From 753fa6b3bd3512ab8e8112e55be2aff667a845c4 Mon Sep 17 00:00:00 2001
From: Gian Merlino <gianmerlino@gmail.com>
Date: Thu, 10 Dec 2020 10:49:40 -0800
Subject: [PATCH] IdUtils: Forbid characters that cannot be used in znodes.
 (#10659)

* IdUtils: Forbid characters that cannot be used in znodes.

* Fix whitespace.
---
 .../org/apache/druid/common/utils/IdUtils.java   | 12 ++++++++++++
 .../apache/druid/common/utils/IdUtilsTest.java   | 16 ++++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/core/src/main/java/org/apache/druid/common/utils/IdUtils.java b/core/src/main/java/org/apache/druid/common/utils/IdUtils.java
index cbafef00e67..549bf458743 100644
--- a/core/src/main/java/org/apache/druid/common/utils/IdUtils.java
+++ b/core/src/main/java/org/apache/druid/common/utils/IdUtils.java
@@ -24,6 +24,7 @@ import com.google.common.base.Joiner;
 import com.google.common.base.Preconditions;
 import com.google.common.base.Strings;
 import org.apache.druid.java.util.common.DateTimes;
+import org.apache.druid.java.util.common.IAE;
 import org.apache.druid.java.util.common.StringUtils;
 import org.joda.time.DateTime;
 import org.joda.time.Interval;
@@ -60,6 +61,17 @@ public class IdUtils
         !m.matches(),
         StringUtils.format("%s cannot contain whitespace character except space.", thingToValidate)
     );
+
+    for (int i = 0; i < stringToValidate.length(); i++) {
+      final char c = stringToValidate.charAt(i);
+
+      // Curator doesn't permit any of the following ranges, so we can't either, because IDs are often used as
+      // znode paths. The first two ranges are control characters, the second two ranges correspond to surrogate
+      // pairs. This means that characters outside the basic multilingual plane, such as emojis, are not allowed. 😢
+      if (c > 0 && c < 31 || c > 127 && c < 159 || c > '\ud800' && c < '\uf8ff' || c > '\ufff0' && c < '\uffff') {
+        throw new IAE("%s cannot contain character #%d (at position %d).", thingToValidate, (int) c, i);
+      }
+    }
   }
 
   public static String getRandomId()
diff --git a/core/src/test/java/org/apache/druid/common/utils/IdUtilsTest.java b/core/src/test/java/org/apache/druid/common/utils/IdUtilsTest.java
index bd34967d4e0..806c74e6161 100644
--- a/core/src/test/java/org/apache/druid/common/utils/IdUtilsTest.java
+++ b/core/src/test/java/org/apache/druid/common/utils/IdUtilsTest.java
@@ -112,6 +112,22 @@ public class IdUtilsTest
     IdUtils.validateId(THINGO, "form\u000cfeed?");
   }
 
+  @Test
+  public void testInvalidUnprintableChars()
+  {
+    expectedException.expect(IllegalArgumentException.class);
+    expectedException.expectMessage("thingToValidate cannot contain character #129 (at position 4).");
+    IdUtils.validateId(THINGO, "form\u0081feed?");
+  }
+
+  @Test
+  public void testInvalidEmojis()
+  {
+    expectedException.expect(IllegalArgumentException.class);
+    expectedException.expectMessage("thingToValidate cannot contain character #55357 (at position 4).");
+    IdUtils.validateId(THINGO, "form💯feed?");
+  }
+
   @Test
   public void testNewTaskIdWithoutInterval()
   {