diff --git a/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4 b/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
index 9a643465a999..ae1853fbfe06 100644
--- a/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
+++ b/sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBase.g4
@@ -16,6 +16,30 @@
 
 grammar SqlBase;
 
+@members {
+  /**
+   * Verify whether current token is a valid decimal token (which contains dot).
+   * Returns true if the character that follows the token is not a digit or letter or underscore.
+   *
+   * For example:
+   * For char stream "2.3", "2." is not a valid decimal token, because it is followed by digit '3'.
+   * For char stream "2.3_", "2.3" is not a valid decimal token, because it is followed by '_'.
+   * For char stream "2.3W", "2.3" is not a valid decimal token, because it is followed by 'W'.
+   * For char stream "12.0D 34.E2+0.12 "  12.0D is a valid decimal token because it is folllowed
+   * by a space. 34.E2 is a valid decimal token because it is followed by symbol '+'
+   * which is not a digit or letter or underscore.
+   */
+  public boolean isValidDecimal() {
+    int nextChar = _input.LA(1);
+    if (nextChar >= 'A' && nextChar <= 'Z' || nextChar >= '0' && nextChar <= '9' ||
+      nextChar == '_') {
+      return false;
+    } else {
+      return true;
+    }
+  }
+}
+
 tokens {
     DELIMITER
 }
@@ -917,23 +941,22 @@ INTEGER_VALUE
     ;
 
 DECIMAL_VALUE
-    : DIGIT+ '.' DIGIT*
-    | '.' DIGIT+
+    : DECIMAL_DIGITS {isValidDecimal()}?
     ;
 
 SCIENTIFIC_DECIMAL_VALUE
-    : DIGIT+ ('.' DIGIT*)? EXPONENT
-    | '.' DIGIT+ EXPONENT
+    : DIGIT+ EXPONENT
+    | DECIMAL_DIGITS EXPONENT {isValidDecimal()}?
     ;
 
 DOUBLE_LITERAL
-    :
-    (INTEGER_VALUE | DECIMAL_VALUE | SCIENTIFIC_DECIMAL_VALUE) 'D'
+    : DIGIT+ EXPONENT? 'D'
+    | DECIMAL_DIGITS EXPONENT? 'D' {isValidDecimal()}?
     ;
 
 BIGDECIMAL_LITERAL
-    :
-    (INTEGER_VALUE | DECIMAL_VALUE | SCIENTIFIC_DECIMAL_VALUE) 'BD'
+    : DIGIT+ EXPONENT? 'BD'
+    | DECIMAL_DIGITS EXPONENT? 'BD' {isValidDecimal()}?
     ;
 
 IDENTIFIER
@@ -944,6 +967,11 @@ BACKQUOTED_IDENTIFIER
     : '`' ( ~'`' | '``' )* '`'
     ;
 
+fragment DECIMAL_DIGITS
+    : DIGIT+ '.' DIGIT*
+    | '.' DIGIT+
+    ;
+
 fragment EXPONENT
     : 'E' [+-]? DIGIT+
     ;
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/ExpressionParserSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/ExpressionParserSuite.scala
index 4e399eef1fed..f319215f0568 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/ExpressionParserSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/ExpressionParserSuite.scala
@@ -18,7 +18,7 @@ package org.apache.spark.sql.catalyst.parser
 
 import java.sql.{Date, Timestamp}
 
-import org.apache.spark.sql.catalyst.FunctionIdentifier
+import org.apache.spark.sql.catalyst.{FunctionIdentifier, TableIdentifier}
 import org.apache.spark.sql.catalyst.analysis.{UnresolvedAttribute, _}
 import org.apache.spark.sql.catalyst.expressions._
 import org.apache.spark.sql.catalyst.plans.PlanTest
@@ -518,4 +518,17 @@ class ExpressionParserSuite extends PlanTest {
     assertEqual("current_date", CurrentDate())
     assertEqual("current_timestamp", CurrentTimestamp())
   }
+
+  test("SPARK-17364, fully qualified column name which starts with number") {
+    assertEqual("123_", UnresolvedAttribute("123_"))
+    assertEqual("1a.123_", UnresolvedAttribute("1a.123_"))
+    // ".123" should not be treated as token of type DECIMAL_VALUE
+    assertEqual("a.123A", UnresolvedAttribute("a.123A"))
+    // ".123E3" should not be treated as token of type SCIENTIFIC_DECIMAL_VALUE
+    assertEqual("a.123E3_column", UnresolvedAttribute("a.123E3_column"))
+    // ".123D" should not be treated as token of type DOUBLE_LITERAL
+    assertEqual("a.123D_column", UnresolvedAttribute("a.123D_column"))
+    // ".123BD" should not be treated as token of type BIGDECIMAL_LITERAL
+    assertEqual("a.123BD_column", UnresolvedAttribute("a.123BD_column"))
+  }
 }
diff --git a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
index dadb8a8def43..793be8953d07 100644
--- a/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
+++ b/sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/TableIdentifierParserSuite.scala
@@ -91,4 +91,17 @@ class TableIdentifierParserSuite extends SparkFunSuite {
       assert(TableIdentifier(nonReserved) === parseTableIdentifier(nonReserved))
     }
   }
+
+  test("SPARK-17364 table identifier - contains number") {
+    assert(parseTableIdentifier("123_") == TableIdentifier("123_"))
+    assert(parseTableIdentifier("1a.123_") == TableIdentifier("123_", Some("1a")))
+    // ".123" should not be treated as token of type DECIMAL_VALUE
+    assert(parseTableIdentifier("a.123A") == TableIdentifier("123A", Some("a")))
+    // ".123E3" should not be treated as token of type SCIENTIFIC_DECIMAL_VALUE
+    assert(parseTableIdentifier("a.123E3_LIST") == TableIdentifier("123E3_LIST", Some("a")))
+    // ".123D" should not be treated as token of type DOUBLE_LITERAL
+    assert(parseTableIdentifier("a.123D_LIST") == TableIdentifier("123D_LIST", Some("a")))
+    // ".123BD" should not be treated as token of type BIGDECIMAL_LITERAL
+    assert(parseTableIdentifier("a.123BD_LIST") == TableIdentifier("123BD_LIST", Some("a")))
+  }
 }