Merge pull request #36 from epage/bin

feat: Ignore binary files

Cargo.lock

@@ -71,8 +71,18 @@ version = "1.0.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
-name = "bytecount"
+name = "bstr"
-version = "0.3.2"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "lazy_static 1.2.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "memchr 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex-automata 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "byteorder"
+version = "1.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
@@ -153,22 +163,6 @@ name = "either"
 version = "1.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
-[[package]]
-name = "encoding_rs"
-version = "0.8.14"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "cfg-if 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
-[[package]]
-name = "encoding_rs_io"
-version = "0.1.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "encoding_rs 0.8.14 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
 [[package]]
 name = "failure"
 version = "0.1.5"
@@ -237,28 +231,6 @@ dependencies = [
  "walkdir 2.2.7 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
-[[package]]
-name = "grep-matcher"
-version = "0.1.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "memchr 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
-[[package]]
-name = "grep-searcher"
-version = "0.1.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "bytecount 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)",
- "encoding_rs 0.8.14 (registry+https://github.com/rust-lang/crates.io-index)",
- "encoding_rs_io 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
- "grep-matcher 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
- "log 0.4.6 (registry+https://github.com/rust-lang/crates.io-index)",
- "memchr 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)",
- "memmap 0.6.2 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
 [[package]]
 name = "heck"
 version = "0.3.1"
@@ -333,15 +305,6 @@ dependencies = [
  "libc 0.2.47 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
-[[package]]
-name = "memmap"
-version = "0.6.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-dependencies = [
- "libc 0.2.47 (registry+https://github.com/rust-lang/crates.io-index)",
- "winapi 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)",
-]
-
 [[package]]
 name = "normalize-line-endings"
 version = "0.2.2"
@@ -568,6 +531,14 @@ dependencies = [
  "utf8-ranges 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
+[[package]]
+name = "regex-automata"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "byteorder 1.3.2 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "regex-syntax"
 version = "0.6.4"
@@ -766,10 +737,10 @@ name = "typos"
 version = "0.1.0"
 dependencies = [
  "assert_fs 0.10.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "bstr 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "clap 2.32.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "csv 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)",
  "failure 0.1.5 (registry+https://github.com/rust-lang/crates.io-index)",
- "grep-searcher 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "ignore 0.4.6 (registry+https://github.com/rust-lang/crates.io-index)",
  "itertools 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "lazy_static 1.2.0 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -885,7 +856,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum backtrace 0.3.13 (registry+https://github.com/rust-lang/crates.io-index)" = "b5b493b66e03090ebc4343eb02f94ff944e0cbc9ac6571491d170ba026741eb5"
 "checksum backtrace-sys 0.1.28 (registry+https://github.com/rust-lang/crates.io-index)" = "797c830ac25ccc92a7f8a7b9862bde440715531514594a6154e3d4a54dd769b6"
 "checksum bitflags 1.0.4 (registry+https://github.com/rust-lang/crates.io-index)" = "228047a76f468627ca71776ecdebd732a3423081fcf5125585bcd7c49886ce12"
-"checksum bytecount 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)" = "f861d9ce359f56dbcb6e0c2a1cb84e52ad732cadb57b806adeb3c7668caccbd8"
+"checksum bstr 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "6cc0572e02f76cb335f309b19e0a0d585b4f62788f7d26de2a13a836a637385f"
+"checksum byteorder 1.3.2 (registry+https://github.com/rust-lang/crates.io-index)" = "a7c3dd8985a7111efc5c80b44e23ecdd8c007de8ade3b96595387e812b957cf5"
 "checksum cc 1.0.28 (registry+https://github.com/rust-lang/crates.io-index)" = "bb4a8b715cb4597106ea87c7c84b2f1d452c7492033765df7f32651e66fcf749"
 "checksum cfg-if 0.1.6 (registry+https://github.com/rust-lang/crates.io-index)" = "082bb9b28e00d3c9d39cc03e64ce4cea0f1bb9b3fde493f0cbc008472d22bdf4"
 "checksum clap 2.32.0 (registry+https://github.com/rust-lang/crates.io-index)" = "b957d88f4b6a63b9d70d5f454ac8011819c6efa7727858f458ab71c756ce2d3e"
