path: root/src/args.rs

use std::cmp;
use std::env;
use std::io;
use std::path::{Path, PathBuf};

use docopt::Docopt;
use env_logger;
use grep::{Grep, GrepBuilder};
use log;
use num_cpus;
use regex;
use term::Terminal;
#[cfg(not(windows))]
use term;
#[cfg(windows)]
use term::WinConsole;
use walkdir::WalkDir;

use atty;
use gitignore::{Gitignore, GitignoreBuilder};
use ignore::Ignore;
use out::{Out, ColoredTerminal};
use printer::Printer;
use search_buffer::BufferSearcher;
use search_stream::{InputBuffer, Searcher};
#[cfg(windows)]
use terminal_win::WindowsBuffer;
use types::{FileTypeDef, Types, TypesBuilder};
use walk;

use Result;

/// The Docopt usage string.
///
/// If you've never heard of Docopt before, see: http://docopt.org
/// (TL;DR: The CLI parser is generated from the usage string below.)
const USAGE: &'static str = "
Usage: rg [options] -e PATTERN ... [<path> ...]
       rg [options] <pattern> [<path> ...]
       rg [options] --files [<path> ...]
       rg [options] --type-list
       rg --help
       rg --version

rg combines the usability of The Silver Searcher with the raw speed of grep.

Common options:
    -a, --text                 Search binary files as if they were text.
    -c, --count                Only show count of line matches for each file.
    --color WHEN               Whether to use coloring in match.
                               Valid values are never, always or auto.
                               [default: auto]
    -e, --regexp PATTERN ...   Use PATTERN to search. This option can be
                               provided multiple times, where all patterns
                               given are searched.
    -F, --fixed-strings        Treat the pattern as a literal string instead of
                               a regular expression.
    -g, --glob GLOB ...        Include or exclude files for searching that
                               match the given glob. This always overrides any
                               other ignore logic. Multiple glob flags may be
                               used. Globbing rules match .gitignore globs.
                               Precede a glob with a '!' to exclude it.
    -h, --help                 Show this usage message.
    -i, --ignore-case          Case insensitive search.
    -n, --line-number          Show line numbers (1-based). This is enabled
                               by default at a tty.
    -N, --no-line-number       Suppress line numbers.
    -q, --quiet                Do not print anything to stdout.
    -r, --replace ARG          Replace every match with the string given.
                               Capture group indices (e.g., $5) and names
                               (e.g., $foo) are supported.
    -t, --type TYPE ...        Only search files matching TYPE. Multiple type
                               flags may be provided. Use the --type-list flag
                               to list all available types.
    -T, --type-not TYPE ...    Do not search files matching TYPE. Multiple
                               not-type flags may be provided.
    -v, --invert-match         Invert matching.
    -w, --word-regexp          Only show matches surrounded by word boundaries.
                               This is equivalent to putting \\b before and
                               after the search pattern.

Less common options:
    -A, --after-context NUM
        Show NUM lines after each match.

    -B, --before-context NUM
        Show NUM lines before each match.

    -C, --context NUM
        Show NUM lines before and after each match.

    --column
        Show column numbers (1 based) in output. This only shows the column
        numbers for the first match on each line. Note that this doesn't try
        to account for Unicode. One byte is equal to one column.

    --context-separator ARG
        The string to use when separating non-continuous context lines. Escape
        sequences may be used. [default: --]

    --debug
        Show debug messages.

    --files
        Print each file that would be searched (but don't search).

    -H, --with-filename
        Prefix each match with the file name that contains it. This is the
        default when more than one file is searched.

    --heading
        Show the file name above clusters of matches from each file.
        This is the default mode at a tty.

    --no-heading
        Don't show any file name heading.

    --hidden
        Search hidden directories and files. (Hidden directories and files are
        skipped by default.)

    -L, --follow
        Follow symlinks.

    --mmap
        Search using memory maps when possible. This is enabled by default
        when ripgrep thinks it will be faster. (Note that mmap searching
        doesn't current support the various context related options.)

    --no-mmap
        Never use memory maps, even when they might be faster.

    --no-ignore
        Don't respect ignore files (.gitignore, .rgignore, etc.)
        This implies --no-ignore-parent.

