Copy command using filter

If requirement is only to skip top 500 records everytime , You can use SKIP with copy command , it will  skip top 500 records always means 1 -500 .

copy abc from /temp/abc.txt delimiter ';' SKIP 500  ;

--> above command will skip top 500 records .

copy abc from /temp/abc.txt delimiter ';' SKIP bytes 500  ;

--> This command will skip 500 bytes from starting of file.

Thanks :

Abhi

 You can do this using pipes in linux and the stdin copy option.

- just pipe the output of the tail or any linux command that can limit the numbers of rows, form head or tail to your copy command.

 - you can put more logic in linux scripting then in the copy command.

 See example:

[dbadmin@~]$ cat /tmp/file.csv
1,1,2
2,1,2
3,1,2
4,1,2
[dbadmin@~]$ tail -2 /tmp/file.csv
3,1,2
4,1,2
[dbadmin@~]$ tail -2 /tmp/file.csv |vsql -d db -U dbadmin -c "COPY test FROM LOCAL STDIN DELIMITER ',' ;"
Password:
Rows Loaded
-------------
2
(1 row)

[dbadmin@~]$ vsql -d db -U dbadmin -c "select * from test;"
Password:
 col1 | col2 | col3
------+------+------
 3 | 1 | 2
 4 | 1 | 2
(2 rows)

 not sure i understand your requirment !