Merge pull request #330 from figsoda/ext

improve extension parsing logic

CHANGELOG.md

@@ -28,6 +28,7 @@ Categories Used:
 
 - Allow ouch to decompress archive into existing folder [\#321](https://github.com/ouch-org/ouch/pull/321) ([a-moreira](https://github.com/a-moreira))
 - Accept inserting subcommand-independent flags in any position [\#329](https://github.com/ouch-org/ouch/pull/329) ([marcospb19](https://github.com/marcospb19))
+- Improve extension parsing logic [\#330](https://github.com/ouch-org/ouch/pull/330) ([figsoda](https://github.com/figsoda))
 
 ## [0.4.0](https://github.com/ouch-org/ouch/compare/0.3.1...0.4.0) (2022-11-20)
 

Cargo.lock

@@ -19,11 +19,11 @@ dependencies = [
 
 [[package]]
 name = "assert_cmd"
-version = "2.0.6"
+version = "2.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ba45b8163c49ab5f972e59a8a5a03b6d2972619d486e19ec9fe744f7c2753d3c"
+checksum = "fa3d466004a8b4cb1bc34044240a2fd29d17607e2e3bd613eb44fd48e8100da3"
 dependencies = [
- "bstr 1.0.1",
+ "bstr 1.1.0",
  "doc-comment",
  "predicates",
  "predicates-core",
@@ -37,7 +37,7 @@ version = "0.2.14"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
 dependencies = [
- "hermit-abi",
+ "hermit-abi 0.1.19",
  "libc",
  "winapi",
 ]
@@ -80,9 +80,9 @@ dependencies = [
 
 [[package]]
 name = "bstr"
-version = "1.0.1"
+version = "1.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fca0852af221f458706eb0725c03e4ed6c46af9ac98e6a689d5e634215d594dd"
+checksum = "b45ea9b00a7b3f2988e9a65ad3917e62123c38dba709b666506207be96d1790b"
 dependencies = [
  "memchr",
  "once_cell",
@@ -119,9 +119,9 @@ dependencies = [
 
 [[package]]
 name = "cc"
-version = "1.0.77"
+version = "1.0.78"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e9f73505338f7d905b19d18738976aae232eb46b8efc15554ffc56deb5d9ebe4"
+checksum = "a20104e2335ce8a659d6dd92a51a767a0c062599c73b343fd152cb401e828c3d"
 dependencies = [
  "jobserver",
 ]
@@ -145,14 +145,14 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "clap"
-version = "4.0.26"
+version = "4.0.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2148adefda54e14492fb9bddcc600b4344c5d1a3123bd666dcb939c6f0e0e57e"
+checksum = "a7db700bc935f9e43e88d00b0850dae18a63773cfbec6d8e070fccf7fef89a39"
 dependencies = [
- "atty",
  "bitflags",
  "clap_derive",
  "clap_lex",
+ "is-terminal",
  "once_cell",
  "strsim",
  "termcolor",
@@ -160,9 +160,9 @@ dependencies = [
 
 [[package]]
 name = "clap_complete"
-version = "4.0.5"
+version = "4.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "96b0fba905b035a30d25c1b585bf1171690712fbb0ad3ac47214963aa4acc36c"
+checksum = "10861370d2ba66b0f5989f83ebf35db6421713fd92351790e7fdd6c36774c56b"
 dependencies = [
  "clap",
 ]
@@ -191,9 +191,9 @@ dependencies = [
 
 [[package]]
 name = "clap_mangen"
-version = "0.2.4"
+version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "aa149477df7854a1497db0def32b8a65bf98f72a14d04ac75b01938285d83420"
+checksum = "904eb24d05ad587557e0f484ddce5c737c30cf81372badb16d13e41c4b8340b1"
 dependencies = [
  "clap",
  "roff",
@@ -235,6 +235,27 @@ version = "1.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "90e5c1c8368803113bf0c9584fc495a58b86dc8a29edbf8fe877d21d9507e797"
 
+[[package]]
+name = "errno"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f639046355ee4f37944e44f60642c6f3a7efa3cf6b78c78a0d989a8ce6c396a1"
+dependencies = [
+ "errno-dragonfly",
+ "libc",
+ "winapi",
+]
+
+[[package]]
+name = "errno-dragonfly"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aa68f1b12764fab894d2755d2518754e71b4fd80ecfb822714a1206c2aab39bf"
+dependencies = [
+ "cc",
+ "libc",
+]
+
 [[package]]
 name = "fastrand"
 version = "1.8.0"
@@ -246,9 +267,9 @@ dependencies = [
 
