[IO-429] Check for long streams in IOUtils.toByteArray #175.

Author: Gary Gregory

src/changes/changes.xml

@@ -126,6 +126,9 @@ The <action> type attribute can be add,update,fix,remove.
       <action dev="ggregory" type="fix" due-to="Rob Spoor, Gary Gregory">
         Prevent infinite loop with AbstractCharacterFilterReader if EOF is filtered out #226.
       </action>
+      <action issue="IO-429" dev="ggregory" type="fix" due-to="Rob Spoor, Ivan Leskin">
+        Check for long streams in IOUtils.toByteArray #175.
+      </action>
       <!-- ADD -->
       <action dev="ggregory" type="add" due-to="Gary Gregory">
         Add FileSystemProviders class.

src/main/java/org/apache/commons/io/IOUtils.java

@@ -55,6 +55,7 @@
 import org.apache.commons.io.output.ByteArrayOutputStream;
 import org.apache.commons.io.output.NullOutputStream;
 import org.apache.commons.io.output.StringBuilderWriter;
+import org.apache.commons.io.output.ThresholdingOutputStream;
 import org.apache.commons.io.output.UnsynchronizedByteArrayOutputStream;
 
 /**
@@ -2394,12 +2395,17 @@ public static BufferedReader toBufferedReader(final Reader reader, final int siz
      * @param inputStream the {@code InputStream} to read.
      * @return the requested byte array.
      * @throws NullPointerException if the InputStream is {@code null}.
-     * @throws IOException if an I/O error occurs.
+     * @throws IOException if an I/O error occurs or reading more than {@link Integer#MAX_VALUE} occurs. 
      */
     public static byte[] toByteArray(final InputStream inputStream) throws IOException {
-        try (final UnsynchronizedByteArrayOutputStream output = new UnsynchronizedByteArrayOutputStream()) {
-            copy(inputStream, output);
-            return output.toByteArray();
+        // We use a ThresholdingOutputStream to avoid reading AND writing more than Integer.MAX_VALUE.
+        try (final UnsynchronizedByteArrayOutputStream ubaOutput = new UnsynchronizedByteArrayOutputStream();
+            final ThresholdingOutputStream thresholdOuput = new ThresholdingOutputStream(Integer.MAX_VALUE, os -> {
+                throw new IllegalArgumentException(
+                    String.format("Cannot read more than %,d into a byte array", Integer.MAX_VALUE));
+            }, os -> ubaOutput)) {
+            copy(inputStream, thresholdOuput);
+            return ubaOutput.toByteArray();
         }
     }
 

src/test/java/org/apache/commons/io/IOUtilsTestCase.java

@@ -62,13 +62,15 @@
 import java.util.List;
 
 import org.apache.commons.io.function.IOConsumer;
+import org.apache.commons.io.input.CircularInputStream;
 import org.apache.commons.io.input.NullInputStream;
 import org.apache.commons.io.output.AppendableWriter;
 import org.apache.commons.io.output.NullOutputStream;
 import org.apache.commons.io.output.StringBuilderWriter;
 import org.apache.commons.io.test.TestUtils;
 import org.apache.commons.io.test.ThrowOnCloseReader;
 import org.junit.jupiter.api.BeforeEach;
+import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
 import org.junit.jupiter.api.io.TempDir;
 
@@ -1435,6 +1437,13 @@ public void testToByteArray_InputStream() throws Exception {
         }
     }
 
+    @Test
+    @Disabled("Disable by default as it uses too much memory and can cause builds to fail.")
+    public void testToByteArray_InputStream_LongerThanIntegerMaxValue() throws Exception {
+        final CircularInputStream cin = new CircularInputStream(IOUtils.byteArray(), Integer.MAX_VALUE + 1L);
+        assertThrows(IllegalArgumentException.class, () -> IOUtils.toByteArray(cin));
+    }
+
     @Test
     public void testToByteArray_InputStream_NegativeSize() throws Exception {