fhueske commented on a change in pull request #6621:  [FLINK-8686] [sql-client] 
Limit result size for prototyping modes
URL: https://github.com/apache/flink/pull/6621#discussion_r213014246
 
 

 ##########
 File path: 
flink-libraries/flink-sql-client/src/main/java/org/apache/flink/table/client/gateway/local/result/MaterializedCollectStreamResult.java
 ##########
 @@ -112,30 +129,43 @@ else if (!isRetrieving()) {
        @Override
        protected void processRecord(Tuple2<Boolean, Row> change) {
                synchronized (resultLock) {
-                       final Row row = change.f1;
                        // insert
                        if (change.f0) {
-                               materializedTable.add(row);
-                               rowPositionCache.put(row, 
materializedTable.size() - 1);
+                               processInsert(change.f1);
                        }
                        // delete
                        else {
-                               // delete the newest record first to minimize 
per-page changes
-                               final Integer cachedPos = 
rowPositionCache.get(row);
-                               final int startSearchPos;
-                               if (cachedPos != null) {
-                                       startSearchPos = Math.min(cachedPos, 
materializedTable.size() - 1);
-                               } else {
-                                       startSearchPos = 
materializedTable.size() - 1;
-                               }
-
-                               for (int i = startSearchPos; i >= 0; i--) {
-                                       if 
(materializedTable.get(i).equals(row)) {
-                                               materializedTable.remove(i);
-                                               rowPositionCache.remove(row);
-                                               break;
-                                       }
-                               }
+                               processDelete(change.f1);
+                       }
+               }
+       }
+
+       // 
--------------------------------------------------------------------------------------------
+
+       private void processInsert(Row row) {
+               // limit the materialized table
+               if (materializedTable.size() >= maxRowCount) {
+                       processDelete(materializedTable.get(0));
 
 Review comment:
   Why should we search for the row if we know that we want to remove it from 
the first position?
   Couldn't we just do `materializedTable.remove(0)`? This will be expensive 
because `materializedTable` is an `ArrayList` but what's the chance that there 
is another row that is equal to the one that is removed?

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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