 [[package]]
 name = "filetime"
-version = "0.2.18"
+version = "0.2.19"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4b9663d381d07ae25dc88dbdf27df458faa83a9b25336bcac83d5e452b5fc9d3"
+checksum = "4e884668cd0c7480504233e951174ddc3b382f7c2666e3b7310b5c4e7b0c37f9"
 dependencies = [
  "cfg-if",
  "libc",
@@ -258,9 +279,9 @@ dependencies = [
 
 [[package]]
 name = "flate2"
-version = "1.0.24"
+version = "1.0.25"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f82b0f4c27ad9f8bfd1f3208d882da2b09c301bc1c828fd3a00d0216d2fbbff6"
+checksum = "a8a2db397cb1c8772f31494cb8917e48cd1e64f0fa7efac59fbd741a0a8ce841"
 dependencies = [
  "crc32fast",
  "libz-sys",
@@ -319,10 +340,19 @@ dependencies = [
 ]
 
 [[package]]
-name = "humansize"
+name = "hermit-abi"
-version = "2.1.2"
+version = "0.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4e682e2bd70ecbcce5209f11a992a4ba001fea8e60acf7860ce007629e6d2756"
+checksum = "ee512640fe35acbfb4bb779db6f0d80704c2cacfa2e39b601ef3e3f47d1ae4c7"
+dependencies = [
+ "libc",
+]
+
+[[package]]
+name = "humansize"
+version = "2.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6cb51c9a029ddc91b07a787f1d86b53ccfa49b0e86688c946ebe8d3555685dd7"
 dependencies = [
  "libm",
 ]
@@ -347,9 +377,9 @@ dependencies = [
 
 [[package]]
 name = "infer"
-version = "0.11.0"
+version = "0.12.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0a6c16b11a665b26aeeb9b1d7f954cdeb034be38dd00adab4f2ae921a8fee804"
+checksum = "a898e4b7951673fce96614ce5751d13c40fc5674bc2d759288e46c3ab62598b3"
 dependencies = [
  "cfb",
 ]
@@ -363,6 +393,28 @@ dependencies = [
  "cfg-if",
 ]
 
+[[package]]
+name = "io-lifetimes"
+version = "1.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "46112a93252b123d31a119a8d1a1ac19deac4fac6e0e8b0df58f0d4e5870e63c"
+dependencies = [
+ "libc",
+ "windows-sys",
+]
+
+[[package]]
+name = "is-terminal"
+version = "0.4.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "28dfb6c8100ccc63462345b67d1bbc3679177c75ee4bf59bf29c8b1d110b8189"
+dependencies = [
+ "hermit-abi 0.2.6",
+ "io-lifetimes",
+ "rustix",
+ "windows-sys",
+]
+
 [[package]]
 name = "is_executable"
 version = "1.0.1"
@@ -383,9 +435,9 @@ dependencies = [
 
 [[package]]
 name = "itoa"
-version = "1.0.4"
+version = "1.0.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4217ad341ebadf8d8e724e264f13e593e0648f5b3e94b3896a5df283be015ecc"
+checksum = "fad582f4b9e86b6caa621cabeb0963332d92eea04729ab12892c2533951e6440"
 
 [[package]]
 name = "jobserver"
@@ -404,9 +456,9 @@ checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
 
 [[package]]
 name = "libc"
-version = "0.2.137"
+version = "0.2.139"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fc7fcc620a3bff7cdd7a365be3376c97191aeaccc2a603e600951e452615bf89"
+checksum = "201de327520df007757c1f0adce6e827fe8562fbc28bfd9c15571c66ca1f5f79"
 
 [[package]]
 name = "libm"
@@ -431,6 +483,12 @@ version = "0.5.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0717cef1bc8b636c6e1c1bbdefc09e6322da8a9321966e8928ef80d20f7f770f"
 
