Usage of the Linux Text Search Command grep

grep (Global Regular Expression Print) is a powerful and commonly used text search tool in Linux/Unix systems, used to search for lines that match a specified pattern (usually a regular expression) in files or standard input, and outputs the matching lines to standard output.

1. Basic Syntax

grep [options] pattern [file...]

• Pattern: The string or regular expression to search for.
• File: The list of files to search. If omitted, it reads from standard input.

2. Common Options Explained

Option	Description
`<span>-i</span>`	Ignore case (case-insensitive)
`<span>-v</span>`	Invert match, show non-matching lines
`<span>-n</span>`	Show line numbers of matching lines
`<span>-c</span>`	Only output the count of matching lines
`<span>-l</span>`	Only show filenames containing matches (for multiple files)
`<span>-L</span>`	Only show filenames not containing matches
`<span>-w</span>`	Whole word match, avoiding partial matches (e.g., “the” does not match “other”)
`<span>-x</span>`	Whole line match (the entire line must match the pattern)
`<span>-r</span>` or `<span>-R</span>`	Recursive search in directories (including subdirectories)
`<span>-A n</span>`	Show matching lines and the following n lines (After)
`<span>-B n</span>`	Show matching lines and the preceding n lines (Before)
`<span>-C n</span>`	Show matching lines and n lines before and after (Context)
`<span>-E</span>`	Use extended regular expressions (equivalent to `<span>egrep</span>`)
`<span>-F</span>`	Treat pattern as a fixed string (do not use regular expressions, equivalent to `<span>fgrep</span>`)
`<span>-o</span>`	Only output the matching part (not the entire line)
`<span>--color=auto</span>`	Highlight matching content (usually enabled by default)
`<span>-h</span>`	Do not show filenames when searching multiple files
`<span>-H</span>`	Force show filenames (even when searching a single file)

3. Basic Usage Examples

1. Simple String Search

grep "error" /var/log/syslog

Search for lines containing “error” in /var/log/syslog.

2. Ignore Case

grep -i "ERROR" /var/log/syslog

Matches “error”, “Error”, “ERROR”, etc.

3. Show Line Numbers

grep -n "main" script.py

Outputs lines containing “main” along with their line numbers.

4. Whole Word Match

grep -w "is" file.txt

Matches the word “is”, but does not match “this” or “list”.

5. Invert Match

grep -v "^#" config.conf

Displays all non-comment lines (lines not starting with #).

6. Recursive Search in Directories

grep -r "TODO" ./src/

Recursively search for files containing “TODO” in the ./src/ directory and its subdirectories.

7. Show Context

grep -C 2 "exception" app.log

Displays matching lines along with 2 lines before and after.

8. Only Output Matching Content

echo "Email: [email protected]" | grep -o "[a-zA-Z0-9._%+-]*@example\.com"
# Output: [email protected]

9. Count Matching Lines

grep -c "success" access.log

10. Match Multiple Patterns

Find lines containing “error” or “warning” or “critical”.

grep "error\|warning\|critical" /var/log/messages

Or use grep -E

grep -E "error|warning|critical" /var/log/messages

Or use egrep

egrep "error|warning|critical" /var/log/messages

11. Filter File Types

When recursively searching large directories, you can combine --include or --exclude to filter file types:

grep -r --include="*.py" "def " .

Only search for lines containing “def ” in .py files.

4. Regular Expression Support

grep uses Basic Regular Expressions (BRE) by default, while grep -E uses Extended Regular Expressions (ERE).

Common Regular Symbols (BRE/ERE):

Symbol	BRE/ERE	Meaning
`<span>.</span>`	Matches any single character
`<span>*</span>`	Matches the previous character 0 or more times
`<span>^</span>`	Line start anchor
`<span>$</span>`	Line end anchor
`<span>[abc]</span>`	Matches a, b, or c
`<span>[^abc]</span>`	Matches any character not a, b, or c
`<span>\?</span>`	BRE	Matches the previous character 0 or 1 time (optional)
`<span>?</span>`	ERE	Matches the previous character 0 or 1 time (optional)
`<span>\+</span>`	BRE	Matches the previous character 1 or more times
`<span>+</span>`	ERE	Matches the previous character 1 or more times
`<span>\ \|</span>`	BRE	Or operation
`<span>\|</span>`	ERE	Or operation
`<span></span>`	BRE	Grouping, `<span>$ab$\{2\}</span>` matches “abab”
`<span>()</span>`	ERE	Grouping, `<span>(ab){2}</span>` matches “abab”
`<span>\{n,m\}</span>`	BRE	Matches the previous character n to m times
`<span>\{n\}</span>`	BRE	Matches exactly n times
`<span>\{n,\}</span>`	BRE	Matches at least n times
`<span>{n,m}</span>`	ERE	Matches the previous character n to m times
`<span>{n}</span>`	ERE	Matches exactly n times
`<span>{n,}</span>`	ERE	Matches at least n times

In Basic Regular Expressions (BRE), +, ?, |, (), and {} need to be escaped to be used as metacharacters.

5. Combining with Other Commands

1. Combining with Pipe

ps aux | grep "nginx"

Find running nginx processes.

2. Combining with find

find . -name "*.log" -exec grep -l "404" {} \;

Find all .log files containing “404”.

6. Common Variants

Command	Description
`<span>grep</span>`	Basic regular expressions
`<span>egrep</span>`	Equivalent to `<span>grep -E</span>`, supports extended regular expressions
`<span>fgrep</span>`	Equivalent to `<span>grep -F</span>`, treats pattern as a fixed string (faster, does not parse regex)

Modern systems recommend using grep -E or grep -F directly, rather than egrep or fgrep.

7. Conclusion

grep is the Swiss Army knife of text processing; mastering its options and regular expression usage can greatly enhance efficiency in handling tasks such as processing logs, code, and configuration files in the command line.