Shell Scripting: Expert Recipes for Linux, Bash, and More

split

When managing and transferring files, there are often limits to the size of files that can be transferred. Whether it’s e-mail systems with 2MB or 10MB limits, or FAT filesystems with 4GB limits, or saving large amounts of data to 4.7GB DVDs, there are always problems when you are used to dealing with a flexible and powerful OS that is perfectly capable of handling much larger files than this, and then discover that a lot of the mainstream infrastructure is not as powerful as a modern Linux or Unix system.

The solution to this problem is the split utility; although it can split files based on line numbers, it is most often used to split a large file into smaller, regular chunks. The -b flag tells it to work in this way. You can specify the chunk size in K, M, G, T, P, E, Z, and even Y. These are in multiples of 1,024; adding a B suffix (KB, MB, GB, and so on) will work in multiples of 1,000.

The default options are quite widely recognized in that it creates a sequence of files called xaa, xab, xac, and so on (until it gets to xax, xay, xaz, xba, xbb, xbc, xbd . . . ). However, it does have some more friendly options; the -d flag uses digits instead of letters — x01, x02, x03 instead of xaa, xab, xac — and you can also define your own prefix instead of the letter x. It is a good habit to add an underscore to the end of the prefix, so that you can more easily see the suffix.

The default size of the suffix is two characters; if the file will be broken into more than 100 chunks, ...