Linux/Unix: Searching, Text Processing & Regex – grep, find, sed, awk, and Pattern Matching

Introduction – Supercharge Your Shell with Search & Pattern Tools

In Linux/Unix, the real power of the command line lies in text processing and file searching. Whether you’re filtering logs, extracting data, or matching patterns with regular expressions, tools like grep, find, awk, and sed turn simple commands into data-processing powerhouses.

In this guide, you’ll learn:

• How to search inside files with grep, fgrep, and egrep
• How to locate files with find, locate, which, and whereis
• How to manipulate structured data using awk and sed
• How to write and apply regular expressions for powerful matching

Topics Covered

Linux/Unix: Text Search (grep, fgrep, egrep)

grep – Search Lines for a Pattern

grep "error" logfile.txt

Displays lines containing “error”.

-i, -n, and -r Options

grep -i "warning" logfile.txt   # Case-insensitive
grep -n "fail" logfile.txt      # Show line numbers
grep -r "config" /etc           # Recursive directory search

fgrep – Literal Search (No Regex)

fgrep "a.b" file.txt

Searches for exact string a.b, not regex.

egrep – Extended Regex Support

egrep "error|fail" logs.txt

Matches either “error” or “fail”.

Linux/Unix: File Location (find, locate, which, whereis)

find – Search Files by Name, Type, Time

find /home -name "*.txt"

Finds all .txt files recursively.

find /etc -type f -size +1M

Finds files over 1MB.

locate – Fast Filename Search (uses database)

locate passwd

Run updatedb to refresh the database.

which & whereis – Command Location

which python3     # Shows full path
whereis python3   # Shows binary, source, man

Linux/Unix: Text Processing (awk, sed)

awk – Process Fields in Files

awk '{print $1, $3}' data.txt

Prints the first and third columns.

awk -F: '{print $1}' /etc/passwd

Uses : as a field separator.

sed – Stream Editor for Replacements

sed 's/error/ERROR/g' logfile.txt

Replaces “error” with “ERROR” in every line.

sed -n '2,4p' file.txt

Prints only lines 2 through 4.

Linux/Unix: Regex Concepts & Practice

Basic Metacharacters

Regex Example with grep -E

grep -E "^[A-Z].*\.log$" files.txt

Matches lines that start with a capital letter and end in .log.

Practice Regex with grep or sed

grep -E "user[0-9]{2}" names.txt
sed -E 's/[0-9]+/NUM/' numbers.txt

Summary – Recap & Next Steps

The tools in this guide are the foundation of shell data manipulation. When combined, they allow for automated parsing, dynamic reporting, and intelligent searching—right from the terminal.

Key Takeaways:

• Use grep, fgrep, and egrep for file content filtering
• Use find, locate, which to locate files or commands
• Use awk and sed to process fields, rows, and text replacements
• Regular expressions let you match patterns and automate logic

Real-World Applications:

• Log parsing for DevOps and monitoring
• Data extraction and reporting from structured files
• Finding configuration files or scripts across directories

Frequently Asked Questions

What’s the difference between grep, fgrep, and egrep?
grep uses basic regex, fgrep searches fixed strings (no regex), and egrep allows extended regex.

How do I search all files for a keyword recursively?
Use:

grep -r "keyword" .

Can I use find to delete files?
Yes:

find . -name "*.log" -delete

How do I replace text in-place using sed?
Use:

sed -i 's/foo/bar/g' file.txt

What’s the best way to count unique values in a column?
Combine awk, sort, and uniq:

awk '{print $2}' data.txt | sort | uniq -c