Complete Feature Guide

Flexible pattern matching for any search task.

• ✓ Regex pattern matching (Rust regex engine by default)
• ✓ Literal fixed-string search (-F / --fixed-strings)
• ✓ Case-insensitive search (-i)
• ✓ Word boundary matching (-w)
• ✓ Multi-pattern search (-e pattern1 -e pattern2)
• ✓ Inverted matching — show lines that do NOT match (-v)

# Regex search
rg 'fn \w+\(' src/

# Literal string (no regex interpretation)
rg -F 'user.name[0]' .

# Case-insensitive
rg -i 'error'

# Word boundary
rg -w 'main'

# Multiple patterns
rg -e 'TODO' -e 'FIXME' -e 'HACK'

# Inverted — lines without "test"
rg -v 'test'

Built in Rust with every optimization that matters.

• ✓ Parallel directory traversal using all CPU cores
• ✓ Memory-mapped I/O for large files
• ✓ SIMD-accelerated literal searching
• ✓ Lazy line-by-line reading — never loads full file into memory
• ✓ Intelligent skipping of binary and very large files
• ✓ Efficient UTF-8 validation built into the search

# Search the entire Linux kernel (25M+ lines) in under 100ms
rg 'EXPORT_SYMBOL' ~/linux/

# Show timing with -Stats flag
rg --stats 'TODO' .

Automatically ignores files you do not want to search.

• ✓ Respects .gitignore, .ignore, and .rgignore files
• ✓ Skips hidden files and directories by default
• ✓ Detects and skips binary files automatically
• ✓ Global ignore file support (--ignore-file)
• ✓ Override any filter with --no-ignore, -u, -uu
• ✓ Custom glob patterns (--glob "*.ts", --glob "!node_modules")

# Use your gitignore rules (default behavior)
rg 'pattern'

# Include hidden files too
rg -. 'pattern'

# Search everything — ignore all ignore files
rg -uu 'pattern'

# Restrict to TypeScript files via glob
rg --glob '*.ts' 'interface'

# Exclude a directory
rg --glob '!dist/**' 'TODO'

Search by language or file type with a single flag.

• ✓ -t / --type to include only a specific type
• ✓ -T / --type-not to exclude a type
• ✓ --type-list to see all 200+ built-in types
• ✓ Define custom types with --type-add
• ✓ Types are glob-based and composable

# Only Python files
rg 'def ' -t py

# Only Rust files
rg 'fn main' -t rust

# Exclude test files
rg 'TODO' -T test

# List all built-in types
rg --type-list

# Define a custom type and use it
rg --type-add 'web:*.{html,css,js}' -t web 'class='

Switch to the full PCRE2 engine when you need look-arounds or backreferences.

• ✓ Default: Rust regex engine — fast, Unicode-aware, no backtracking catastrophe
• ✓ PCRE2 mode (-P) unlocks look-ahead, look-behind, and backreferences
• ✓ Automatic engine selection with --auto-hybrid-regex
• ✓ Full Unicode character class support (\p{Letter}, \p{Han})
• ✓ Named capture groups ((?P...))

# Look-behind (requires PCRE2)
rg -P '(?<=fn )\w+'

# Backreference
rg -P '(\w+) \1'

# Unicode: find Chinese characters
rg '\p{Han}+'

# Named capture group
rg -P '(?P<year>\d{4})-(?P<month>\d{2})'

Control exactly how results are displayed.

• ✓ Colored output by default (file, line number, match highlighted)
• ✓ Context lines — before (-B), after (-A), or both (-C)
• ✓ Show only file names (-l) or count of matches (-c)
• ✓ JSON output (--json) for piping to other tools
• ✓ Statistics mode (--stats) — total matches, files searched, elapsed time
• ✓ Quiet mode (-q) — exit code only, no output

# 3 lines of context around each match
rg 'error' -C 3

# Only list file names with matches
rg -l 'TODO'

# Count matches per file
rg -c 'import'

# JSON output for scripting
rg --json 'pattern' | jq '.data.lines.text'

# Print stats summary
rg --stats 'fn ' src/

Search files in any encoding, not just UTF-8.

• ✓ UTF-8 by default with strict validation
• ✓ UTF-16 auto-detection via BOM
• ✓ Explicit encoding override with -E / --encoding
• ✓ Supports Latin-1, GBK, EUC-JP, Shift-JIS, and more
• ✓ --text (-a) to treat binary as text (useful for Latin-1 logs)

# Force Latin-1 encoding
rg -E latin-1 'pattern' legacy-file.txt

# Search a UTF-16 file (auto-detected if BOM present)
rg 'pattern' windows-file.txt

# Treat all files as text (including binary)
rg -a 'pattern'

Search inside archives without extracting them first.

• ✓ Enable with -z / --search-zip
• ✓ Supports: gzip (.gz), bzip2 (.bz2), xz (.xz), LZ4, Brotli, Zstandard, LZMA
• ✓ Automatically decompresses on the fly
• ✓ Works with file type filtering

# Search inside .gz log files
rg -z 'ERROR' /var/log/app.log.gz

# Search all compressed files in a directory
rg -z 'pattern' /var/log/

Preview or apply replacements with capture group support.

• ✓ -r / --replace to substitute matches in output
• ✓ Capture group references ($1, $name)
• ✓ Preview replacements before writing (pipe to file or use sd)
• ✓ Works with all filtering options

# Show what a replacement would look like
rg 'fn (\w+)\(' -r 'function $1(' src/

# Combine with capture groups
rg '(\d{4})-(\d{2})-(\d{2})' -r '$3/$2/$1' dates.txt

# Write replacement output to a new file
rg 'old_api' -r 'new_api' main.go > main_updated.go

ripgrep is designed to work with other tools and editors.

• ✓ JSON output (--json) for structured consumption by scripts
• ✓ Non-zero exit code when no matches found — scriptable in if statements
• ✓ Shell completions for bash, zsh, fish, Elvish, and PowerShell
• ✓ Configuration file (.ripgreprc or RIPGREP_CONFIG_PATH)
• ✓ Used internally by VS Code, Neovim (telescope.nvim), Helix, and many more
• ✓ Preprocessing support — pipe through any command before searching

# Use in a shell if statement
if rg -q 'TODO' src/; then
  echo "Found TODOs!"
fi

# Generate shell completions
rg --generate complete-bash >> ~/.bashrc

# Set permanent flags via config file
echo '--smart-case' >> ~/.ripgreprc
echo '--hidden' >> ~/.ripgreprc

# Preprocess PDFs before searching
rg --pre pdftotext 'keyword' docs/