Did you have the need to sometimes split an SQL extra large files to few pieces in order to be able to later upload it via phpmyadmin?
Did you needed an extra large video or data file to be cut in few pieces in order to transfer it in few pieces over an USB stick?
Or just to give you an another scenario where I sometimes need to have an enormous file let’s say 3G split in few pieces, in order to later read it in vim or mcedit .
I sometimes need to achieve this on FreeBSD and Linux hosts thus I thought it will be helpful to somebody to give a very quick tutorial on the way large files can be cut in pieces on Linux and BSD hosts.
GNU/Linux and FreeBSD are equipped with the split command. The purpose of this command is exactly the cutting of a file to a number of pieces.
On Linux the split command comes by default install to the system with the coreutils package on most Debian (deb) based and Redhat based (rpm) distributions, theerefore Linux’s version of split is GNU/split since it’s part of the GNU Coreutils package. An interesting fact about Linux split is that one of the two programmers who has coded it is Richard Stallman 😉
On BSD Unix split is the AT&T UNIX (BSD) split
In the past splitting files in pieces was much more needed than today, as people used floppy drives to transfer data, though today with the bloom of Internet and the improve of the data carriers transferring even an extra large files from one place to another is a way more trivial task still at many occasions splitting it in pieces is needed.
Even though today splitting file is very rarely required, still there are times when being able to split a file in X number of parts is very much needed.
Maybe the most common use of splitting a file today is necessery when a large SQL file dumps, like let’s say 200 MBytes of info database needs to be moved from ane hosting provider to another one.
Many hosting providers does disallow direct access with standard mySQL client programs to the database directly and only allow a user to connect only via phpMyAdmin or some other web interface like Cpanel to improve data into the SQL or PostgreSQL server.
In such times, having knowledge on the Unix split command is a priceless asset.
Even though on Linux and BSD the code for the split command is not identical and GNU/split and BSD/split has some basic differences, the use of split on both of these Unices is identical.
The way to split a file in few pieces using on both Linux and BSD OSes is being done with one and the same command, here is how:
1. Splitting file in size of 40 mb On Linux
linux:~# split -b 40m SQL-Backup-Data.sql SQL-Backup-Data_split
2. Splitting file in size of 40mb on BSD (FreeBSD, OpenBSD, NetBSD)
freebsd# split -b 40m SQL-Backup-Data.sql SQL-Backup-Data_split
The Second argument the split command takes is actually called a prefix, the prefix is used as a basis name for the creation of the newly generated files cut in pieces file based on SQL-Backup-Data.sql.
As I said identical command will split the SQL-Backup-Data.sql files in a couple of parts which of it will be sized 40 megas.
These command will generate few files output like:
freebsd# ls -1 SQL-Backup-Dat*SQL-Backup-Data.sql
As you see the SQL-Backup-Data.sql with size 200MB is being split in four files each of which is sized 40mbytes.
After the files are transfered to another Linux or BSD host, they can easily be again united in the original file with the command:
linux:~# for i in $(ls -1 SQL-Backup-Data_split*); echo $i >> SQL-Backup-Data.sql
Alternatively in most Unices also using cat should be enough to collect back the pieces into the original file, like so:
freebsd# cat SQL-Backup-Data_split* >> SQL-Backup-Data.sql