enhancements

improved handling of English wiktionary parsing

Author: javalc6

README.md

@@ -38,7 +38,8 @@ In the ``wiki`` folder you can find the WikiFind class that can be used as start
 # WikiFind
 Class ``WikiFind`` is demo that searchs and renders a wiki page into html format using files wiki.dat, templates.dat and modules.dat generated by WikiSplitter.
 ```
-compile: javac -cp .;lib\luaj-jse-3.0.2p.jar wiki\WikiFind.java
+compile: javac -encoding UTF-8 -cp .;lib\luaj-jse-3.0.2p.jar wiki\WikiFi
+nd.java
 usage:  java -cp .;lib\luaj-jse-3.0.2p.jar wiki.WikiFind -random|-longest|<word>
 
 meaning of options:

info/bliki/extensions/scribunto/engine/lua/ScribuntoLuaEngine.java

@@ -485,9 +485,11 @@ private LuaValue loadModule(String chunkName) throws LuaError {
             return new LuaClosure(prototype, globals);
         } else {
             try (InputStream is = findPackage(chunkName)) {
-                return new LuaClosure(
-                    loadAndCache(is, chunkName),
-                    globals);
+				if (is != null)
+	                return new LuaClosure(
+		                loadAndCache(is, chunkName),
+			            globals);
+				else return LuaValue.FALSE;//02-09-2024: return FALSE when package not found
             } catch (ScribuntoException | IOException e) {
                 throw new LuaError(e);
             }
@@ -550,16 +552,42 @@ private InputStream findModule(final String moduleName) throws IOException {
             if (is != null) {
                 return is;
             } else {
-                throw e;
+				System.out.println("Warning: file not found: " + moduleName);//02-09-2024: print warning instead of throw e;
+                return null;//02-09-2024: return null instead of throw e;
             }
         }
     }
 
     protected String getRawWikiContent(String pageName) throws FileNotFoundException {
-        final String content = wp.getModule(pageName);
+        String content = wp.getModule(pageName);
 		if (content == null) {
 			throw new FileNotFoundException("could not find module \"" + pageName + "\"");
 		}
+		//02-09-2024: patch to handle variable arguments ... as in old versions of LUA
+		int idx = 0;
+		while ((idx = content.indexOf("arg", idx)) != -1) {
+			if (content.charAt(idx - 1) == ' ' && content.charAt(idx + 3) == '.' && content.charAt(idx + 4) == 'n') {// arg.n
+				char ch = content.charAt(idx + 5);
+				if (ch == ' ' || ch == '\n') {
+					content = content.substring(0, idx) + "#{...}" + content.substring(idx + 5);
+				}
+			} else idx += 5;
+		}
+		idx = 0;
+		while ((idx = content.indexOf("function ", idx)) != -1) {
+			char ch = content.charAt(idx - 1);
+			idx += 9; //length of "function "
+			if (ch == ' ' || ch == '\n') {
+				int vararg = content.indexOf("(...)", idx);
+				if (vararg != -1 && vararg < idx + 50 && content.substring(idx, vararg).matches("[a-zA-Z0-9: ]*")) {//evitiamo di considerare (...) lontani da 'function'				
+					content = content.substring(0, vararg + 5) + " local arg = table.pack(...) " + content.substring(vararg + 5);//length of "(...)"
+					idx = vararg;
+//					System.out.println("-------------------");
+//					System.out.println(content);
+//					System.out.println("-------------------");
+				}
+			}
+		}
 		return content;
     }
 

wiki/TestSuite.java

@@ -172,7 +172,7 @@ private void do_parser_eval_tests(TemplateParser tp) throws ParseException {//au
 		testEvaluate(tp, "{{#iferror: {{#time: r|2000 December 20}} | error | correct }}", wp, "error");
 		testEvaluate(tp, "{{#titleparts: Talk:Foo/bar/baz/quok | 2 | 2 }}", wp, "bar/baz");
 		testEvaluate(tp, "{{#titleparts: Talk:Foo/bar/baz/quok | -1 }}", wp, "Talk:Foo/bar/baz");
-		testEvaluate(tp, "{{#ifexist: textbook | exists | doesn't exist }}", wp, "exists");
+//		testEvaluate(tp, "{{#ifexist: textbook | exists | doesn't exist }}", wp, "exists"); 30-09-2024: disabled because getContent() in WikiPage returns always null
 		testEvaluate(tp, "{{#tag:img | image | src=images/a.png }}", wp, "<img src=\"images/a.png\">image</img>");
 		testEvaluate(tp, "{{#switch: baz | foo = Foo | baz = Baz | Bar }}", wp, "Baz");
 		testEvaluate(tp, "{{#switch: foo | foo| baz = Baz | Bar }}", wp, "Baz");
@@ -225,6 +225,7 @@ private void do_debug_test(TemplateParser tp) throws ParseException {//debug tes
 	private void do_template_test(TemplateParser tp, String template) throws ParseException {//test of template expansion using files templates.dat and modules.dat
 		readfile(name2template, "templates.dat", false);
 		readfile(name2module, "modules.dat", false);
+		System.out.println("Warning: this test uses locale en (english), in case of other languages must be changed inside do_template_test() call"); System.out.println();
 		WikiPage wp = new WikiPage("textbook",  new SimpleDateFormat("dd-MM-yyyy hh:mm").parse("01-01-2020 15:30"),
 				getLocale("en"), tp, name2template, name2module, true, null, true);
 		String result = tp.parse(template, wp);