⚡ fsearch — Fast File Search & Duplicate Finder

A blazingly fast, cross-platform library and CLI tool for:

• 🔍 Searching files by name (glob) or content (in-file grep)
• 🔁 Finding duplicate files — by content hash, name, or size

📦 Features

📦 Installation

CLI binary

cargo install fast-search

Library dependency

# Cargo.toml
[dependencies]
fast-search = "1.1"

Pre-built binaries

Download from GitHub Releases:

From source

git clone https://github.com/cumulus13/fsearch
cd fsearch
cargo build --release
# binary → ./target/release/fsearch  (or fsearch.exe on Windows)

🔧 Usage

fsearch <SUBCOMMAND> [OPTIONS]

Subcommands:
  find    🔍 Search files by name or content
  dup     🔁 Find duplicate files
  config  ⚙️  Configuration helpers

fsearch find

# Find all Rust files (depth 5)
fsearch find '*.rs' -d 5

# Search file contents for "TODO" in Python/TS files
fsearch find TODO -f -i '*.py,*.ts' -d 10

# Case-sensitive name search
fsearch find README -C -d 3

# Search in a specific directory, verbose
fsearch find config -p ~/projects -d 3 -v

fsearch dup

# Content duplicates (default — exact byte match via SHA-256)
fsearch dup ~/Downloads -d 5

# Same filename in different directories
fsearch dup . --mode name -d 10

# Same size (fast but imprecise)
fsearch dup . --mode size

# Use MD5, only images, files >= 100 KiB
fsearch dup ~/Photos --algo md5 -i '*.jpg,*.png' --min-size 102400

# Skip binary files, limit to 20 groups
fsearch dup . --skip-binary -n 20

Options

💡 Examples

# Find all Rust source files (depth 5)
fsearch '*.rs' -d 5

# Find files containing "TODO" in Python/JS/TS source
fsearch TODO -f -i '*.py,*.js,*.ts' -d 10

# Case-sensitive search for exact name
fsearch README -C

# Search in a specific directory, verbose output
fsearch config -p ~/projects -d 3 -v

# Exclude extra dirs on top of defaults
fsearch main -x dist,build -d 5

# Limit to first 20 results
fsearch '*.log' -n 20 -d 4

# Generate your config file
fsearch --init-config

# Inspect active config
fsearch --show-config

fsearch config

fsearch config init    # write ~/.config/fsearch/config.toml
fsearch config show    # print active config as TOML
fsearch config path    # print config file location

📚 Library usage

Add to Cargo.toml:

[dependencies]
fast-search = "1.1"

File search

use fsearch::searcher::{fast_find, SearchOptions};
use std::sync::{Arc, atomic::AtomicBool};

let opts = SearchOptions::builder("*.rs")
    .base_dir("./src")
    .max_depth(5)
    .case_insensitive(true)
    .build();

let interrupted = Arc::new(AtomicBool::new(false));
let results = fast_find(&opts, interrupted)?;
for m in &results {
    println!("{}", m.path().display());
}

Content search

use fsearch::searcher::{fast_find, SearchOptions};
use fsearch::searcher::SearchMatch;
use std::sync::{Arc, atomic::AtomicBool};

let opts = SearchOptions::builder("TODO")
    .base_dir(".")
    .max_depth(10)
    .search_in_files(true)
    .include_patterns(vec!["*.rs".into()])
    .build();

let interrupted = Arc::new(AtomicBool::new(false));
let results = fast_find(&opts, Arc::clone(&interrupted))?;

for m in &results {
    if let SearchMatch::Content { path, lines } = m {
        for (line_num, text) in lines {
            println!("{}:{} {}", path.display(), line_num, text);
        }
    }
}

Duplicate detection

use fsearch::duplicates::{find_duplicates, DuplicateOptions, DuplicateMode, HashAlgorithm};
use std::sync::{Arc, atomic::AtomicBool};

let opts = DuplicateOptions::builder(".")
    .max_depth(10)
    .mode(DuplicateMode::Content)
    .algorithm(HashAlgorithm::Sha256)
    .min_size(1024)           // skip files < 1 KiB
    .skip_binary(false)
    .build();

let interrupted = Arc::new(AtomicBool::new(false));
let (groups, summary) = find_duplicates(&opts, interrupted)?;

println!("{} groups · {} wasted",
    summary.groups_found,
    summary.wasted_human());

for group in &groups {
    println!("\n[{}] {} × {}",
        &group.hash[..16], group.paths.len(), group.size_human());
    for p in &group.paths {
        println!("  {}", p.display());
    }
}

⚙️ Configuration

fsearch config init    # write ~/.config/fsearch/config.toml
fsearch config show    # print active config as TOML
fsearch config path    # print config file location
# → ~/.config/fsearch/config.toml

Config lookup order (highest priority first):

1. ./fsearch.toml — project-local override
2. ~/.config/fsearch/config.toml — user config
3. Built-in defaults

See fsearch.toml.example for all available keys with documentation.

All config keys

# Search behaviour
default_depth       = 1       # default -d value
default_method      = 1       # 1 or 2
case_insensitive    = true
include_dirs        = true
binary_check_bytes  = 1024
max_line_length     = 10000
threads             = 0       # 0 = all logical CPUs

# Output
verbose             = false
show_size           = false
show_modified       = false
max_results         = 0       # 0 = unlimited

# Colours — any CSS hex string
color_index         = "#FF88FF"
color_path          = "#FFFF00"
color_line_num      = "#FF4444"
color_line_text     = "#00FFFF"
color_header        = "#FFFFFF"
color_count         = "#00FFFF"
color_error         = "#FF3333"
color_warn          = "#FFAA00"
color_info          = "#00FF88"
color_pattern       = "#FF00FF"

# Filters
exclude_dirs        = ".git,node_modules,.svn,__pycache__,.hg,target,.cache"
default_include     = ""      # e.g. "*.py,*.rs"

📊 Exit Codes

🏗️ Building for Windows from Linux

# Install rustup (https://rustup.rs), then:
rustup target add x86_64-pc-windows-gnu
sudo apt install gcc-mingw-w64-x86-64

# Uncomment the [target.x86_64-pc-windows-gnu] block in .cargo/config.toml, then:
cargo build --release --target x86_64-pc-windows-gnu
# → target/x86_64-pc-windows-gnu/release/fsearch.exe

Or simply push a version tag — GitHub Actions builds all platforms automatically.

🤝 Contributing

1. Fork & create a feature branch
2. Ensure cargo clippy -- -D warnings and cargo test both pass
3. Open a pull request

📄 License

MIT — see LICENSE

👤 Author

Hadi Cahyadi — cumulus13@gmail.com
GitHub: @cumulus13