vlsi commented on a change in pull request #1740: [CALCITE-3713] Remove column 
names from Project#digest
URL: https://github.com/apache/calcite/pull/1740#discussion_r365118346
 
 

 ##########
 File path: core/src/main/java/org/apache/calcite/rex/RexCall.java
 ##########
 @@ -127,11 +137,83 @@ protected final StringBuilder 
appendOperands(StringBuilder sb) {
           includeType = RexDigestIncludeType.NO_TYPE;
         }
       }
-      sb.append(((RexLiteral) operand).computeDigest(includeType));
+      operandDigests.add(((RexLiteral) operand).computeDigest(includeType));
+    }
+    int totalLength = (operandDigests.size() - 1) * 2; // commas
+    for (String s : operandDigests) {
+      totalLength += s.length();
+    }
+    sb.ensureCapacity(sb.length() + totalLength);
+    sortOperandsIfNeeded(sb, operands, operandDigests);
+    for (int i = 0; i < operandDigests.size(); i++) {
+      String op = operandDigests.get(i);
+      if (i != 0) {
+        sb.append(", ");
+      }
+      sb.append(op);
     }
     return sb;
   }
 
+  private void sortOperandsIfNeeded(StringBuilder sb,
+      List<RexNode> operands, List<String> operandDigests) {
+    if (operands.isEmpty() || !needNormalize()) {
+      return;
+    }
+    final SqlKind kind = op.getKind();
+    if (SqlKind.SYMMETRICAL_SAME_ARG_TYPE.contains(kind)) {
+      final RelDataType firstType = operands.get(0).getType();
+      for (int i = 1; i < operands.size(); i++) {
+        if (!equalSansNullability(firstType, operands.get(i).getType())) {
+          // Arguments have different type, thus they must not be sorted
 
 Review comment:
   I've no idea. We could probably add a logger or counter to so it would count 
the number of times the normalization was skipped for + / *.
   
   If you have production-like queries/test-cases, it would be nice if you 
could add them.
   
   However, I still think refining cost estimates and uniqueness metadata (see 
https://github.com/apache/calcite/pull/1702 ) has better benefits than enhanced 
+ * normalization.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to