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

Summary

Arrays are a powerful feature of the shell that can be used in many different ways. They expand the power of the shell quite considerably, although requiring arrays as a supported feature restricts the portability of a shell script to systems that support arrays. Sparse arrays can be very useful, too, with none of the memory management necessary for lower-level languages such as C.

Associative arrays are even more flexible than regular arrays. They are, in effect, like sparse arrays with a name instead of a number as an index. This allows you to store metadata within the data’s key itself. The ability to treat each element in an array as an individual string and use the more advanced bash features on those strings makes for great levels of flexibility, where one would otherwise reach for perl, awk, or sed.

The next chapter looks at processes and how the kernel manages them. This covers foreground and background processes, signals, exec, as well as I/O redirection and pipelines. The kernel is responsible for controlling all the processes on the system, how they run and how they interact with each other. The Linux kernel also exposes a lot of its internal workings via the /proc pseudo-filesystem, which is a very useful way for shell scripts to get access to internal kernel data structures.