Search and Filter

Working Code

Example 1: Searching text with grep

Start with a file to search:

cat Documents/notes.md

Output:

# Notes
- item 1
- item 2

Find lines containing a specific word:

grep "item" Documents/notes.md

Output:

- item 1
- item 2

Show line numbers too:

grep -n "item" Documents/notes.md

Output:

2:- item 1
3:- item 2

Example 2: Case-insensitive search

grep -i "hello" Documents/hello.txt

Output:

Hello, World!

The -i option ignores case. It matches Hello, hello, and HELLO alike.

Example 3: Chaining commands with pipes

# Filter ls output for a specific pattern
ls | grep ".md"

Output:

notes.md

# Count items in a file listing
ls | wc -l

Output:

       4

# 3-stage pipe: ls -> grep -> wc
ls | grep ".txt" | wc -l

Output:

       1

Try It Yourself

grep with Various Options

# Search across multiple files
grep "Hello" Documents/hello.txt Documents/notes.md

# Recursive search through subdirectories
grep -r "item" Documents/

# Show only filenames (not content)
grep -rl "Hello" .

# Show match count only
grep -c "item" Documents/notes.md

Output:

2

# Show non-matching lines (inverse search)
grep -v "item" Documents/notes.md

Output:

# Notes

Finding Files with find

# Find files by name
find . -name "*.txt"

Example output:

./Documents/hello.txt
./Downloads/readme.txt

# Filter by type (f=file, d=directory)
find . -type f -name "*.md"

# Files modified within the last day
find . -mtime -1

# Files larger than a specific size
find . -size +1k

# Run a command on each found file (-exec)
find . -name "*.txt" -exec cat {} \;

"Why?" — Why Pipes Are Powerful

The pipe (|) connects one command's stdout to the next command's stdin.

ls -la | grep ".txt" | wc -l
   |          |           |
file list  .txt filter  count lines

Step by step:

1. ls -la — list files
2. grep ".txt" — keep only lines containing .txt
3. wc -l — count remaining lines

Result: the number of .txt files in the current directory.

The core principle of pipes: combine simple tools to tackle complex tasks. Each command does one thing well, and pipes connect them to solve bigger problems.

grep Options Summary

| Option | Description | Example | | ------ | ---------------------- | ----------------- | | -i | Ignore case | grep -i "hello" | | -n | Show line numbers | grep -n "error" | | -r | Include subdirectories | grep -r "bug" . | | -l | Show filenames only | grep -rl "todo" | | -v | Exclude matches | grep -v "#" | | -c | Count matching lines | grep -c "error" |

grep and Pipes in Practice

# Find errors in a log
cat app.log | grep "ERROR"

# Count error lines
cat app.log | grep "ERROR" | wc -l

# View everything except debug lines
cat app.log | grep -v "DEBUG"

# Find a function in source code
grep -rn "getUserById" src/

Common Mistakes

Mistake 1: Special characters in grep patterns

# . means "any character" in regex
grep "file.txt" .    # matches file.txt AND file_txt

# To search literally, escape it
grep "file\.txt" .
# Or use -F (fixed string)
grep -F "file.txt" .

Mistake 2: Searching a directory without -r

grep "text" Documents/   # error
# grep: Documents/: Is a directory

grep -r "text" Documents/  # correct

Mistake 3: Piping find results to grep incorrectly

# Wrong: this searches filenames, not file contents
find . -name "*.txt" | grep "hello"   # finds files with "hello" in the name

# Correct: search file contents
grep -r "hello" --include="*.txt" .

Deep Dive

# ^ : start of line
grep "^-" Documents/notes.md

# $ : end of line
grep "2$" Documents/notes.md

# . : any single character
grep "ite.1" Documents/notes.md

# []: one character from a set
grep "[12]" Documents/notes.md

# Extended regex (-E)
grep -E "item [12]" Documents/notes.md

# Delete all .log files
find . -name "*.log" -exec rm {} \;

# Print contents of all .txt files
find . -name "*.txt" -exec cat {} \;

# Save found file list to a file (using redirection)
find . -name "*.md" > md-files.txt

brew install ripgrep

# Basic usage (similar to grep -r)
rg "functionName" src/

# Automatically respects .gitignore
# Color output by default
# Tens of times faster than grep