Command line and power tools

ripgrep, and when grep is still right

Updated 2026-07-31 · 4 min read

If you search inside files regularly, ripgrep is one of the few tools that is simply better in almost every dimension.

Install

brew install ripgrep

Why it is faster

• Skips anything in .gitignore by default
• Skips binaries and hidden files unless told otherwise
• Searches directories in parallel
• Uses a genuinely fast regex engine with SIMD literal matching

The default exclusions matter as much as the speed: not searching node_modules is why it finishes.

The commands worth knowing

rg "searchterm"                # recursive from here
rg -i "searchterm"             # case-insensitive
rg -t py "def main"            # only Python files
rg -l "TODO"                   # filenames only
rg -C 3 "error"                # 3 lines of context
rg --hidden --no-ignore "key"  # search everything, ignore nothing

When to use grep instead

• On a machine where you cannot install anything
• In scripts that must run anywhere, since grep is guaranteed present
• For POSIX-exact behaviour in a pipeline

Contents versus names

rg searches inside files. That is a different job from finding a file by name, and it is the expensive one — every byte must be read. A filename index answers "where is X" in milliseconds; rg answers "which files mention X" in seconds. Most workflows want both, used for different questions.

Common questions

Does ripgrep search hidden files?

Not by default. Use --hidden, and --no-ignore to also ignore .gitignore rules.

Can ripgrep search PDFs?

Not directly — it searches text. Use pdftotext first, or a tool with a PDF preprocessor.