\xFCzerinde gezinme"], normalized: "hover" },
+    submit: { primary: "g\xF6nderme", alternatives: ["g\xF6nder"], normalized: "submit" },
+    input: { primary: "giri\u015F", alternatives: ["girdi"], normalized: "input" },
+    change: { primary: "de\u011Fi\u015Fiklik", alternatives: ["de\u011Fi\u015Fim"], normalized: "change" },
     // Navigation
     go: { primary: "git", normalized: "go" },
     // Async
@@ -1207,7 +1243,16 @@ var turkishProfile = {
     // Event modifiers (for repeat until event)
     until: { primary: "kadar", normalized: "until" },
     event: { primary: "olay", normalized: "event" },
-    from: { primary: "-den", alternatives: ["-dan"], normalized: "from" }
+    from: { primary: "den", alternatives: ["dan"], normalized: "from" }
+  },
+  eventHandler: {
+    // Event marker: da/de/ta/te (locative case suffix with vowel harmony), used in SOV pattern
+    // Pattern: [event] da [patient] i [action]
+    // Example: tıklama da .active i değiştir
+    // Note: Vowel harmony variants (da/de/ta/te) should be handled by vowel harmony expansion
+    eventMarker: { primary: "da", alternatives: ["de", "ta", "te"], position: "after" },
+    temporalMarkers: ["d\u0131\u011F\u0131nda", "di\u011Finde"]
+    // temporal converb suffixes (when)
   }
 };
 
@@ -1321,15 +1366,9 @@ var TURKISH_EXTRAS = [
   { native: "tus_bas", normalized: "keydown" },
   { native: "tu\u015F_b\u0131rak", normalized: "keyup" },
   { native: "tus_birak", normalized: "keyup" },
-  // References
-  { native: "ben", normalized: "me" },
+  // References (possessive forms not in profile)
   { native: "benim", normalized: "my" },
-  { native: "o", normalized: "it" },
   { native: "onun", normalized: "its" },
-  { native: "sonu\xE7", normalized: "result" },
-  { native: "sonuc", normalized: "result" },
-  { native: "olay", normalized: "event" },
-  { native: "hedef", normalized: "target" },
   // Time units
   { native: "saniye", normalized: "s" },
   { native: "milisaniye", normalized: "ms" },
@@ -1339,59 +1378,9 @@ var TURKISH_EXTRAS = [
   { native: "ve", normalized: "and" },
   { native: "veya", normalized: "or" },
   { native: "de\u011Fil", normalized: "not" },
-  { native: "degil", normalized: "not" },
-  // Event triggers (on)
-  { native: "\xFCzerinde", normalized: "on" },
-  { native: "uzerinde", normalized: "on" },
-  { native: "oldu\u011Funda", normalized: "on" },
-  { native: "oldugunda", normalized: "on" },
-  // Command overrides (ensure correct mapping when profile has multiple meanings)
-  { native: "ekle", normalized: "add" },
-  // Profile may have this as 'append'
-  { native: "de\u011Fi\u015Ftir", normalized: "toggle" },
-  // Profile has this as 'swap'
-  // Diacritic-free variants of commands
-  { native: "de\u011Fistir", normalized: "toggle" },
-  { native: "kaldir", normalized: "remove" },
-  { native: "yerlestir", normalized: "put" },
-  { native: "olustur", normalized: "make" },
-  { native: "artir", normalized: "increment" },
-  { native: "yazdir", normalized: "log" },
-  { native: "goster", normalized: "show" },
-  { native: "gecis", normalized: "transition" },
-  { native: "atesle", normalized: "trigger" },
-  { native: "gonder", normalized: "send" },
-  { native: "bulaniklastir", normalized: "blur" },
-  { native: "odak_kaldir", normalized: "blur" },
-  { native: "yonlendir", normalized: "go" },
-  { native: "cek", normalized: "fetch" },
-  { native: "yerles", normalized: "settle" },
-  { native: "eger", normalized: "if" },
-  { native: "degilse", normalized: "else" },
-  { native: "firlat", normalized: "throw" },
-  { native: "cagir", normalized: "call" },
-  { native: "don", normalized: "return" },
-  { native: "dondur", normalized: "return" },
-  { native: "eszamansiz", normalized: "async" },
-  { native: "soyle", normalized: "tell" },
-  { native: "varsayilan", normalized: "default" },
-  { native: "baslat", normalized: "init" },
-  { native: "basla", normalized: "init" },
-  { native: "davranis", normalized: "behavior" },
-  { native: "yukle", normalized: "install" },
-  { native: "olc", normalized: "measure" },
-  { native: "icine", normalized: "into" },
-  { native: "once", normalized: "before" },
-  { native: "icin", normalized: "for" },
-  // Colloquial forms
-  { native: "al", normalized: "get" },
-  { native: "yap", normalized: "set" },
-  // Control flow helpers
-  { native: "o_zaman", normalized: "then" },
-  { native: "bitir", normalized: "end" },
-  // Case suffix modifiers
-  { native: "-den", normalized: "from" },
-  { native: "-dan", normalized: "from" }
+  { native: "degil", normalized: "not" }
+  // Note: Command synonyms and diacritic-free variants should be in profile alternatives.
