grep is one of the most powerful and flexible commands in Linux. Whether you’re searching through code, analyzing logs, or filtering outputs in pipelines, grep can save you hours of manual work. It’s fast, supports regular expressions, and works beautifully with other Unix tools. In this guide, we’ll look at 10 practical and advanced grep examples that will take your searching skills to the next level. Each command is explained with clear use-cases and real-world outputs, helping you master text searching like a pro.
1. Search for Exact Matches in Files
This is the most basic use of grep—search for a literal string inside a file.
$ grep "main()" /home/myuser/code/main.c
int main() {
It shows lines containing the string “main()”. grep highlights results when your terminal supports it.
2. Search Recursively in a Directory
Use -r or -R to search through all files under a directory recursively.
$ grep -R "api_key" /home/myuser/projects /home/myuser/projects/config/settings.py:api_key = "XYZ123"
This is very useful when you’re looking for variable names or constants across a codebase.
3. Show Line Numbers with Matches
To include the line number where the match appears, use the -n flag.
$ grep -n "server" config.yaml 42: server: "192.168.1.100"
This is helpful when you’re editing files based on search results.
4. Search Case-Insensitive
The -i flag makes the search case-insensitive.
$ grep -i "error" /var/log/syslog Jul 6 14:32:01 server CRON[29219]: (root) CMD (some_script.sh) — Error occurred
Great for searching logs or user input where case may vary.
5. Count Matching Lines
Use -c to count how many lines matched the pattern.
$ grep -c "ERROR" /var/log/app.log 58
This is ideal for summarizing log data or checking how many times an event occurred.
6. Show Only the Matching Part of the Line
-o extracts only the part of the line that matches the pattern—not the full line.
$ grep -o "http[s]*://[^ ]*" urls.txt https://example.com http://testsite.net
This is excellent when you’re extracting URLs, IP addresses, or any pattern using regex.
7. Use Regular Expressions
grep supports full regular expressions by default, or use grep -E for extended regex (equivalent to egrep).
$ grep -E "error|fail|fatal" /var/log/app.log 2025-07-06 12:01:12 - fatal: service failed to start
This allows you to match multiple patterns in a single command.
8. Invert the Match (Exclude Lines)
Use -v to exclude lines that match a pattern. It inverts the result.
$ grep -v "DEBUG" app.log > app_no_debug.log
This is helpful when you want to filter out noise, like debug lines in logs.
9. Display Lines Before or After a Match
Use -A, -B, or -C to show context around matches.
$ grep -C 2 "panic" server.log 123: Starting service... 124: Connecting to database 125: panic: unexpected connection failure 126: Restarting service... 127: Alert triggered
-C 2 shows 2 lines before and after. Use -A or -B for just after or before, respectively.
10. Combine with Other Commands Using Pipes
grep becomes even more powerful when combined with other commands like ps, netstat, or ls using pipes.
$ ps aux | grep firefox myuser 3192 5.0 3.2 998344 132000 ? Sl 10:10 0:40 /usr/lib/firefox/firefox
This technique lets you filter real-time output to locate exactly what you need.
Bonus: Highlight Pattern Matches
Most modern versions of grep highlight matches in color when outputting to a terminal. If it’s not working, try adding --color=auto.
$ grep --color=auto "auth" auth.log
This visually helps to spot matches quickly in a large chunk of output.
Conclusion
grep is much more than a basic search command—it’s a versatile powerhouse that can handle everything from simple string searches to complex pattern matching across massive files. Whether you’re debugging logs, analyzing code, or filtering structured output from pipelines, knowing how to wield grep effectively can dramatically boost your productivity in the Linux terminal.