    --no-ignore-parent
        Don't respect ignore files in parent directories.

    -p, --pretty
        Alias for --color=always --heading -n.

    -j, --threads ARG
        The number of threads to use. Defaults to the number of logical CPUs
        (capped at 6). [default: 0]

    --version
        Show the version number of ripgrep and exit.

File type management options:
    --type-list
        Show all supported file types and their associated globs.

    --type-add ARG ...
        Add a new glob for a particular file type.
        Example: --type-add html:*.html,*.htm

    --type-clear TYPE ...
        Clear the file type globs for TYPE.
";

/// RawArgs are the args as they are parsed from Docopt. They aren't used
/// directly by the rest of ripgrep.
#[derive(Debug, RustcDecodable)]
pub struct RawArgs {
    arg_pattern: String,
    arg_path: Vec<String>,
    flag_after_context: usize,
    flag_before_context: usize,
    flag_color: String,
    flag_column: bool,
    flag_context: usize,
    flag_context_separator: String,
    flag_count: bool,
    flag_debug: bool,
    flag_files: bool,
    flag_follow: bool,
    flag_glob: Vec<String>,
    flag_heading: bool,
    flag_hidden: bool,
    flag_ignore_case: bool,
    flag_invert_match: bool,
    flag_line_number: bool,
    flag_fixed_strings: bool,
    flag_mmap: bool,
    flag_no_heading: bool,
    flag_no_ignore: bool,
    flag_no_ignore_parent: bool,
    flag_no_line_number: bool,
    flag_no_mmap: bool,
    flag_pretty: bool,
    flag_quiet: bool,
    flag_regexp: Vec<String>,
    flag_replace: Option<String>,
    flag_text: bool,
    flag_threads: usize,
    flag_type: Vec<String>,
    flag_type_not: Vec<String>,
    flag_type_list: bool,
    flag_type_add: Vec<String>,
    flag_type_clear: Vec<String>,
    flag_with_filename: bool,
    flag_word_regexp: bool,
}

/// Args are transformed/normalized from RawArgs.
#[derive(Debug)]
pub struct Args {
    pattern: String,
    paths: Vec<PathBuf>,
    after_context: usize,
    before_context: usize,
    color: bool,
    column: bool,
    context_separator: Vec<u8>,
    count: bool,
    eol: u8,
    files: bool,
    follow: bool,
    glob_overrides: Option<Gitignore>,
    grep: Grep,
    heading: bool,
    hidden: bool,
    ignore_case: bool,
    invert_match: bool,
    line_number: bool,
    mmap: bool,
    no_ignore: bool,
    no_ignore_parent: bool,
    quiet: bool,
    replace: Option<Vec<u8>>,
    text: bool,
    threads: usize,
    type_defs: Vec<FileTypeDef>,
    type_list: bool,
    types: Types,
    with_filename: bool,
}

impl RawArgs {
    /// Convert arguments parsed into a configuration used by ripgrep.
    fn to_args(&self) -> Result<Args> {
        let pattern = self.pattern();
        let paths =
            if self.arg_path.is_empty() {
                if atty::on_stdin()
                    || self.flag_files
                    || self.flag_type_list {
                    vec![Path::new("./").to_path_buf()]
                } else {
                    vec![Path::new("-").to_path_buf()]
                }
            } else {
                self.arg_path.iter().map(|p| {
                    Path::new(p).to_path_buf()
                }).collect()
            };
        let (after_context, before_context) =
            if self.flag_context > 0 {
                (self.flag_context, self.flag_context)
            } else {
                (self.flag_after_context, self.flag_before_context)
            };
        let mmap =
            if before_context > 0 || after_context > 0 || self.flag_no_mmap {
                false
            } else if self.flag_mmap {
                true
            } else {
                // If we're only searching a few paths and all of them are
                // files, then memory maps are probably faster.
                paths.len() <= 10 && paths.iter().all(|p| p.is_file())
            };
        if mmap {
            debug!("will try to use memory maps");
        }
        let glob_overrides =
            if self.flag_glob.is_empty() {
                None
            } else {
                let cwd = try!(env::current_dir());
                let mut bgi = GitignoreBuilder::new(cwd);
                for pat in &self.flag_glob {
                    </