+  // Event triggers (üzerinde, olduğunda) should be in profile as 'on' alternatives.
 ];
 var TURKISH_TIME_UNITS = [
   { pattern: "milisaniye", suffix: "ms", length: 10, caseInsensitive: true },
@@ -1463,6 +1452,12 @@ var TurkishTokenizer = class extends BaseTokenizer {
         pos = varToken.position.end;
         continue;
       }
+      const phraseToken = this.tryMultiWordPhrase(input, pos);
+      if (phraseToken) {
+        tokens.push(phraseToken);
+        pos = phraseToken.position.end;
+        continue;
+      }
       if (isTurkishLetter2(input[pos])) {
         const wordToken = this.extractTurkishWord(input, pos);
         if (wordToken) {
@@ -1485,6 +1480,32 @@ var TurkishTokenizer = class extends BaseTokenizer {
     if (/^\d/.test(token)) return "literal";
     return "identifier";
   }
+  /**
+   * Try to match multi-word phrases that function as single units.
+   * Multi-word phrases are included in profileKeywords and sorted longest-first,
+   * so they'll be matched before their constituent words.
+   *
+   * Examples: "üzerine gelme" (hover), "fare üzerinde" (mouseover)
+   */
+  tryMultiWordPhrase(input, pos) {
+    for (const entry of this.profileKeywords) {
+      if (!entry.native.includes(" ")) continue;
+      const phrase = entry.native;
+      const candidate = input.slice(pos, pos + phrase.length).toLowerCase();
+      if (candidate === phrase.toLowerCase()) {
+        const nextPos = pos + phrase.length;
+        if (nextPos >= input.length || isWhitespace(input[nextPos]) || !isTurkishLetter2(input[nextPos])) {
+          return createToken(
+            input.slice(pos, pos + phrase.length),
+            "keyword",
+            createPosition(pos, nextPos),
+            entry.normalized
+          );
+        }
+      }
+    }
+    return null;
+  }
   /**
    * Extract a Turkish word.
    * Uses morphological normalization to handle verb conjugations.
@@ -1497,13 +1518,16 @@ var TurkishTokenizer = class extends BaseTokenizer {
     }
     if (!word) return null;
     const lowerWord = word.toLowerCase();
-    const keywordEntry = this.lookupKeyword(lowerWord);
-    if (keywordEntry) {
-      return createToken(word, "keyword", createPosition(startPos, pos), keywordEntry.normalized);
+    if (CASE_SUFFIXES.has(lowerWord)) {
+      return createToken(word, "particle", createPosition(startPos, pos));
     }
     if (POSTPOSITIONS.has(lowerWord)) {
       return createToken(word, "particle", createPosition(startPos, pos));
     }
+    const keywordEntry = this.lookupKeyword(lowerWord);
+    if (keywordEntry) {
+      return createToken(word, "keyword", createPosition(startPos, pos), keywordEntry.normalized);
+    }
     const morphToken = this.tryMorphKeywordMatch(lowerWord, startPos, pos);
     if (morphToken) return morphToken;
     return createToken(word, "identifier", createPosition(startPos, pos));