Fixed issue with illegal URLs containing spaces being returned by Sea…

…rXNG (#42)

web-search-engines/langchain4j-community-web-search-engine-searxng/src/main/java/dev/langchain4j/community/web/search/searxng/SearXNGWebSearchEngine.java

@@ -49,6 +49,13 @@ private static String toCSV(List<?> values) {
         return String.join(",", values.toString());
     }
 
+    static URI makeURI(String urlString) {
+        if (urlString == null || urlString.isBlank()) {
+            throw new IllegalArgumentException("urlString can not be null or blank");
+        }
+        return URI.create(urlString.replaceAll("\\s+", "%20"));
+    }
+
     private static Map<String, String> extractMetadata(SearXNGResult result) {
         final Map<String, String> metadata = new HashMap<>();
         metadata.put("engine", result.getEngine());
@@ -60,13 +67,8 @@ private static Map<String, String> extractMetadata(SearXNGResult result) {
     }
 
     private static WebSearchOrganicResult toWebSearchOrganicResult(SearXNGResult result) {
-        // FIXME: temporarily fix URI illegal character, raise a issue to solve it.
-        String url = result.getUrl();
-        int illegalChar = url.indexOf('#');
-        url = illegalChar == -1 ? url : url.substring(0, illegalChar);
-
         return WebSearchOrganicResult.from(
-                result.getTitle(), URI.create(url), result.getContent(), null, extractMetadata(result));
+                result.getTitle(), makeURI(result.getUrl()), result.getContent(), null, extractMetadata(result));
     }
 
     private static boolean hasValue(String value) {

web-search-engines/langchain4j-community-web-search-engine-searxng/src/test/java/dev/langchain4j/community/web/search/searxng/SearXNGWebSearchEngineTest.java

@@ -0,0 +1,55 @@
+package dev.langchain4j.community.web.search.searxng;
+
+import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertNotEquals;
+import static org.junit.Assert.assertNotNull;
+import static org.junit.jupiter.api.Assertions.assertDoesNotThrow;
+import static org.junit.jupiter.api.Assertions.assertThrows;
+
+import java.net.URI;
+import org.junit.jupiter.api.Test;
+
+class SearXNGWebSearchEngineTest {
+
+    private static void testURI(String uriString) {
+        assertDoesNotThrow(() -> SearXNGWebSearchEngine.makeURI(uriString));
+        final URI uri = SearXNGWebSearchEngine.makeURI(uriString);
+        assertNotNull(uri);
+        if (uriString.matches(".*\\s+.*")) {
+            assertNotEquals(uriString, uri.toString());
+        } else {
+            assertEquals(uriString, uri.toString());
+        }
+    }
+
+    @Test
+    void test_malformed_urls() {
+        assertThrows(
+                IllegalArgumentException.class,
+                () -> URI.create(
+                        "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j is a groundbreaking Java,by Python and JavaScript libraries."));
+        assertDoesNotThrow(
+                () -> URI.create(
+                        "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j"));
+        assertDoesNotThrow(() -> URI.create(
+                "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j is a groundbreaking Java,by Python and JavaScript libraries."
+                        .replaceAll(" ", "%20")));
+        assertDoesNotThrow(() -> URI.create(
+                "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j is a groundbreaking Java,by Python and JavaScript libraries."
+                        .replaceAll("\\s+", "%20")));
+        assertDoesNotThrow(() -> URI.create(
+                "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f"));
+    }
+
+    @Test
+    void test_make_uri() {
+        assertThrows(IllegalArgumentException.class, () -> SearXNGWebSearchEngine.makeURI(null));
+        assertThrows(IllegalArgumentException.class, () -> SearXNGWebSearchEngine.makeURI(""));
+        assertThrows(IllegalArgumentException.class, () -> SearXNGWebSearchEngine.makeURI("  \\t"));
+        testURI(
+                "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j");
+        testURI(
+                "https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f#:~:text=LangChain4j is a groundbreaking Java,by Python and JavaScript libraries.");
+        testURI("https://www.linkedin.com/pulse/introduction-langchain4j-supercharging-java-llms-ibrahim-jimoh-kyj4f");
+    }
+}