apache · hvanhovell · Jan 7, 2016 · Jan 7, 2016 · Jan 7, 2016 · Jan 7, 2016
diff --git a/python/pyspark/sql/tests.py b/python/pyspark/sql/tests.py
@@ -1081,8 +1081,7 @@ def test_replace(self):
     def test_capture_analysis_exception(self):
         self.assertRaises(AnalysisException, lambda: self.sqlCtx.sql("select abc"))
         self.assertRaises(AnalysisException, lambda: self.df.selectExpr("a + b"))
-        # RuntimeException should not be captured
-        self.assertRaises(py4j.protocol.Py4JJavaError, lambda: self.sqlCtx.sql("abc"))
+        self.assertRaises(AnalysisException, lambda: self.sqlCtx.sql("abc"))
 
     def test_capture_illegalargument_exception(self):
         self.assertRaisesRegexp(IllegalArgumentException, "Setting negative mapred.reduce.tasks",

diff --git a/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/ExpressionParser.g b/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/ExpressionParser.g
@@ -123,7 +123,6 @@ constant
     | SmallintLiteral
     | TinyintLiteral
     | DecimalLiteral
-    | charSetStringLiteral
     | booleanValue
     ;
 
@@ -132,13 +131,6 @@ stringLiteralSequence
     StringLiteral StringLiteral+ -> ^(TOK_STRINGLITERALSEQUENCE StringLiteral StringLiteral+)
     ;
 
-charSetStringLiteral
-@init { gParent.pushMsg("character string literal", state); }
-@after { gParent.popMsg(state); }
-    :
-    csName=CharSetName csLiteral=CharSetLiteral -> ^(TOK_CHARSETLITERAL $csName $csLiteral)
-    ;
-
 dateLiteral
     :
     KW_DATE StringLiteral ->
@@ -163,22 +155,38 @@ timestampLiteral
 
 intervalLiteral
     :
-    KW_INTERVAL StringLiteral qualifiers=intervalQualifiers ->
-    {
-      adaptor.create($qualifiers.tree.token.getType(), $StringLiteral.text)
+    (KW_INTERVAL intervalConstant KW_YEAR KW_TO KW_MONTH) => KW_INTERVAL intervalConstant KW_YEAR KW_TO KW_MONTH
+      -> ^(TOK_INTERVAL_YEAR_MONTH_LITERAL intervalConstant)
+    | (KW_INTERVAL intervalConstant KW_DAY KW_TO KW_SECOND) => KW_INTERVAL intervalConstant KW_DAY KW_TO KW_SECOND
+      -> ^(TOK_INTERVAL_DAY_TIME_LITERAL intervalConstant)
+    | KW_INTERVAL
+      ((intervalConstant KW_YEAR)=> year=intervalConstant KW_YEAR)?
+      ((intervalConstant KW_MONTH)=> month=intervalConstant KW_MONTH)?
+      ((intervalConstant KW_WEEK)=> week=intervalConstant KW_WEEK)?
+      ((intervalConstant KW_DAY)=> day=intervalConstant KW_DAY)?
+      ((intervalConstant KW_HOUR)=> hour=intervalConstant KW_HOUR)?
+      ((intervalConstant KW_MINUTE)=> minute=intervalConstant KW_MINUTE)?
+      ((intervalConstant KW_SECOND)=> second=intervalConstant KW_SECOND)?
+      (millisecond=intervalConstant KW_MILLISECOND)?
+      (microsecond=intervalConstant KW_MICROSECOND)?
+      -> ^(TOK_INTERVAL
+          ^(TOK_INTERVAL_YEAR_LITERAL $year?)
+          ^(TOK_INTERVAL_MONTH_LITERAL $month?)
+          ^(TOK_INTERVAL_WEEK_LITERAL $week?)
+          ^(TOK_INTERVAL_DAY_LITERAL $day?)
+          ^(TOK_INTERVAL_HOUR_LITERAL $hour?)
+          ^(TOK_INTERVAL_MINUTE_LITERAL $minute?)
+          ^(TOK_INTERVAL_SECOND_LITERAL $second?)
+          ^(TOK_INTERVAL_MILLISECOND_LITERAL $millisecond?)
+          ^(TOK_INTERVAL_MICROSECOND_LITERAL $microsecond?))
+    ;
+
+intervalConstant
+    :
+    sign=(MINUS|PLUS)? value=Number -> {
+      adaptor.create(Number, ($sign != null ? $sign.getText() : "") + $value.getText())
     }
-    ;
-
-intervalQualifiers
-    :
-    KW_YEAR KW_TO KW_MONTH -> TOK_INTERVAL_YEAR_MONTH_LITERAL
-    | KW_DAY KW_TO KW_SECOND -> TOK_INTERVAL_DAY_TIME_LITERAL
-    | KW_YEAR -> TOK_INTERVAL_YEAR_LITERAL
-    | KW_MONTH -> TOK_INTERVAL_MONTH_LITERAL
-    | KW_DAY -> TOK_INTERVAL_DAY_LITERAL
-    | KW_HOUR -> TOK_INTERVAL_HOUR_LITERAL
-    | KW_MINUTE -> TOK_INTERVAL_MINUTE_LITERAL
-    | KW_SECOND -> TOK_INTERVAL_SECOND_LITERAL
+    | StringLiteral
     ;
 