@@ -896,8 +868,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum csv-core 0.1.5 (registry+https://github.com/rust-lang/crates.io-index)" = "fa5cdef62f37e6ffe7d1f07a381bc0db32b7a3ff1cac0de56cb0d81e71f53d65"
 "checksum difference 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "524cbf6897b527295dff137cec09ecf3a05f4fddffd7dfcd1585403449e74198"
 "checksum either 1.5.0 (registry+https://github.com/rust-lang/crates.io-index)" = "3be565ca5c557d7f59e7cfcf1844f9e3033650c929c6566f511e8005f205c1d0"
-"checksum encoding_rs 0.8.14 (registry+https://github.com/rust-lang/crates.io-index)" = "a69d152eaa438a291636c1971b0a370212165ca8a75759eb66818c5ce9b538f7"
-"checksum encoding_rs_io 0.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "098f6a0ab73a9ba256b71344dc82c6d7e252736ad9db7f4e35345f3a1f8713f5"
 "checksum failure 0.1.5 (registry+https://github.com/rust-lang/crates.io-index)" = "795bd83d3abeb9220f257e597aa0080a508b27533824adf336529648f6abf7e2"
 "checksum failure_derive 0.1.5 (registry+https://github.com/rust-lang/crates.io-index)" = "ea1063915fd7ef4309e222a5a07cf9c319fb9c7836b1f89b85458672dbb127e1"
 "checksum float-cmp 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "134a8fa843d80a51a5b77d36d42bc2def9edcb0262c914861d08129fd1926600"
@@ -906,8 +876,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum fuchsia-zircon-sys 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)" = "3dcaa9ae7725d12cdb85b3ad99a434db70b468c09ded17e012d86b5c1010f7a7"
 "checksum globset 0.4.2 (registry+https://github.com/rust-lang/crates.io-index)" = "4743617a7464bbda3c8aec8558ff2f9429047e025771037df561d383337ff865"
 "checksum globwalk 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)" = "89fa2e29859da05acd066bd45996f05c271b271d7ec4a781f909682328f65d25"
-"checksum grep-matcher 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "820946e0c314510779a8d86c5cd03240e0ae0993dabcdb98733a8f6a9001b607"
-"checksum grep-searcher 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "6eb23805170ff0e96894a24847019500de11e9baaabe3dafed75b35a897636e1"
 "checksum heck 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)" = "20564e78d53d2bb135c343b3f47714a56af2061f1c928fdb541dc7b9fdd94205"
 "checksum ignore 0.4.6 (registry+https://github.com/rust-lang/crates.io-index)" = "ad03ca67dc12474ecd91fdb94d758cbd20cb4e7a78ebe831df26a9b7511e1162"
 "checksum itertools 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "5b8467d9c1cebe26feb08c640139247fac215782d35371ade9a2136ed6085358"
@@ -917,7 +885,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum lock_api 0.1.5 (registry+https://github.com/rust-lang/crates.io-index)" = "62ebf1391f6acad60e5c8b43706dde4582df75c06698ab44511d15016bc2442c"
 "checksum log 0.4.6 (registry+https://github.com/rust-lang/crates.io-index)" = "c84ec4b527950aa83a329754b01dbe3f58361d1c5efacd1f6d68c494d08a17c6"
 "checksum memchr 2.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "e1dd4eaac298c32ce07eb6ed9242eda7d82955b9170b7d6db59b2e02cc63fcb8"
-"checksum memmap 0.6.2 (registry+https://github.com/rust-lang/crates.io-index)" = "e2ffa2c986de11a9df78620c01eeaaf27d94d3ff02bf81bfcca953102dd0c6ff"
 "checksum normalize-line-endings 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)" = "2e0a1a39eab95caf4f5556da9289b9e68f0aafac901b2ce80daaf020d3b733a8"
 "checksum num-traits 0.2.6 (registry+https://github.com/rust-lang/crates.io-index)" = "0b3a5d7cc97d6d30d8b9bc8fa19bf45349ffe46241e8816f50f62f6d6aaabee1"
 "checksum owning_ref 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "49a4b8ea2179e6a2e27411d3bca09ca6dd630821cf6894c6c7c8467a8ee7ef13"
@@ -944,6 +911,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum redox_syscall 0.1.51 (registry+https://github.com/rust-lang/crates.io-index)" = "423e376fffca3dfa06c9e9790a9ccd282fafb3cc6e6397d01dbf64f9bacc6b85"
 "checksum redox_termios 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "7e891cfe48e9100a70a3b6eb652fef28920c117d366339687bd5576160db0f76"
 "checksum regex 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "37e7cbbd370869ce2e8dff25c7018702d10b21a20ef7135316f8daecd6c25b7f"
+"checksum regex-automata 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)" = "3ed09217220c272b29ef237a974ad58515bde75f194e3ffa7e6d0bf0f3b01f86"
 "checksum regex-syntax 0.6.4 (registry+https://github.com/rust-lang/crates.io-index)" = "4e47a2ed29da7a9e1960e1639e7a982e6edc6d49be308a3b02daf511504a16d1"
 "checksum remove_dir_all 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)" = "3488ba1b9a2084d38645c4c08276a1752dcbf2c7130d74f1569681ad5d2799c5"
 "checksum rustc-demangle 0.1.13 (registry+https://github.com/rust-lang/crates.io-index)" = "adacaae16d02b6ec37fdc7acfcddf365978de76d1983d3ee22afc260e1ca9619"

