> and I’m curious why the one-line fix to SqlToRelConverter you propose in the > Jira case is not sufficient.
I personally believe that this is sufficient. I think that commentor would like to see a “better” solution to handling correlation in SqlToRelConverter as a whole. I don’t think that’s generally unreasonable, but I don’t think that should stop us from solving this bug now; especially since this change is so low impact. -Ian From: Julian Hyde <[email protected]> Date: Friday, March 6, 2026 at 09:53 To: [email protected] <[email protected]> Subject: Re: CALCITE-7405 Reviews or thoughts? This Message Is From an External Sender This message came from outside your organization. Report Suspicious<https://us-phishalarm-ewt.proofpoint.com/EWT/v1/Iz9xO38YGHZK!bJNDSyAEGWPNDtTPe03mfepdnaFZEvpOPzfHTXAUDGaivwZGyrbbpuICXwTiCPLLF4dGfVA_yKGCbUbvwbXIliEdNzuijexavdLEWw$> My quick thoughts are… this seems to be a serious issue… we should fix it… you have done a lot of good research into this… and I’m curious why the one-line fix to SqlToRelConverter you propose in the Jira case is not sufficient. > On Mar 6, 2026, at 9:39 AM, Ian Bertolacci via dev <[email protected]> > wrote: > > Hi, we discovered CALCITE-7405 a few weeks back and provided a minimal PR to > address it. > (https://urldefense.com/v3/__https://github.com/apache/calcite/pull/4779__;!!Iz9xO38YGHZK!4VAy4M9rswq9EJBpae-JFJ3byx80TNW84wDsp7pN4ctzRdsYPm6Zw_xEIhxl_hzFR0Tu0fGC3UBPNhMVMJu7Z7IyKW8$). > I was hoping to get some feedback on it. > There’s a proposal that we approach correlation differently in > SqlToRelConverter, but I would prefer to solve this problem now, using this > minimal approach,, especially since that new approach would take a long time > to materialize. > > Thanks! > -Ian J. Bertolacci
