How to split a 4GB SQL file into queries using the command line

Question

I am trying to split a 4GB SQL file into smaller files so that I can import them one by one to reduce errors.

Each INSERT statement starts on a new line and is preceded by );

I have tried using split -p \0\;INSERT but this does not work.

I have read about csplit however I am still unssure as to what RegularExpression is needed.

There are 117 tables across the 4GB of data.

A sample INSERT statement is as follows:

INSERT INTO 'puid' VALUES (2196534,'ee1d9e3f-b3ae-e36b-cfea-2499a18bd86e',163);
INSERT INTO 'puid' VALUES (2196535,'e2c05e1c-73bd-cdc7-0bab-610b6afe4bd7',163);

One `INSERT` statement is just over 1MB in size and is too large to post here. — Michael, Oct 21 '13 at 10:39
You can just trim most of it, and show minimal examples. Such as `INSERT INTO table ('test');` or something — justhalf, Oct 21 '13 at 10:40
`INSERT INTO 'puid' VALUES (2196534,'ee1d9e3f-b3ae-e36b-cfea-2499a18bd86e',163);` `INSERT INTO 'puid' VALUES (2196535,'e2c05e1c-73bd-cdc7-0bab-610b6afe4bd7',163);` — Michael, Oct 21 '13 at 10:43
If that's the case, it's quite simple to do just like written in this answer, no? http://stackoverflow.com/questions/7764755/unix-how-to-split-a-file-into-equal-parts-without-breaking-individual-lines?rq=1 — justhalf, Oct 21 '13 at 10:53

score 1 · Accepted Answer · answered Oct 21 '13 at 10:56

1

Have you tried using split -l <line_count>? That'll split on line boundaries.

answered Oct 21 '13 at 10:56

Marcelo Cantos

181,030
38
327
365

How to split a 4GB SQL file into queries using the command line

1 Answers1