Cargo.toml

@@ -27,12 +27,12 @@ ignore = "0.4"
 phf = { version = "0.7", features = ["unicase"] }
 regex = "1.0"
 lazy_static = "1.2.0"
-grep-searcher = "0.1"
 serde = "1.0"
 serde_derive = "1.0"
 serde_json = "1.0"
 itertools = "0.8"
 unicase = "1.1"
+bstr = "0.2"
 
 [dev-dependencies]
 assert_fs = "0.10"

benches/file.rs

@@ -18,6 +18,7 @@ fn process_empty(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });
@@ -37,6 +38,7 @@ fn process_no_tokens(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });
@@ -56,6 +58,7 @@ fn process_single_token(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });
@@ -75,6 +78,7 @@ fn process_sherlock(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });
@@ -94,6 +98,7 @@ fn process_code(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });
@@ -113,6 +118,7 @@ fn process_corpus(b: &mut test::Bencher) {
             sample_path.path(),
             &corrections,
             true,
+            false,
             typos::report::print_silent,
         )
     });

src/lib.rs

@@ -12,15 +12,22 @@ pub use crate::dict::*;
 use std::fs::File;
 use std::io::Read;
 
+use bstr::ByteSlice;
+
 pub fn process_file(
     path: &std::path::Path,
     dictionary: &Dictionary,
     ignore_hex: bool,
+    binary: bool,
     report: report::Report,
 ) -> Result<(), failure::Error> {
     let mut buffer = Vec::new();
     File::open(path)?.read_to_end(&mut buffer)?;
-    for (line_idx, line) in grep_searcher::LineIter::new(b'\n', &buffer).enumerate() {
+    if !binary && buffer.find_byte(b'\0').is_some() {
+        return Ok(());
+    }
+
+    for (line_idx, line) in buffer.lines().enumerate() {
         let line_num = line_idx + 1;
         for ident in tokens::Identifier::parse(line) {
             if !ignore_hex && is_hex(ident.token()) {

src/main.rs

@@ -55,6 +55,12 @@ struct Options {
     /// The approximate number of threads to use.
     threads: usize,
 
+    #[structopt(long, raw(overrides_with = r#""no-binary""#))]
+    /// Search binary files.
+    binary: bool,
+    #[structopt(long, raw(overrides_with = r#""binary""#), raw(hidden = "true"))]
+    no_binary: bool,
+
     #[structopt(long, raw(overrides_with = r#""no-hidden""#))]
     /// Search hidden files and directories.
     hidden: bool,
@@ -118,6 +124,15 @@ impl Options {
         }
     }
 
+    pub fn binary(&self) -> Option<bool> {
+        match (self.binary, self.no_binary) {
+            (true, false) => Some(true),
+            (false, true) => Some(false),
+            (false, false) => None,
+            (_, _) => unreachable!("StructOpt should make this impossible"),
+        }
+    }
+
     pub fn ignore_hidden(&self) -> Option<bool> {
         match (self.hidden, self.no_hidden) {
             (true, false) => Some(false),
@@ -183,6 +198,7 @@ fn run() -> Result<(), failure::Error> {
 
     let dictionary = typos::Dictionary::new();
     let ignore_hex = options.ignore_hex().unwrap_or(true);
+    let binary = options.binary().unwrap_or(false);
 
     let first_path = &options
         .path
@@ -207,6 +223,7 @@ fn run() -> Result<(), failure::Error> {
                 entry.path(),
                 &dictionary,
                 ignore_hex,
+                binary,
                 options.format.report(),
             )?;
         }