With a reference to my previous report of memory leak (
The memory leak appears to has been fixed when testing with the script provided in the issue report. Thank you very much for the quick fix !
I've been running the script for 10 minutes on my PC and the memory usage reported by the Windows Task Manager started from 6336K to reach stable level of 6492K. This test was run after a slight modification of the test script, namely: added gc.collect() at every 100th iteration.
Then I modified the script thus:
cursor = conn.cursor(dictionary=False)
cursor = conn.cursor(dictionary=True)
Still gc.collect() every 100th iteration.
Running this modified script for 10 minutes showed memory usage increase from initial 6348K to 7064K.
IMHO this indicates some outstanding allocations related to the result type as dictionary.
Modified script is attached.