+[[package]]
+name = "linux-raw-sys"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f051f77a7c8e6957c0696eac88f26b0117e54f52d3fc682ab19397a8812846a4"
+
 [[package]]
 name = "log"
 version = "0.4.17"
@@ -468,9 +526,9 @@ checksum = "2dffe52ecf27772e601905b7522cb4ef790d2cc203488bbd0e2fe85fcb74566d"
 
 [[package]]
 name = "miniz_oxide"
-version = "0.5.4"
+version = "0.6.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "96590ba8f175222643a85693f33d26e9c8a015f599c216509b1a6894af675d34"
+checksum = "b275950c28b37e794e8c55d88aeb5e139d0ce23fdbbeda68f8d7174abdf9e8fa"
 dependencies = [
  "adler",
 ]
@@ -486,9 +544,9 @@ dependencies = [
 
 [[package]]
 name = "once_cell"
-version = "1.16.0"
+version = "1.17.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "86f0b0d4bf799edbc74508c1e8bf170ff5f41238e5f8225603ca7caaae2b7860"
+checksum = "6f61fba1741ea2b3d6a1e3178721804bb716a68a6aeba1149b5d52e3d464ea66"
 
 [[package]]
 name = "os_str_bytes"
@@ -502,6 +560,7 @@ version = "0.4.0"
 dependencies = [
  "assert_cmd",
  "atty",
+ "bstr 1.1.0",
  "bzip2",
  "clap",
  "clap_complete",
@@ -534,9 +593,9 @@ dependencies = [
 
 [[package]]
 name = "parse-display"
-version = "0.6.0"
+version = "0.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3b25af4ef94a8528b41fb49a696e361dc6ef975c782417268072d987ac327964"
+checksum = "6f96cc033d72896bb9a2c239a14e1141c3e2eae6d649e7c10ef4e598d66bc86c"
 dependencies = [
  "once_cell",
  "parse-display-derive",
@@ -545,9 +604,9 @@ dependencies = [
 
 [[package]]
 name = "parse-display-derive"
-version = "0.6.0"
+version = "0.8.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "73f106cced1f4b645e3fca6125105cdf7407e35d1af710f290aac530f6b826b9"
+checksum = "e5587062be441f3d868f7c4c9d13c67f286b03aa679d7f8176ef80bf2ee79e5d"
 dependencies = [
  "once_cell",
  "proc-macro2",
@@ -572,9 +631,9 @@ checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
 
 [[package]]
 name = "predicates"
-version = "2.1.3"
+version = "2.1.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ed6bd09a7f7e68f3f0bf710fb7ab9c4615a488b58b5f653382a687701e458c92"
+checksum = "59230a63c37f3e18569bdb90e4a89cbf5bf8b06fea0b84e65ea10cc4df47addd"
 dependencies = [
  "difflib",
  "itertools",
@@ -623,9 +682,9 @@ dependencies = [
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.47"
+version = "1.0.49"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5ea3d908b0e36316caf9e9e2c4625cdde190a7e6f440d794667ed17a1855e725"
+checksum = "57a8eca9f9c4ffde41714334dee777596264c7825420f521abc92b5b5deb63a5"
 dependencies = [
  "unicode-ident",
 ]
@@ -664,9 +723,9 @@ checksum = "a993555f31e5a609f617c12db6250dedcac1b0a85076912c436e6fc9b2c8e6a3"
 
 [[package]]
 name = "quote"
-version = "1.0.21"
+version = "1.0.23"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bbe448f377a7d6961e30f5955f9b8d106c3f5e449d493ee1b125c1d43c2b5179"
+checksum = "8856d8364d252a14d474036ea1358d63c9e6965c8e5c1885c18f73d70bff9c7b"
 dependencies = [
  "proc-macro2",
 ]
@@ -757,6 +816,20 @@ version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b833d8d034ea094b1ea68aa6d5c740e0d04bad9d16568d08ba6f76823a114316"
 
+[[package]]
+name = "rustix"
+version = "0.36.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4feacf7db682c6c329c4ede12649cd36ecab0f3be5b7d74e6a20304725db4549"
+dependencies = [
+ "bitflags",
+ "errno",
+ "io-lifetimes",
+ "libc",
+ "linux-raw-sys",
+ "windows-sys",
+]
+
 [[package]]
 name = "rusty-fork"
 version = "0.3.0"
@@ -780,9 +853,9 @@ dependencies = [
 
