Re: Condition pushdown: why (=) is pushed down into join, but BETWEEN or >= is not? - Mailing list pgsql-hackers

From Justin Pryzby
Subject Re: Condition pushdown: why (=) is pushed down into join, but BETWEEN or >= is not?
Date
Msg-id 20220201171329.GA23027@telsasoft.com
Whole thread Raw
In response to Re: Condition pushdown: why (=) is pushed down into join, but BETWEEN or >= is not?  (Andy Fan <zhihui.fan1213@gmail.com>)
Responses Re: Condition pushdown: why (=) is pushed down into join, but BETWEEN or >= is not?  (Andy Fan <zhihui.fan1213@gmail.com>)
List pgsql-hackers
> Subject: [PATCH v1 1/6] Rebaee David's patch against the latest code.

If you use git-am, then the author/commit information is preserved.
It's probably good to include a link to the patch in any case.

> Subject: [PATCH v1 4/6] remove duplicated qual executing.

                                                                        
 
---
                                                                        
 
 src/backend/optimizer/path/equivclass.c | 22 +++++++++++++++++++
                                                                        
 
 src/backend/optimizer/plan/createplan.c | 29 +++++++++++++++++++++++--
                                                                        
 
 src/include/optimizer/paths.h           |  2 ++
                                                                        
 
 src/test/regress/parallel_schedule      |  2 ++
                                                                        
 
 4 files changed, 53 insertions(+), 2 deletions(-)
                                                                        
 

I think the ./ec_filter test is missing from from this patch.

> Subject: [PATCH v1 6/6] adding some test cases for this feature and fix the existing case

                            
 
The tests should be updated with the corresponding patches.  It's common for
the patches to be commited separately, like if 0001 is ready but the others are
still evolving.

I'm not sure whether you think this patch is ready to be added to a commitfest,
but do you know about the CI infrastructure ?  It allows running all the cfbot
tests for a github branch against 4 OS, which helps catch portability issues,
including memory errors and unstable explain output.  See: src/tools/ci/README.
There's an failure in postgres_fdw, probably the output needs to be updated.

-- 
Justin



pgsql-hackers by date:

Previous
From: Fujii Masao
Date:
Subject: Re: Add checkpoint and redo LSN to LogCheckpointEnd log message
Next
From: Robert Treat
Date:
Subject: Re: autovacuum prioritization