Fix validation failures of URIs with multiple hash signs.

Author: MortimerGoro

app/src/common/shared/org/mozilla/vrbrowser/telemetry/GleanMetricsService.java

@@ -26,6 +26,7 @@
 import org.mozilla.vrbrowser.utils.UrlUtils;
 
 import java.net.URI;
+import java.net.URISyntaxException;
 import java.util.HashMap;
 import java.util.HashSet;
 import java.util.Hashtable;
@@ -108,7 +109,7 @@ public static void stopPageLoadTimeWithURI(String uri) {
         }
 
         try {
-            URI uriLink = URI.create(uri);
+            URI uriLink = UrlUtils.parseUri(uri);
             if (uriLink.getHost() == null) {
                 return;
             }
@@ -117,7 +118,7 @@ public static void stopPageLoadTimeWithURI(String uri) {
                 Url.INSTANCE.domains().add();
             }
             Url.INSTANCE.visits().add();
-        } catch (IllegalArgumentException e) {
+        } catch (URISyntaxException e) {
             Log.e(LOGTAG, "Invalid URL", e);
         }
     }

app/src/common/shared/org/mozilla/vrbrowser/telemetry/TelemetryWrapper.java

@@ -21,6 +21,7 @@
 import org.mozilla.telemetry.schedule.jobscheduler.JobSchedulerTelemetryScheduler;
 import org.mozilla.telemetry.serialize.JSONPingSerializer;
 import org.mozilla.telemetry.storage.FileTelemetryStorage;
+import org.mozilla.thirdparty.com.google.android.exoplayer2.util.UriUtil;
 import org.mozilla.vrbrowser.BuildConfig;
 import org.mozilla.vrbrowser.R;
 import org.mozilla.vrbrowser.browser.SettingsStore;
@@ -31,6 +32,7 @@
 import org.mozilla.vrbrowser.utils.UrlUtils;
 
 import java.net.URI;
+import java.net.URISyntaxException;
 import java.util.HashMap;
 import java.util.HashSet;
 import java.util.Map;
@@ -307,7 +309,7 @@ public static void uploadPageLoadToHistogram(String uri) {
         }
 
         try {
-            URI uriLink = URI.create(uri);
+            URI uriLink = UrlUtils.parseUri(uri);
             if (uriLink.getHost() == null) {
                 return;
             }
@@ -332,7 +334,7 @@ public static void uploadPageLoadToHistogram(String uri) {
 
             loadingTimeHistogram[histogramLoadIndex]++;
 
-        } catch (IllegalArgumentException e) {
+        } catch (URISyntaxException e) {
             Log.e(LOGTAG, "Invalid URL", e);
         }
     }

app/src/common/shared/org/mozilla/vrbrowser/utils/StringUtils.java

@@ -76,6 +76,10 @@ public static boolean contains(String[] aTarget, String aText) {
         return false;
     }
 
+    public static long charCount(@NonNull String aString, char target) {
+        return aString.chars().filter(ch -> ch == target).count();
+    }
+
     /**
      * The version code is composed like: yDDDHHmm
      *  * y   = Double digit year, with 16 substracted: 2017 -> 17 -> 1

app/src/common/shared/org/mozilla/vrbrowser/utils/UrlUtils.java

@@ -22,6 +22,7 @@
 import java.io.File;
 import java.net.MalformedURLException;
 import java.net.URI;
+import java.net.URISyntaxException;
 import java.net.URL;
 import java.util.regex.Pattern;
 
@@ -139,14 +140,14 @@ public static String titleBarUrl(@Nullable String aUri) {
 
             } else {
                 try {
-                    URI uri = URI.create(aUri);
+                    URI uri = parseUri(aUri);
                     URL url = new URL(
                             uri.getScheme() != null ? uri.getScheme() : "",
                             uri.getAuthority() != null ? uri.getAuthority() : "",
                             "");
                     return url.toString();
 
-                } catch (MalformedURLException | IllegalArgumentException e) {
+                } catch (MalformedURLException | URISyntaxException e) {
                     return "";
                 }
             }
@@ -202,6 +203,21 @@ public static String getHost(String uri) {
         }
     }
 
+    public static URI parseUri(String aUri) throws URISyntaxException {
+        try {
+            return new URI(aUri);
+        } catch (URISyntaxException e) {
+            if (!StringUtils.isEmpty(aUri) && StringUtils.charCount(aUri, '#') >= 2) {
+                // Browsers are able to handle URLs with double # by ignoring everything after the
+                // second # occurrence. But Java implementation considers it an invalid URL.
+                // Remove everything after the second #.
+                int index = aUri.indexOf("#", aUri.indexOf("#") + 1);
+                return parseUri(aUri.substring(0, index));
+            }
+            throw e;
+        }
+    }
+
     public static String urlForText(@NonNull Context context, @NonNull String text) {
         String url = text.trim();
         if ((UrlUtils.isDomain(text) || UrlUtils.isIPUri(text)) && !text.contains(" ")) {