ripgrep: add support for lz4 decompression

This uses the lz4 binary for decompression.

Closes #898

9 files changed, 30 insertions, 3 deletions

diff --git a/.travis.yml b/.travis.yml
index 7c2c7f93..5fc57d60 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -16,6 +16,7 @@ addons:
       - zsh
       # Needed for testing decompression search.
       - xz-utils
+      - liblz4-tool
 matrix:
   fast_finish: true
   include:
diff --git a/CHANGELOG.md b/CHANGELOG.md
index c7105c55..b2130854 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -42,6 +42,8 @@ Feature enhancements:
   Add `--count-matches` flag, which is like `--count`, but for each match.
 * [FEATURE #880](https://github.com/BurntSushi/ripgrep/issues/880):
   Add a `--no-column` flag, which disables column numbers in the output.
+* [FEATURE #898](https://github.com/BurntSushi/ripgrep/issues/898):
+  Add support for `lz4` when using the `-z/--search-zip` flag.
 * [FEATURE #924](https://github.com/BurntSushi/ripgrep/issues/924):
   `termcolor` has moved to its own repository:
   https://github.com/BurntSushi/termcolor
diff --git a/FAQ.md b/FAQ.md
index c98bafc5..868c4723 100644
--- a/FAQ.md
+++ b/FAQ.md
@@ -135,7 +135,7 @@ How do I search compressed files?
 </h3>
 
 ripgrep's `-z/--search-zip` flag will cause it to search compressed files
-automatically. Currently, this supports gzip, bzip2, lzma and xz only and
+automatically. Currently, this supports gzip, bzip2, lzma, lz4 and xz only and
 requires the corresponding `gzip`, `bzip2` and `xz` binaries to be installed on
 your system. (That is, ripgrep does decompression by shelling out to another
 process.)
diff --git a/README.md b/README.md
index dfa1c94d..57ad0d7c 100644
--- a/README.md
+++ b/README.md
@@ -106,7 +106,7 @@ increases the times to `2.640s` for ripgrep and `10.277s` for GNU grep.
   automatically detecting UTF-16 is provided. Other text encodings must be
   specifically specified with the `-E/--encoding` flag.)
 * ripgrep supports searching files compressed in a common format (gzip, xz,
-  lzma or bzip2 current) with the `-z/--search-zip` flag.
+  lzma, bzip2 or lz4) with the `-z/--search-zip` flag.
 
 In other words, use ripgrep if you like speed, filtering by default, fewer
 bugs, and Unicode support.
diff --git a/ignore/src/types.rs b/ignore/src/types.rs
index 3f96f8e6..8ecbb2b0 100644
--- a/ignore/src/types.rs
+++ b/ignore/src/types.rs
@@ -193,6 +193,7 @@ const DEFAULT_TYPES: &'static [(&'static str, &'static [&'static str])] = &[
     ("log", &["*.log"]),
     ("lua", &["*.lua"]),
     ("lzma", &["*.lzma"]),
+    ("lz4", &["*.lz4"]),
     ("m4", &["*.ac", "*.m4"]),
     ("make", &[
         "gnumakefile", "Gnumakefile", "GNUmakefile",
diff --git a/src/app.rs b/src/app.rs
index 5728371f..a0fdf946 100644
--- a/src/app.rs
+++ b/src/app.rs
@@ -1445,7 +1445,7 @@ This flag can be used with the -o/--only-matching flag.
 fn flag_search_zip(args: &mut Vec<RGArg>) {
     const SHORT: &str = "Search in compressed files.";
     const LONG: &str = long!("\
-Search in compressed files. Currently gz, bz2, xz, and lzma files are
+Search in compressed files. Currently gz, bz2, xz, lzma and lz4 files are
 supported. This option expects the decompression binaries to be available in
 your PATH.
 
diff --git a/src/decompressor.rs b/src/decompressor.rs
index a94948af..235849b7 100644
--- a/src/decompressor.rs
+++ b/src/decompressor.rs
@@ -44,6 +44,7 @@ lazy_static! {
         m.insert("gz", DecompressionCommand::new("gzip", ARGS));
         m.insert("bz2", DecompressionCommand::new("bzip2", ARGS));
         m.insert("xz", DecompressionCommand::new("xz", ARGS));
+        m.insert("lz4", DecompressionCommand::new("lz4", ARGS));
 
         const LZMA_ARGS: &[&str] = &["--format=lzma", "-d", "-c"];
         m.insert("lzma", DecompressionCommand::new("xz", LZMA_ARGS));
@@ -55,6 +56,7 @@ lazy_static! {
         builder.add(Glob::new("*.gz").unwrap());
         builder.add(Glob::new("*.bz2").unwrap());
         builder.add(Glob::new("*.xz").unwrap());
+        builder.add(Glob::new("*.lz4").unwrap());
         builder.add(Glob::new("*.lzma").unwrap());
         builder.build().unwrap()
     };
@@ -63,6 +65,7 @@ lazy_static! {
         builder.add(Glob::new("*.tar.gz").unwrap());
         builder.add(Glob::new("*.tar.xz").unwrap());
         builder.add(Glob::new("*.tar.bz2").unwrap());
+        builder.add(Glob::new("*.tar.lz4").unwrap());
         builder.add(Glob::new("*.tgz").unwrap());
         builder.add(Glob::new("*.txz").unwrap());
         builder.add(Glob::new("*.tbz2").unwrap());
diff --git a/tests/data/sherlock.lz4 b/tests/data/sherlock.lz4
new file mode 100644
index 00000000..5913dbba
--- /dev/null
+++ b/tests/data/sherlock.lz4
diff --git a/tests/tests.rs b/tests/tests.rs
index f3cc8f06..9920c118 100644
--- a/tests/tests.rs
+++ b/tests/tests.rs
@@ -1793,6 +1793,26 @@ be, to a very large extent, the result of luck. Sherlock Holmes
 }
 
 #[test]
+fn compressed_lz4() {
+    if !cmd_exists("lz4") {
+        return;
+    }
+    let lz4_file = include_bytes!("./data/sherlock.lz4");
+
+    let wd = WorkDir::new("feature_search_compressed");
+    wd.create_bytes("sherlock.lz4", lz4_file);
+
+    let mut cmd = wd.command();
+    cmd.arg("-z").arg("Sherlock").arg("sherlock.lz4");
+    let lines: String = wd.stdout(&mut cmd);
+    let expected = "\
+For the Doctor Watsons of this world, as opposed to the Sherlock
+be, to a very large extent, the result of luck. Sherlock Holmes
+";
+    assert_eq!(lines, expected);
+}
+
+#[test]
 fn compressed_lzma() {
     if !cmd_exists("xz") {
         return;