Essential Ripgrep Search Patterns

Beginnerv1.0.0

Master everyday ripgrep search patterns — literal search, regex, file type filtering, context lines, and finding code patterns efficiently in any project.

Content

Overview

Ripgrep (rg) is the fastest code search tool. It respects .gitignore, searches in parallel, and uses smart defaults. Learn the essential patterns that cover 90% of daily code search needs.

Why This Matters

-Speed — 10-100x faster than grep on large codebases
-Smart defaults — ignores .gitignore patterns, binary files, hidden dirs
-Regex — powerful pattern matching for code constructs
-Integration — works with editors, CI, and shell pipelines

How It Works

Step 1: Basic Search

bash

# Simple text search
rg "TODO"
rg "console.log"

# Case-insensitive
rg -i "error"

# Whole word (avoid partial matches)
rg -w "user"          # matches "user" not "username"

# Fixed string (no regex interpretation)
rg -F "file.ts"       # treats dots as literal, not regex
rg -F "$variable"     # treats $ as literal

# Search specific directory
rg "pattern" src/
rg "pattern" src/ tests/    # multiple directories

Step 2: File Type Filtering

bash

# Search only TypeScript files
rg -t ts "useState"

# Search only Python files
rg -t py "import requests"

# Exclude test files
rg -T test "deprecated"

# Multiple types
rg -t ts -t js "fetch"

# Custom glob
rg -g "*.config.*" "port"
rg -g "!*.test.*" "function"    # exclude pattern

# List available types
rg --type-list

Step 3: Context & Output Control

bash

# Lines around matches
rg -C 3 "panic!"              # 3 lines before and after
rg -B 2 "FIXME"               # 2 lines before
rg -A 5 "function main"       # 5 lines after

# File list only (no content)
rg -l "TODO"                   # files with matches
rg -l "TODO" | wc -l           # count files

# Count matches
rg -c "console.log"            # matches per file
rg -c "console.log" --sort-files

# Suppress filename
rg -N "pattern"                # no line numbers
rg --no-filename "pattern"     # no filenames (just content)

# Max results
rg -m 5 "TODO"                 # first 5 matches per file
rg "TODO" | head -20           # first 20 total matches

Step 4: Common Search Patterns

bash

# Find function definitions
rg "function w+" --type ts
rg "def w+" --type py
rg "func w+" --type go

# Find imports
rg "^import" --type ts
rg "^from .* import" --type py
rg "require\(" --type js

# Find TODO/FIXME/HACK
rg "(TODO|FIXME|HACK|XXX)"

# Find console.log (for cleanup)
rg "console\.(log|debug|warn)" --type ts

# Find environment variable usage
rg "process\.env\." --type ts
rg "os\.environ" --type py

Best Practices

-Use -w for identifier searches (avoids partial word matches)
-Use -t for type filtering (cleaner than glob patterns)
-Use -F for literal strings containing regex chars (. $ * etc)
-Use -l to get file list for batch operations
-Use --sort-files for reproducible output order

Common Mistakes

-Searching without -w (too many partial matches)
-Not using -F for literal strings ("file.ts" matches "filexts" without -F)
-Forgetting ripgrep auto-ignores .gitignore patterns (search missed files)
-Not using -t for type filtering (noisy results from wrong file types)
-Using -i when not needed (slower, more false positives)

FAQ

Discussion

Loading comments...

How It Works

Step 1: Basic Search

bash

# Simple text search
rg "TODO"
rg "console.log"

# Case-insensitive
rg -i "error"

# Whole word (avoid partial matches)
rg -w "user"          # matches "user" not "username"

# Fixed string (no regex interpretation)
rg -F "file.ts"       # treats dots as literal, not regex
rg -F "$variable"     # treats $ as literal

# Search specific directory
rg "pattern" src/
rg "pattern" src/ tests/    # multiple directories

Step 2: File Type Filtering

bash

# Search only TypeScript files
rg -t ts "useState"

# Search only Python files
rg -t py "import requests"

# Exclude test files
rg -T test "deprecated"

# Multiple types
rg -t ts -t js "fetch"

# Custom glob
rg -g "*.config.*" "port"
rg -g "!*.test.*" "function"    # exclude pattern

# List available types
rg --type-list

Step 3: Context & Output Control

bash

# Lines around matches
rg -C 3 "panic!"              # 3 lines before and after
rg -B 2 "FIXME"               # 2 lines before
rg -A 5 "function main"       # 5 lines after

# File list only (no content)
rg -l "TODO"                   # files with matches
rg -l "TODO" | wc -l           # count files

# Count matches
rg -c "console.log"            # matches per file
rg -c "console.log" --sort-files

# Suppress filename
rg -N "pattern"                # no line numbers
rg --no-filename "pattern"     # no filenames (just content)

# Max results
rg -m 5 "TODO"                 # first 5 matches per file
rg "TODO" | head -20           # first 20 total matches

Step 4: Common Search Patterns

bash

# Find function definitions
rg "function w+" --type ts
rg "def w+" --type py
rg "func w+" --type go

# Find imports
rg "^import" --type ts
rg "^from .* import" --type py
rg "require\(" --type js

# Find TODO/FIXME/HACK
rg "(TODO|FIXME|HACK|XXX)"

# Find console.log (for cleanup)
rg "console\.(log|debug|warn)" --type ts

# Find environment variable usage
rg "process\.env\." --type ts
rg "os\.environ" --type py

Common Mistakes

-Searching without -w (too many partial matches)

-Not using -F for literal strings ("file.ts" matches "filexts" without -F)

-Forgetting ripgrep auto-ignores .gitignore patterns (search missed files)

-Not using -t for type filtering (noisy results from wrong file types)

-Using -i when not needed (slower, more false positives)