MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date: Tue, 08 Sep 2020 12:29:14 +0300
From: Alexey Kondratov <a.kondratov@postgrespro.ru>
To: Amit Langote <amitlangote09@gmail.com>
Cc: "Andrey V. Lepikhov" <a.lepikhov@postgrespro.ru>, Michael Paquier
 <michael@paquier.xyz>, Ashutosh Bapat <ashutosh.bapat@2ndquadrant.com>,
 Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>, Etsuro Fujita
 <etsuro.fujita@gmail.com>, PostgreSQL-Dev <pgsql-hackers@postgresql.org>
Subject: Re: [POC] Fast COPY FROM command for the table with foreign
 partitions
In-Reply-To: 
 <CA+HiwqEoSMpzMhzz1Dv7qFsFtzG71t5Wrgy4F+MoVUVR_5HdDQ@mail.gmail.com>
References: <ae25e79e-6122-720e-1349-fb832c6021df@postgrespro.ru>
 <CA+HiwqH2ehGurL9bYUydXBxVUH6aHN5yXXxESAOs9O6-f_+Qeg@mail.gmail.com>
 <cf1e92eb-9a8e-f5c1-f0cc-598882f1e546@postgrespro.ru>
 <CA+HiwqHRqG8X9W6-v-E6d0nw1k6aC7QSMk-SOvv4puOjWsQMmA@mail.gmail.com>
 <9a65d03b-250f-b086-5a3c-541d8e2de30d@postgrespro.ru>
 <CA+HiwqE6M4mwH=CDOKmy5DUHRDP-knLdAfmCh+GW7D59s3QDRg@mail.gmail.com>
 <CA+HiwqHtQaZp_t5M3z1+M0T9zd=0s3r_oOmi9TxHDdyW0_cxvg@mail.gmail.com>
 <ab62d228-cf9a-e517-8667-7df1184adf2a@postgrespro.ru>
 <CA+HiwqG1o5BDi0Hx6Uoxteyra2FbiSvaN4Fzs-i0pfKo1Zp_fg@mail.gmail.com>
 <CA+HiwqHT9UFhVHyKsavL5u=iVdW9k92pHHN=FYL0sEysGtvYsw@mail.gmail.com>
 <20200907072651.GF4867@paquier.xyz>
 <19ba27e7-ee1e-9117-68ad-fd609b5763ca@postgrespro.ru>
 <CA+HiwqEoSMpzMhzz1Dv7qFsFtzG71t5Wrgy4F+MoVUVR_5HdDQ@mail.gmail.com>
User-Agent: Roundcube Webmail/1.4.0
Message-ID: <0c67618c2e43b1dccb260dd6f90eeac2@postgrespro.ru>
Precedence: bulk

Hi,

I've started doing a review of v7 yesterday.

On 2020-09-08 10:34, Amit Langote wrote:
> On Mon, Sep 7, 2020 at 7:31 PM Andrey V. Lepikhov
> <a.lepikhov@postgrespro.ru> wrote:
>> >
>> v.7 (in attachment) fixes this problem.
>> I also accepted Amit's suggestion to rename all fdwapi routines such 
>> as
>> ForeignCopyIn to *ForeignCopy.
> 

It seems that naming is quite inconsistent now:

+	/* COPY a bulk of tuples into a foreign relation */
+	BeginForeignCopyIn_function BeginForeignCopy;
+	EndForeignCopyIn_function EndForeignCopy;
+	ExecForeignCopyIn_function ExecForeignCopy;

You get rid of this 'In' in the function names, but the types are still 
with it:

+typedef void (*BeginForeignCopyIn_function) (ModifyTableState *mtstate,
+		ResultRelInfo *rinfo);
+
+typedef void (*EndForeignCopyIn_function) (EState *estate,
+		ResultRelInfo *rinfo);
+
+typedef void (*ExecForeignCopyIn_function) (ResultRelInfo *rinfo,
+		TupleTableSlot **slots,
+		int nslots);

Also docs refer to old function names:

+void
+BeginForeignCopyIn(ModifyTableState *mtstate,
+                   ResultRelInfo *rinfo);

I think that it'd be better to choose either of these two naming schemes 
and use it everywhere for consistency.

> 
> Any thoughts on the taking out the refactoring changes out of the main
> patch as I suggested?
> 

+1 for splitting the patch. It was rather difficult for me to 
distinguish changes required by COPY via postgres_fdw from this 
refactoring.

Another ambiguous part of the refactoring was in changing 
InitResultRelInfo() arguments:

@@ -1278,6 +1280,7 @@ InitResultRelInfo(ResultRelInfo *resultRelInfo,
  				  Relation resultRelationDesc,
  				  Index resultRelationIndex,
  				  Relation partition_root,
+				  bool use_multi_insert,
  				  int instrument_options)

Why do we need to pass this use_multi_insert flag here? Would it be 
better to set resultRelInfo->ri_usesMultiInsert in the 
InitResultRelInfo() unconditionally like it is done for 
ri_usesFdwDirectModify? And after that it will be up to the caller 
whether to use multi-insert or not based on their own circumstances. 
Otherwise now we have a flag to indicate that we want to check for 
another flag, while this check doesn't look costly.


Regards
-- 
Alexey Kondratov

Postgres Professional https://www.postgrespro.com
Russian Postgres Company