 expression
@@ -219,7 +227,8 @@ nullCondition
 
 precedenceUnaryPrefixExpression
     :
-    (precedenceUnaryOperator^)* precedenceFieldExpression
+    (precedenceUnaryOperator+)=> precedenceUnaryOperator^ precedenceUnaryPrefixExpression
+    | precedenceFieldExpression
     ;
 
 precedenceUnarySuffixExpression

diff --git a/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/FromClauseParser.g b/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/FromClauseParser.g
@@ -206,11 +206,8 @@ tableName
 @init { gParent.pushMsg("table name", state); }
 @after { gParent.popMsg(state); }
     :
-    db=identifier DOT tab=identifier
-    -> ^(TOK_TABNAME $db $tab)
-    |
-    tab=identifier
-    -> ^(TOK_TABNAME $tab)
+    id1=identifier (DOT id2=identifier)?
+    -> ^(TOK_TABNAME $id1 $id2?)
     ;
 
 viewName

diff --git a/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/SparkSqlLexer.g b/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/SparkSqlLexer.g
@@ -307,12 +307,12 @@ KW_AUTHORIZATION: 'AUTHORIZATION';
 KW_CONF: 'CONF';
 KW_VALUES: 'VALUES';
 KW_RELOAD: 'RELOAD';
-KW_YEAR: 'YEAR';
-KW_MONTH: 'MONTH';
-KW_DAY: 'DAY';
-KW_HOUR: 'HOUR';
-KW_MINUTE: 'MINUTE';
-KW_SECOND: 'SECOND';
+KW_YEAR: 'YEAR'|'YEARS';
+KW_MONTH: 'MONTH'|'MONTHS';
+KW_DAY: 'DAY'|'DAYS';
+KW_HOUR: 'HOUR'|'HOURS';
+KW_MINUTE: 'MINUTE'|'MINUTES';
+KW_SECOND: 'SECOND'|'SECONDS';
 KW_START: 'START';
 KW_TRANSACTION: 'TRANSACTION';
 KW_COMMIT: 'COMMIT';
@@ -324,6 +324,9 @@ KW_ISOLATION: 'ISOLATION';
 KW_LEVEL: 'LEVEL';
 KW_SNAPSHOT: 'SNAPSHOT';
 KW_AUTOCOMMIT: 'AUTOCOMMIT';
+KW_WEEK: 'WEEK'|'WEEKS';
+KW_MILLISECOND: 'MILLISECOND'|'MILLISECONDS';
+KW_MICROSECOND: 'MICROSECOND'|'MICROSECONDS';
 
 // Operators
 // NOTE: if you add a new function/operator, add it to sysFuncNames so that describe function _FUNC_ will work.
@@ -400,12 +403,6 @@ StringLiteral
     )+
     ;
 
-CharSetLiteral
-    :
-    StringLiteral
-    | '0' 'X' (HexDigit|Digit)+
-    ;
-
 BigintLiteral
     :
     (Digit)+ 'L'
@@ -433,7 +430,7 @@ ByteLengthLiteral
 
 Number
     :
-    (Digit)+ ( DOT (Digit)* (Exponent)? | Exponent)?
+    ((Digit+ (DOT Digit*)?) | (DOT Digit+)) Exponent?
     ;
 
 /*
@@ -456,10 +453,10 @@ An Identifier can be:
 - macro name
 - hint name
 - window name
-*/    
+*/
 Identifier
     :
-    (Letter | Digit) (Letter | Digit | '_')*
+    (Letter | Digit | '_')+
     | {allowQuotedId()}? QuotedIdentifier  /* though at the language level we allow all Identifiers to be QuotedIdentifiers; 
                                               at the API level only columns are allowed to be of this form */
     | '`' RegexComponent+ '`'
@@ -471,11 +468,6 @@ QuotedIdentifier
     '`'  ( '``' | ~('`') )* '`' { setText(getText().substring(1, getText().length() -1 ).replaceAll("``", "`")); }
     ;
 
