@lokascript/semantic 1.3.0 → 2.0.0

package/dist/languages/sw.d.ts

@@ -1,35 +1,9 @@
-import { B as BaseTokenizer, T as TokenStream, a as TokenKind, L as LanguageProfile } from '../types-BY3Id07j.js';
-
 /**
- * Swahili Tokenizer
+ * Swahili Language Module
  *
- * Tokenizes Swahili (Kiswahili) hyperscript input.
- * Swahili characteristics:
- * - SVO word order
- * - Agglutinative morphology
- * - Noun class prefixes (m-, wa-, ki-, vi-, etc.)
- * - Verb prefixes for subject/object agreement
- * - No grammatical gender, but noun classes
+ * Self-registering module for Swahili language support.
+ * Importing this module registers Swahili tokenizer and profile.
  */
-
-declare class SwahiliTokenizer extends BaseTokenizer {
-    readonly language = "sw";
-    readonly direction: "ltr";
-    constructor();
-    tokenize(input: string): TokenStream;
-    classifyToken(token: string): TokenKind;
-    private extractWord;
-    private extractNumber;
-}
-declare const swahiliTokenizer: SwahiliTokenizer;
-
-/**
- * Swahili Language Profile
- *
- * SVO word order, prepositions, space-separated, agglutinative.
- * Features noun class system (18 classes) and verb agreement prefixes.
- */
-
-declare const swahiliProfile: LanguageProfile;
-
-export { swahiliProfile, swahiliTokenizer };
+export { swahiliTokenizer } from '../tokenizers/swahili';
+export { swahiliProfile } from '../generators/profiles/swahili';
+//# sourceMappingURL=sw.d.ts.map

package/dist/languages/sw.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"sw.d.ts","sourceRoot":"","sources":["../../src/languages/sw.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAMH,OAAO,EAAE,gBAAgB,EAAE,MAAM,uBAAuB,CAAC;AACzD,OAAO,EAAE,cAAc,EAAE,MAAM,gCAAgC,CAAC"}

package/dist/languages/sw.js

@@ -8,7 +8,7 @@ function registerLanguage(code, tokenizer, profile) {
   patternCache.delete(code);
 }
 