 [[package]]
 name = "serde"
-version = "1.0.147"
+version = "1.0.152"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d193d69bae983fc11a79df82342761dfbf28a99fc8d203dca4c3c1b590948965"
+checksum = "bb7d1f0d3021d347a83e556fc4683dea2ea09d87bccdf88ff5c12545d89d5efb"
 
 [[package]]
 name = "snap"
@@ -821,9 +894,9 @@ dependencies = [
 
 [[package]]
 name = "syn"
-version = "1.0.103"
+version = "1.0.107"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a864042229133ada95abf3b54fdc62ef5ccabe9515b64717bcb9a1919e59445d"
+checksum = "1f4064b5b16e03ae50984a5a8ed5d4f8803e6bc1fd170a3cda91a1be4b18e3f5"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -920,9 +993,9 @@ dependencies = [
 
 [[package]]
 name = "unicode-ident"
-version = "1.0.5"
+version = "1.0.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6ceab39d59e4c9499d4e5a8ee0e2735b891bb7308ac83dfb4e80cad195c9f6f3"
+checksum = "84a22b9f218b40614adcb3f4ff08b703773ad44fa9423e4e0d346d5db86e4ebc"
 
 [[package]]
 name = "uuid"
@@ -1089,9 +1162,9 @@ dependencies = [
 
 [[package]]
 name = "zstd"
-version = "0.12.0+zstd.1.5.2"
+version = "0.12.1+zstd.1.5.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b8148aa921e9d53217ab9322f8553bd130f7ae33489db68b381d76137d2e6374"
+checksum = "5c947d2adc84ff9a59f2e3c03b81aa4128acf28d6ad7d56273f7e8af14e47bea"
 dependencies = [
  "zstd-safe",
 ]

Cargo.toml

@@ -12,24 +12,25 @@ description = "A command-line utility for easily compressing and decompressing f
 
 [dependencies]
 atty = "0.2.14"
+bstr = "1.1.0"
 bzip2 = "0.4.3"
-clap = { version = "4.0.26", features = ["derive", "env"] }
+clap = { version = "4.0.32", features = ["derive", "env"] }
-filetime = "0.2.18"
+filetime = "0.2.19"
-flate2 = { version = "1.0.24", default-features = false }
+flate2 = { version = "1.0.25", default-features = false }
 fs-err = "2.9.0"
-humansize = "2.1.2"
+humansize = "2.1.3"
 ignore = "0.4.18"
-libc = "0.2.137"
+libc = "0.2.139"
 linked-hash-map = "0.5.6"
 lzzzz = "1.0.4"
-once_cell = "1.16.0"
+once_cell = "1.17.0"
 same-file = "1.0.6"
 snap = "1.1.0"
 tar = "0.4.38"
 tempfile = "3.3.0"
 xz2 = "0.1.7"
 zip = { version = "0.6.3", default-features = false, features = ["time"] }
-zstd = { version = "0.12.0", default-features = false }
+zstd = { version = "0.12.1", default-features = false }
 # zstd-sys > 2.0.1 unconditionally enables thin LTO and causes CI to fail
 # https://github.com/gyscos/zstd-rs/pull/155
 zstd-sys = "=2.0.1"
@@ -41,14 +42,14 @@ time = { version = "0.3.17", default-features = false }
 is_executable = "1.0.1"
 
 [build-dependencies]
-clap = { version = "4.0.26", features = ["derive", "env", "string"] }
+clap = { version = "4.0.32", features = ["derive", "env", "string"] }
-clap_complete = "4.0.5"
+clap_complete = "4.0.7"
-clap_mangen = "0.2.4"
+clap_mangen = "0.2.6"
 
 [dev-dependencies]
-assert_cmd = "2.0.6"
+assert_cmd = "2.0.7"
-infer = "0.11.0"
+infer = "0.12.0"
-parse-display = "0.6.0"
+parse-display = "0.8.0"
 proptest = "1.0.0"
 rand = { version = "0.8.5", default-features = false, features = ["small_rng", "std"] }
 test-strategy = "0.2.1"

src/commands/mod.rs

@@ -259,7 +259,7 @@ pub fn run(
             let mut formats = vec![];
 
             for path in files.iter() {
-                let (_, file_formats) = extension::separate_known_extensions_from_name(path);
+                let file_formats = extension::extensions_from_path(path);
                 formats.push(file_formats);
             }
 

src/extension.rs

@@ -1,6 +1,8 @@
 //! Our representation of all the supported compression formats.
 
