Pipes & Redirection

stdin/stdout/stderr, pipes, here-docs, process substitution, and tee for multi-destination output.

Overview

Overview Every process inherits three file descriptors: 0 (stdin), 1 (stdout), 2 (stderr). Redirection changes where these point.

Redirection Operators

> file — redirect stdout to file (truncating)
>> file — append stdout to file
2> file — redirect stderr to file
2>&1 — merge stderr into stdout
&> file — redirect both stdout and stderr to file

Pipes and Subshells

The pipe operator | connects stdout of the left command to stdin of the right. Commands in a pipeline run in subshells — variable assignments in a pipeline do not affect the parent shell.

tee and Named Pipes

tee copies stdin to both stdout and a file:

Named pipes (FIFOs) created with mkfifo allow producer-consumer patterns between separate processes.

Key concepts

File descriptors 0/1/2: stdin (0), stdout (1), stderr (2). Redirection changes which file or pipe these descriptors point to.
2>&1: Redirect stderr (fd 2) to wherever stdout (fd 1) currently points. Order matters: cmd > file 2>&1 is correct; cmd 2>&1 > file sends stderr to terminal.
Pipeline subshell: Each command in a pipeline runs in a subshell. Variable assignments inside pipelines do not affect the parent shell.
Process substitution: <(cmd) substitutes command output as a named pipe. Enables comparing live output without temp files.
tee: Reads stdin and writes to both stdout and one or more files. tee -a appends instead of truncating.

When to use it

Building complex data-processing pipelines in shell scripts
Explaining why variables set inside a pipeline do not persist after the pipeline
Logging output while still displaying it to the terminal

Interview questions

Why does this script not work: count=0; cat file.txt | while read line; do count=$((count+1)); done; echo $count?
You need to log all output of a script to a file while still displaying it on the terminal. How?