Message-ID: <gh-pgjdbc-pgjdbc-2861@github.com>
From: "SuperPat45 (@SuperPat45)" <noreply+SuperPat45@github.com>
To: "pgjdbc/pgjdbc" <noreply+pgjdbc-pgjdbc@github.com>
Date: Tue, 21 Mar 2023 15:48:48 +0000
Subject: [pgjdbc/pgjdbc] issue #2861: setFetchSize is ignored?
List-Id: <gh-pgjdbc-pgjdbc.github.com>
X-GitHub-Author-Id: 7791600
X-GitHub-Author-Login: SuperPat45
X-GitHub-Issue: 2861
X-GitHub-Repo: pgjdbc/pgjdbc
X-GitHub-State: open
X-GitHub-Type: issue
X-GitHub-Url: https://github.com/pgjdbc/pgjdbc/issues/2861
Content-Type: text/plain; charset=utf-8

Please read https://stackoverflow.com/help/minimal-reproducible-example 

**Describe the issue**
Due to a bug, I have a query that returned 1 million rows instead of 10000....
I specified that I wanted the database to return only 100 rows at a time.

I noticed that the memory was full before the executeQuery() function returned the ResultSet...
Looking with JVisualVM I saw hundreds of thousands instances of the org.postgresql.core.Tuple class
So, It seem that all rows are fetched ignoring the fetchSize limit.

**Driver Version?** 
42.5.0

**Java Version?**
17.0.4

**OS Version?**
Windows 10

**PostgreSQL Version?**
14.5

**To Reproduce**
```
Statement statement = connection.createStatement();
statement.setFetchSize(100);
ResultSet resultSet = statement.executeQuery("select * from one_million_row_table");
System.out.println("OK");
```

**Expected behaviour**
The result set should have been returned by the executeQuery() function without exploding the memory as I specified to only fetch 100 rows at a time.
I use the result set to write data in a text file so, in this case, the memory should never have been filled completely.

**Logs**
java.lang.OutOfMemoryError: Java heap space