-CharSetName
-    :
-    '_' (Letter | Digit | '_' | '-' | '.' | ':' )+
-    ;
-
 WS  :  (' '|'\r'|'\t'|'\n') {$channel=HIDDEN;}
     ;
 

diff --git a/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/SparkSqlParser.g b/sql/catalyst/src/main/antlr3/org/apache/spark/sql/catalyst/parser/SparkSqlParser.g
@@ -116,16 +116,20 @@ TOK_DATELITERAL;
 TOK_DATETIME;
 TOK_TIMESTAMP;
 TOK_TIMESTAMPLITERAL;
+TOK_INTERVAL;
 TOK_INTERVAL_YEAR_MONTH;
 TOK_INTERVAL_YEAR_MONTH_LITERAL;
 TOK_INTERVAL_DAY_TIME;
 TOK_INTERVAL_DAY_TIME_LITERAL;
 TOK_INTERVAL_YEAR_LITERAL;
 TOK_INTERVAL_MONTH_LITERAL;
+TOK_INTERVAL_WEEK_LITERAL;
 TOK_INTERVAL_DAY_LITERAL;
 TOK_INTERVAL_HOUR_LITERAL;
 TOK_INTERVAL_MINUTE_LITERAL;
 TOK_INTERVAL_SECOND_LITERAL;
+TOK_INTERVAL_MILLISECOND_LITERAL;
+TOK_INTERVAL_MICROSECOND_LITERAL;
 TOK_STRING;
 TOK_CHAR;
 TOK_VARCHAR;
@@ -228,7 +232,6 @@ TOK_TMP_FILE;
 TOK_TABSORTCOLNAMEASC;
 TOK_TABSORTCOLNAMEDESC;
 TOK_STRINGLITERALSEQUENCE;
-TOK_CHARSETLITERAL;
 TOK_CREATEFUNCTION;
 TOK_DROPFUNCTION;
 TOK_RELOADFUNCTION;
@@ -509,7 +512,9 @@ import java.util.HashMap;
     xlateMap.put("KW_UPDATE", "UPDATE");
     xlateMap.put("KW_VALUES", "VALUES");
     xlateMap.put("KW_PURGE", "PURGE");
-
+    xlateMap.put("KW_WEEK", "WEEK");
+    xlateMap.put("KW_MILLISECOND", "MILLISECOND");
+    xlateMap.put("KW_MICROSECOND", "MICROSECOND");
 
     // Operators
     xlateMap.put("DOT", ".");
@@ -2078,6 +2083,7 @@ primitiveType
     | KW_SMALLINT      ->    TOK_SMALLINT
     | KW_INT           ->    TOK_INT
     | KW_BIGINT        ->    TOK_BIGINT
+    | KW_LONG          ->    TOK_BIGINT
     | KW_BOOLEAN       ->    TOK_BOOLEAN
     | KW_FLOAT         ->    TOK_FLOAT
     | KW_DOUBLE        ->    TOK_DOUBLE

diff --git a/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java b/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/parser/ParseUtils.java
@@ -18,46 +18,17 @@
 
 package org.apache.spark.sql.catalyst.parser;
 
-import java.io.UnsupportedEncodingException;
-
 /**
  * A couple of utility methods that help with parsing ASTs.
  *
- * Both methods in this class were take from the SemanticAnalyzer in Hive:
+ * The 'unescapeSQLString' method in this class was take from the SemanticAnalyzer in Hive:
  * ql/src/java/org/apache/hadoop/hive/ql/parse/BaseSemanticAnalyzer.java
  */
 public final class ParseUtils {
   private ParseUtils() {
     super();
   }
 
-  public static String charSetString(String charSetName, String charSetString)
-          throws UnsupportedEncodingException {
-    // The character set name starts with a _, so strip that
-    charSetName = charSetName.substring(1);
-    if (charSetString.charAt(0) == '\'') {
-      return new String(unescapeSQLString(charSetString).getBytes(), charSetName);
-    } else // hex input is also supported
-    {
-      assert charSetString.charAt(0) == '0';
-      assert charSetString.charAt(1) == 'x';
-      charSetString = charSetString.substring(2);
-
-      byte[] bArray = new byte[charSetString.length() / 2];
-      int j = 0;
-      for (int i = 0; i < charSetString.length(); i += 2) {
-        int val = Character.digit(charSetString.charAt(i), 16) * 16
-                + Character.digit(charSetString.charAt(i + 1), 16);
-        if (val > 127) {
-          val = val - 256;
-        }
-        bArray[j++] = (byte)val;
-      }
-
-      return new String(bArray, charSetName);
-    }
-  }
-
   private static final int[] multiplier = new int[] {1000, 100, 10, 1};
 
   @SuppressWarnings("nls")

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/AbstractSparkSQLParser.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/AbstractSparkSQLParser.scala
@@ -26,9 +26,9 @@ import scala.util.parsing.input.CharArrayReader.EofCh
 import org.apache.spark.sql.catalyst.plans.logical._
 
 private[sql] abstract class AbstractSparkSQLParser
-  extends StandardTokenParsers with PackratParsers {
+  extends StandardTokenParsers with PackratParsers with ParserDialect {
 
-  def parse(input: String): LogicalPlan = synchronized {
+  def parsePlan(input: String): LogicalPlan = synchronized {
     // Initialize the Keywords.
     initLexical
     phrase(start)(new lexical.Scanner(input)) match {