-// src/tokenizers/base.ts
+// src/tokenizers/token-utils.ts
 var TokenStreamImpl = class {
   constructor(tokens, language) {
     this.pos = 0;
@@ -74,17 +74,14 @@ function createToken(value, kind, position, 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,
+      ...normalized !== void 0 && { normalized },
+      ...stem !== void 0 && { stem },
+      ...stemConfidence !== void 0 && { stemConfidence }
+    };
   }
   return { value, kind, position };
 }
@@ -106,11 +103,15 @@ function isAsciiLetter(char) {
 function isAsciiIdentifierChar(char) {
   return /[a-zA-Z0-9_-]/.test(char);
 }
+
+// src/tokenizers/char-classifiers.ts
 function createLatinCharClassifiers(letterPattern) {
   const isLetter = (char) => letterPattern.test(char);
   const isIdentifierChar = (char) => isLetter(char) || /[0-9_-]/.test(char);
   return { isLetter, isIdentifierChar };
 }
+
+// src/tokenizers/extractors.ts
 function extractCssSelector(input, startPos) {
   if (startPos >= input.length) return null;
   const char = input[startPos];
@@ -368,6 +369,8 @@ function extractNumber(input, startPos) {
   }
   return number;
 }
+
+// src/tokenizers/base-tokenizer.ts
 var _BaseTokenizer = class _BaseTokenizer {
   constructor() {
     /** Keywords derived from profile, sorted longest-first for greedy matching */
@@ -375,6 +378,34 @@ var _BaseTokenizer = class _BaseTokenizer {
     /** Map for O(1) keyword lookups by lowercase native word */
     this.profileKeywordMap = /* @__PURE__ */ new Map();
   }
+  /**
+   * Check if current position is a property access (obj.prop) vs CSS selector (.active).
+   * Property access: no whitespace before '.', previous token is identifier/keyword/selector.
+   * Also detects standalone method calls: .identifier( pattern.
+   *
+   * Returns true if '.' was emitted as an operator token and pos should advance by 1.
+   * Returns false if this is a CSS selector and should be handled by trySelector().
+   */
+  tryPropertyAccess(input, pos, tokens) {
+    if (input[pos] !== ".") return false;
+    const lastToken = tokens[tokens.length - 1];
+    const hasWhitespaceBefore = lastToken && lastToken.position.end < pos;
+    const isPropertyAccess = lastToken && !hasWhitespaceBefore && (lastToken.kind === "identifier" || lastToken.kind === "keyword" || lastToken.kind === "selector");
+    if (isPropertyAccess) {
+      tokens.push(createToken(".", "operator", createPosition(pos, pos + 1)));
+      return true;
+    }
+    const methodStart = pos + 1;
+    let methodEnd = methodStart;
+    while (methodEnd < input.length && isAsciiIdentifierChar(input[methodEnd])) {
+      methodEnd++;
+    }
+    if (methodEnd < input.length && input[methodEnd] === "(") {
+      tokens.push(createToken(".", "operator", createPosition(pos, pos + 1)));
+      return true;
+    }
+    return false;
+  }
   /**
    * Initialize keyword mappings from a language profile.
    * Builds a list of native→english mappings from:
@@ -410,6 +441,11 @@ var _BaseTokenizer = class _BaseTokenizer {
       for (const [normalized, native] of Object.entries(profile.references)) {
         keywordMap.set(native, { native, normalized });
       }
+      for (const canonical of Object.keys(profile.references)) {
+        if (!keywordMap.has(canonical)) {
+          keywordMap.set(canonical, { native: canonical, normalized: canonical });
+        }
+      }
     }
     if (profile.roleMarkers) {
       for (const [role, marker] of Object.entries(profile.roleMarkers)) {
@@ -855,9 +891,9 @@ var swahiliProfile = {
     clone: { primary: "nakili", alternatives: ["rudufu"], normalized: "clone" },
     swap: { primary: "badilisha", alternatives: ["badalisha"], normalized: "swap" },
     morph: { primary: "geuza", alternatives: ["badilisha umbo"], normalized: "morph" },
-    set: { primary: "weka", alternatives: ["seti"], normalized: "set" },
+    set: { primary: "seti", alternatives: ["weka"], normalized: "set" },
     get: { primary: "pata", alternatives: ["pokea"], normalized: "get" },
-    increment: { primary: "ongeza", normalized: "increment" },
+    increment: { primary: "ongezeko", alternatives: ["ongeza"], normalized: "increment" },
     decrement: { primary: "punguza", normalized: "decrement" },
     log: { primary: "andika", alternatives: ["rekodi"], normalized: "log" },
     show: { primary: "onyesha", normalized: "show" },
@@ -893,8 +929,14 @@ var swahiliProfile = {
     default: { primary: "chaguo-msingi", normalized: "default" },
     init: { primary: "anzisha", alternatives: ["anza"], normalized: "init" },
     behavior: { primary: "tabia", normalized: "behavior" },
-    install: { primary: "weka", alternatives: ["sakinisha"], normalized: "install" },
+    install: { primary: "sakinisha", alternatives: ["weka"], normalized: "install" },
     measure: { primary: "pima", normalized: "measure" },
+    beep: { primary: "lia", normalized: "beep" },
+    break: { primary: "vunja", normalized: "break" },
+    copy: { primary: "nakili", normalized: "copy" },
+    exit: { primary: "toka", normalized: "exit" },
+    pick: { primary: "chagua", normalized: "pick" },
+    render: { primary: "chora", normalized: "render" },
     into: { primary: "ndani", normalized: "into" },
     before: { primary: "kabla", normalized: "before" },
     after: { primary: "baada", normalized: "after" },
@@ -906,7 +948,12 @@ var swahiliProfile = {
   eventHandler: {
     keyword: { primary: "wakati", alternatives: ["kwenye", "kwa"] },
     sourceMarker: { primary: "kutoka", position: "before" },
-    conditionalKeyword: { primary: "unapo", alternatives: ["anapo", "tunapo", "mnapo", "wanapo"] }
+    conditionalKeyword: { primary: "unapo", alternatives: ["anapo", "tunapo", "mnapo", "wanapo"] },
+    eventMarker: {
+      primary: "wakati",
+      alternatives: ["kwenye", "kwa", "unapo"],
+      position: "before"
+    }
   }
 };
 
@@ -1018,6 +1065,10 @@ var SwahiliTokenizer = class extends BaseTokenizer {
           pos = modifierToken.position.end;
           continue;
         }
+        if (this.tryPropertyAccess(input, pos, tokens)) {
+          pos++;
+          continue;
+        }
         const selectorToken = this.trySelector(input, pos);
         if (selectorToken) {
           tokens.push(selectorToken);
@@ -1077,7 +1128,8 @@ var SwahiliTokenizer = class extends BaseTokenizer {
     const lower = token.toLowerCase();
     if (PREPOSITIONS.has(lower)) return "particle";
     if (this.isKeyword(lower)) return "keyword";
-    if (token.startsWith("#") || token.startsWith(".") || token.startsWith("[")) return "selector";
+    if (token.startsWith("#") || token.startsWith(".") || token.startsWith("[") || token.startsWith("*") || token.startsWith("<"))
+      return "selector";
     if (token.startsWith('"') || token.startsWith("'")) return "literal";
     if (/^\d/.test(token)) return "literal";
     return "identifier";