HDFS-7816. Unable to open webhdfs paths with "+". Contributed by Haohui Mai

(cherry picked from commit e79be0ee123d05104eb34eb854afcf9fa78baef2)

Conflicts:
	hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt

hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt

@@ -908,6 +908,8 @@ Release 2.7.0 - UNRELEASED
     HDFS-7929. inotify unable fetch pre-upgrade edit log segments once upgrade
     starts (Zhe Zhang via Colin P. McCabe)
 
+    HDFS-7816. Unable to open webhdfs paths with "+". (wheat9 via kihwal)
+
     BREAKDOWN OF HDFS-7584 SUBTASKS AND RELATED JIRAS
 
       HDFS-7720. Quota by Storage Type API, tools and ClientNameNode

hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/web/webhdfs/ParameterParser.java

@@ -18,6 +18,7 @@
 package org.apache.hadoop.hdfs.server.datanode.web.webhdfs;
 
 import io.netty.handler.codec.http.QueryStringDecoder;
+import org.apache.commons.io.Charsets;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.permission.FsPermission;
 import org.apache.hadoop.hdfs.HAUtil;
@@ -39,6 +40,7 @@ import org.apache.hadoop.security.token.Token;
 
 import java.io.IOException;
 import java.net.URI;
+import java.nio.charset.Charset;
 import java.util.List;
 import java.util.Map;
 
@@ -51,7 +53,8 @@ class ParameterParser {
   private final Map<String, List<String>> params;
 
   ParameterParser(QueryStringDecoder decoder, Configuration conf) {
-    this.path = QueryStringDecoder.decodeComponent(decoder.path().substring(WEBHDFS_PREFIX_LENGTH));
+    this.path = decodeComponent(decoder.path().substring
+        (WEBHDFS_PREFIX_LENGTH), Charsets.UTF_8);
     this.params = decoder.parameters();
     this.conf = conf;
   }
@@ -127,4 +130,78 @@ class ParameterParser {
     List<String> p = params.get(key);
     return p == null ? null : p.get(0);
   }
+
+  /**
+   * The following function behaves exactly the same as netty's
+   * <code>QueryStringDecoder#decodeComponent</code> except that it
+   * does not decode the '+' character as space. WebHDFS takes this scheme
+   * to maintain the backward-compatibility for pre-2.7 releases.
+   */
+  private static String decodeComponent(final String s, final Charset charset) {
+    if (s == null) {
+      return "";
+    }
+    final int size = s.length();
+    boolean modified = false;
+    for (int i = 0; i < size; i++) {
+      final char c = s.charAt(i);
+      if (c == '%' || c == '+') {
+        modified = true;
+        break;
+      }
+    }
+    if (!modified) {
+      return s;
+    }
+    final byte[] buf = new byte[size];
+    int pos = 0;  // position in `buf'.
+    for (int i = 0; i < size; i++) {
+      char c = s.charAt(i);
+      if (c == '%') {
+        if (i == size - 1) {
+          throw new IllegalArgumentException("unterminated escape sequence at" +
+                                                 " end of string: " + s);
+        }
+        c = s.charAt(++i);
+        if (c == '%') {
+          buf[pos++] = '%';  // "%%" -> "%"
+          break;
+        }
+        if (i == size - 1) {
+          throw new IllegalArgumentException("partial escape sequence at end " +
+                                                 "of string: " + s);
+        }
+        c = decodeHexNibble(c);
+        final char c2 = decodeHexNibble(s.charAt(++i));
+        if (c == Character.MAX_VALUE || c2 == Character.MAX_VALUE) {
+          throw new IllegalArgumentException(
+              "invalid escape sequence `%" + s.charAt(i - 1) + s.charAt(
+                  i) + "' at index " + (i - 2) + " of: " + s);
+        }
+        c = (char) (c * 16 + c2);
+        // Fall through.
+      }
+      buf[pos++] = (byte) c;
+    }
+    return new String(buf, 0, pos, charset);
+  }
+
+  /**
+   * Helper to decode half of a hexadecimal number from a string.
+   * @param c The ASCII character of the hexadecimal number to decode.
+   * Must be in the range {@code [0-9a-fA-F]}.
+   * @return The hexadecimal value represented in the ASCII character
+   * given, or {@link Character#MAX_VALUE} if the character is invalid.
+   */
+  private static char decodeHexNibble(final char c) {
+    if ('0' <= c && c <= '9') {
+      return (char) (c - '0');
+    } else if ('a' <= c && c <= 'f') {
+      return (char) (c - 'a' + 10);
+    } else if ('A' <= c && c <= 'F') {
+      return (char) (c - 'A' + 10);
+    } else {
+      return Character.MAX_VALUE;
+    }
+  }
 }

hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/server/datanode/web/webhdfs/TestParameterParser.java

@@ -56,13 +56,12 @@ public class TestParameterParser {
 
   @Test
   public void testDecodePath() {
-    final String SCAPED_PATH = "hdfs-6662/test%25251%26%3Dtest?op=OPEN";
-    final String EXPECTED_PATH = "/hdfs-6662/test%251&=test";
+    final String ESCAPED_PATH = "/test%25+1%26%3Dtest?op=OPEN&foo=bar";
+    final String EXPECTED_PATH = "/test%+1&=test";
 
-    Configuration conf = DFSTestUtil.newHAConfiguration(LOGICAL_NAME);
+    Configuration conf = new Configuration();
     QueryStringDecoder decoder = new QueryStringDecoder(
-      WebHdfsHandler.WEBHDFS_PREFIX + "/"
-      + SCAPED_PATH);
+      WebHdfsHandler.WEBHDFS_PREFIX + ESCAPED_PATH);
     ParameterParser testParser = new ParameterParser(decoder, conf);
     Assert.assertEquals(EXPECTED_PATH, testParser.path());
   }