-use std::{ffi::OsStr, fmt, path::Path};
+use std::{fmt, path::Path};
+
+use bstr::ByteSlice;
 
 use self::CompressionFormat::*;
 use crate::warning;
@@ -104,54 +106,61 @@ pub const SUPPORTED_EXTENSIONS: &[&str] = &[
     "zst",
 ];
 
+pub fn to_extension(ext: &[u8]) -> Option<Extension> {
+    Some(Extension::new(
+        match ext {
+            b"tar" => &[Tar],
+            b"tgz" => &[Tar, Gzip],
+            b"tbz" | b"tbz2" => &[Tar, Bzip],
+            b"tlz4" => &[Tar, Lz4],
+            b"txz" | b"tlzma" => &[Tar, Lzma],
+            b"tsz" => &[Tar, Snappy],
+            b"tzst" => &[Tar, Zstd],
+            b"zip" => &[Zip],
+            b"bz" | b"bz2" => &[Bzip],
+            b"gz" => &[Gzip],
+            b"lz4" => &[Lz4],
+            b"xz" | b"lzma" => &[Lzma],
+            b"sz" => &[Snappy],
+            b"zst" => &[Zstd],
+            _ => return None,
+        },
+        ext.to_str_lossy(),
+    ))
+}
+
+pub fn split_extension<'a>(name: &mut &'a [u8]) -> Option<&'a [u8]> {
+    let (new_name, ext) = name.rsplit_once_str(b".")?;
+    if matches!(new_name, b"" | b"." | b"..") {
+        return None;
+    }
+    *name = new_name;
+    Some(ext)
+}
+
 /// Extracts extensions from a path.
 ///
 /// Returns both the remaining path and the list of extension objects
-pub fn separate_known_extensions_from_name(mut path: &Path) -> (&Path, Vec<Extension>) {
+pub fn separate_known_extensions_from_name(path: &Path) -> (&Path, Vec<Extension>) {
     let mut extensions = vec![];
 
-    if let Some(file_stem) = path.file_stem().and_then(OsStr::to_str) {
+    let Some(mut name) = path.file_name().and_then(<[u8] as ByteSlice>::from_os_str) else {
-        let file_stem = file_stem.trim_matches('.');
+        return (path, extensions);
+    };
 
+    // While there is known extensions at the tail, grab them
+    while let Some(extension) = split_extension(&mut name).and_then(to_extension) {
+        extensions.insert(0, extension);
+    }
+
+    if let Ok(name) = name.to_str() {
+        let file_stem = name.trim_matches('.');
         if SUPPORTED_EXTENSIONS.contains(&file_stem) {
             warning!("Received a file with name '{file_stem}', but {file_stem} was expected as the extension.");
         }
     }
 
-    // While there is known extensions at the tail, grab them
+    (name.to_path().unwrap(), extensions)
-    while let Some(extension) = path.extension().and_then(OsStr::to_str) {
-        let formats: &[CompressionFormat] = match extension {
-            "tar" => &[Tar],
-            "tgz" => &[Tar, Gzip],
-            "tbz" | "tbz2" => &[Tar, Bzip],
-            "tlz4" => &[Tar, Lz4],
-            "txz" | "tlzma" => &[Tar, Lzma],
-            "tsz" => &[Tar, Snappy],
-            "tzst" => &[Tar, Zstd],
-            "zip" => &[Zip],
-            "bz" | "bz2" => &[Bzip],
-            "gz" => &[Gzip],
-            "lz4" => &[Lz4],
-            "xz" | "lzma" => &[Lzma],
-            "sz" => &[Snappy],
-            "zst" => &[Zstd],
-            _ => break,
-        };
-
-        let extension = Extension::new(formats, extension);
-        extensions.push(extension);
-
-        // Update for the next iteration
-        path = if let Some(stem) = path.file_stem() {
-            Path::new(stem)
-        } else {
-            Path::new("")
-        };
-    }
-    // Put the extensions in the correct order: left to right
-    extensions.reverse();
-
-    (path, extensions)
 }
 
 /// Extracts extensions from a path, return only the list of extension objects