Re: Reg: BULK COLLECT - Mailing list pgsql-general

From Allan Kamau
Subject Re: Reg: BULK COLLECT
Date
Msg-id CAF3N6oQEK6=mBR7r4sZiBuFusiRMfPoyZ7dSHqzs_yjfKtBAMA@mail.gmail.com
Whole thread Raw
In response to Reg: BULK COLLECT  (Medhavi Mahansaria <medhavi.mahansaria@tcs.com>)
List pgsql-general
It seems you are fetching from a table then sequentially inserting each record to another table.
In PostgreSQL, you could use cursors in PL/pgSQL ("http://www.postgresql.org/docs/9.4/interactive/plpgsql-cursors.html").

Alternatively you may write a single query which selects from the table and performs an insert into the other table, it is usually more efficient than sequential single record inserts.

If the amount of data are significant you may use ORDER BY and LIMIT in your SELECT query in order to limit the count of records you are fetching for insertion.
You may place this SQL query in a loop within a bash script as shown below, this bash script and the pseudo query updates a few records of a given table at a time, you may convert this example into an INSERT query for piecemeal inserts keeping the transaction size manageable.


#!/bin/sh
#date;time sh scripts/sample_piecemeal_db_updation.sh 10000 2>&1|tee /tmp/sample_piecemeal_db_updation.sh.sh.log;date;
export PSQL_PATH="/apps/postgreSQL/pgsql-9.1.4/bin/psql";
export DB_NAME="my_db";
export PGPORT="5432";
export DB_USER="some_user";
export DB_HOST="some_ip"; #put the actual ip here.
step=$1;

echo `date +%Y-%m-%d.%H%Mhrs:%S.%N`'-DB update commences';
trim() { echo $1; }

        _sql_query2="
            SELECT count(a.*)AS cnt FROM main_table a;
        ";
        echo $_sql_query2;
        counter=0;
        for cnt in `$PSQL_PATH -d$DB_NAME -U$DB_USER --port=$PGPORT --host=$DB_HOST --quiet --tuples-only -c"$_sql_query2"`
        do
            echo '$cnt is:'$cnt;
            counter=$(( $counter + 1 ));
            for i in $(seq 0 $step $cnt)
            do
                _sql_query1="
                    SET work_mem='1GB';
                    UPDATE main_table b
                    SET some_field=a.some_field
                    FROM
                    (
                    SELECT
                    a.id
                    ,b.some_field2
                    FROM
                    (
                    SELECT
                    a.*
                    FROM
                    main_table a
                    ORDER BY a.id ASC
                    OFFSET $i
                    LIMIT $step
                    )a
                    JOIN
                    some_other_table b
                    ON
                    b.some_other_field=a.some_other_field
                    )a
                    WHERE
                    b.id=a.id
                    ;
                ";
                #echo "_sql_query1 is:$_sql_query1";
                results=`$PSQL_PATH -d$DB_NAME -U$DB_USER --port=$PGPORT --host=$DB_HOST --quiet --tuples-only -c"$_sql_query1"`
                echo "cnt is:$cnt, step is:$step, i is:$i, results:$results";
            done;
        done;
echo `date +%Y-%m-%d.%H%Mhrs:%S.%N`'-DB updation concludes.';


Regards,
Allan.

On Mon, May 25, 2015 at 3:24 PM, Medhavi Mahansaria <medhavi.mahansaria@tcs.com> wrote:
Hello,

I am porting my application from Oracle to PostgreSQL.

We are using BULK COLLECT functionality of oracle.
How can i change the 'BULK COLLECT' fetch of the data from the cursor to make if compatible for pg/plsql?

A small example is as below (This is just an example and the query is much more complex which returns huge amount of data)


CREATE OR REPLACE FUNCTION abc() RETURNS VOID AS $body$


DECLARE
l_data b%ROWTYPE;

POPULATE_STATS CURSOR IS
(
SELECT * from a
)
;                        // query returning a huge amount of data

BEGIN
      OPEN POPULATE_STATS;
      LOOP
            FETCH POPULATE_STATS BULK COLLECT INTO l_data LIMIT 1000;
        IF POPULATE_STATS%ROWCOUNT > 0
        THEN
        FORALL i IN 1..l_data.COUNT
                INSERT INTO b VALUES l_data(i);
        END IF;
      IF NOT FOUND THEN EXIT; END IF;
      END LOOP;
      CLOSE POPULATE_STATS;
EXCEPTION
      WHEN OTHERS THEN
      CODE := SQLSTATE;
      MSG := SQLERRM;
     INSERT INTO tracker VALUES (CODE,MSG,LOCALTIMESTAMP);
      RAISE NOTICE 'SQLERRM';
      RAISE NOTICE '%', SQLSTATE;
      RAISE NOTICE '%', MSG;
END;

$body$
LANGUAGE PLPGSQL;

How can i change the 'BULK COLLECT' fetch of the data from the cursor to make if compatible for pg/plsql?


Thanks & Regards
Medhavi Mahansaria
Mailto: medhavi.mahansaria@tcs.com

=====-----=====-----=====
Notice: The information contained in this e-mail
message and/or attachments to it may contain
confidential or privileged information. If you are
not the intended recipient, any dissemination, use,
review, distribution, printing or copying of the
information contained in this e-mail message
and/or attachments to it are strictly prohibited. If
you have received this communication in error,
please notify us by reply e-mail or telephone and
immediately and permanently delete the message
and any attachments. Thank you


pgsql-general by date:

Previous
From: Piotr Gasidło
Date:
Subject: Re: Strange replication problem - segment restored from archive but still requested from master
Next
From: Adrian Klaver
Date:
Subject: Re: Reg: BULK COLLECT