utils, test: helper for extracting http[s] URLs from text

When sharing URLs through KDE Connect, the incoming "url" usually contains some
additional description that was added by the source application. Use regext to
pick up any valid http[s] URLs that may be found in the text.

meson.build

@@ -84,6 +84,20 @@ test_mconn_crypt = executable('test-mconn-crypt', test_mconn_crypt_src,
                               install: false)
 test('mconn-crypt', test_mconn_crypt)
 
+test_mconn_utils_src = [
+  'test/mconn-utils-test.vala',
+  'src/mconnect/utils.vala',
+  'src/mconnect/logging.vala',
+]
+test_mconn_utils = executable('test-mconn-utils', test_mconn_utils_src,
+                              dependencies: [
+                                glib_dep, gobject_dep,
+                                gio_dep, gio_unix_dep,
+                                posix_dep,
+                              ],
+                              install: false)
+test('mconn-utils', test_mconn_utils)
+
 # other files
 applicationsdir = join_paths(get_option('datadir'),
                              'applications')

src/mconnect/utils.vala

@@ -149,4 +149,41 @@ using Posix;
 			});
 		return tls_conn;
 	}
+
+	/**
+	 * find_urls:
+	 *
+	 * Locate and extract URL like patterns in the text. URLs are assumed to
+	 * start with http or https.
+	 *
+	 * @text: input test
+	 * @return array of matches, if there were none then array if of length 0
+	 */
+	string[] find_urls(string text) {
+		try {
+		// regex taken from SO
+			Regex r = /https?:\/\/(www\.)?[-a-zA-Z0-9@:%._\+~#=]{2,256}\.[a-z]{2,6}\b([-a-zA-Z0-9@:%_\+.~#?&\/=]*)/;
+
+			MatchInfo mi;
+
+			string[] matches = {};
+
+			if (r.match(text, RegexMatchFlags.NOTEMPTY, out mi)) {
+				while (mi.matches()) {
+					if (mi.is_partial_match() == false) {
+						var m = mi.fetch(0);
+						debug("found match %s", m);
+						matches += m;
+					}
+					mi.next();
+				}
+			} else {
+				debug("no match");
+			}
+			return matches;
+		} catch (RegexError e) {
+			warning("failed to compile regex: %s", e.message);
+			return null;
+		}
+	}
 }

test/mconn-utils-test.vala

@@ -0,0 +1,46 @@
+void test_find_urls_simple() {
+	var urls = Utils.find_urls("https://en.m.wikipedia.org/wiki/Isle_of_Man via DuckDuckGo for Android");
+
+	assert(urls != null);
+	assert(urls.length == 1);
+
+	assert(urls[0] == "https://en.m.wikipedia.org/wiki/Isle_of_Man");
+}
+
+void test_find_urls_extract() {
+	var urls = Utils.find_urls("Foo bar baz?\n\nhttp://foo.bar.com/123/345/abcd\n\nShared from my Google cards");
+
+	assert(urls != null);
+	assert(urls.length == 1);
+
+	assert(urls[0] == "http://foo.bar.com/123/345/abcd");
+}
+
+void test_find_urls_many() {
+	var urls = Utils.find_urls("https://foo.bar.com http://google.biz http://www.funny.io");
+
+	assert(urls != null);
+	assert(urls.length == 3);
+
+	assert(urls[0] == "https://foo.bar.com");
+	assert(urls[1] == "http://google.biz");
+	assert(urls[2] == "http://www.funny.io");
+}
+
+void test_find_urls_none() {
+	var urls = Utils.find_urls("baz bar \nbar.com foo ");
+
+	assert(urls != null);
+	assert(urls.length == 0);
+}
+
+
+public static void main(string[] args) {
+	Test.init(ref args);
+
+	Test.add_func("/mconn-utils/find-urls/simple", test_find_urls_simple);
+	Test.add_func("/mconn-utils/find-urls/extract", test_find_urls_extract);
+	Test.add_func("/mconn-utils/find-urls/many", test_find_urls_many);
+	Test.add_func("/mconn-utils/find-urls/none", test_find_urls_none);
+	Test.run();
+}