feat(parser): add support for reading GZIP files

Cargo.lock

@@ -2,6 +2,12 @@
 # It is not intended for manual editing.
 version = 3
 
+[[package]]
+name = "adler"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
+
 [[package]]
 name = "aho-corasick"
 version = "0.7.18"
@@ -66,6 +72,15 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "crc32fast"
+version = "1.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "738c290dfaea84fc1ca15ad9c168d083b05a714e1efddd8edaab678dc28d2836"
+dependencies = [
+ "cfg-if",
+]
+
 [[package]]
 name = "crossbeam-channel"
 version = "0.5.1"
@@ -137,6 +152,18 @@ version = "1.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e78d4f1cc4ae33bbfc157ed5d5a5ef3bc29227303d595861deb238fcec4e9457"
 
+[[package]]
+name = "flate2"
+version = "1.0.22"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e6988e897c1c9c485f43b47a529cef42fde0547f9d8d41a7062518f1d8fc53f"
+dependencies = [
+ "cfg-if",
+ "crc32fast",
+ "libc",
+ "miniz_oxide",
+]
+
 [[package]]
 name = "fnv"
 version = "1.0.7"
@@ -256,6 +283,16 @@ dependencies = [
  "autocfg",
 ]
 
+[[package]]
+name = "miniz_oxide"
+version = "0.4.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a92518e98c078586bc6c934028adcca4c92a53d6a958196de835170a01d84e4b"
+dependencies = [
+ "adler",
+ "autocfg",
+]
+
 [[package]]
 name = "num_cpus"
 version = "1.13.0"
@@ -455,6 +492,7 @@ dependencies = [
 name = "systeroid-parser"
 version = "0.1.0"
 dependencies = [
+ "flate2",
  "globwalk",
  "regex",
  "thiserror",

systeroid-parser/Cargo.toml

@@ -8,3 +8,4 @@ edition = "2021"
 regex = "1.5.4"
 globwalk = "0.8.1"
 thiserror = "1.0.29"
+flate2 = "1.0.22"

systeroid-parser/src/reader.rs

@@ -1,7 +1,12 @@
+use flate2::read::GzDecoder;
 use std::fs::File;
-use std::io::{BufRead, BufReader as IoBufReader, Result as IoResult};
+use std::io::{
+    BufRead, BufReader as IoBufReader, Error as IoError, ErrorKind as IoErrorKind, Read,
+    Result as IoResult,
+};
 use std::path::Path;
 use std::rc::Rc;
+use std::str;
 
 /// Default buffer size of the reader.
 const DEFAULT_BUFFER_SIZE: usize = 1024;
@@ -11,7 +16,7 @@ pub struct BufReader {
     /// Inner type.
     reader: IoBufReader<File>,
     /// Buffer.
-    buffer: Rc<String>,
+    buffer: Rc<Vec<u8>>,
 }
 
 impl BufReader {
@@ -24,15 +29,15 @@ impl BufReader {
     }
 
     /// Creates a new buffer with the given size.
-    fn new_buffer(buffer_size: Option<usize>) -> Rc<String> {
-        Rc::new(String::with_capacity(
+    fn new_buffer(buffer_size: Option<usize>) -> Rc<Vec<u8>> {
+        Rc::new(Vec::with_capacity(
             buffer_size.unwrap_or(DEFAULT_BUFFER_SIZE),
         ))
     }
 }
 
 impl Iterator for BufReader {
-    type Item = IoResult<Rc<String>>;
+    type Item = IoResult<Rc<Vec<u8>>>;
     fn next(&mut self) -> Option<Self::Item> {
         let buffer = match Rc::get_mut(&mut self.buffer) {
             Some(rc_buffer) => {
@@ -45,7 +50,7 @@ impl Iterator for BufReader {
             }
         };
         self.reader
-            .read_line(buffer)
+            .read_until(b'\n', buffer)
             .map(|u| {
                 if u == 0 {
                     None
@@ -63,11 +68,29 @@ impl Iterator for BufReader {
 pub fn read_to_string<P: AsRef<Path>>(path: P) -> IoResult<String> {
     let mut lines = Vec::<String>::new();
     for line in BufReader::open(path, None)? {
-        lines.push(line?.to_string());
+        lines.push(
+            str::from_utf8(&line?)
+                .map_err(|e| IoError::new(IoErrorKind::Other, e))?
+                .to_string(),
+        );
     }
     Ok(lines.join(""))
 }
 
+/// Reads (decodes) the given gzip file into a string.
+///
+/// Uses [`BufReader`] under the hood.
+pub fn read_gzip<P: AsRef<Path>>(path: P) -> IoResult<String> {
+    let mut bytes = Vec::<u8>::new();
+    for read_bytes in BufReader::open(path, None)? {
+        bytes.extend(read_bytes?.to_vec());
+    }
+    let mut gz = GzDecoder::new(&bytes[..]);
+    let mut data = String::new();
+    gz.read_to_string(&mut data)?;
+    Ok(data)
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;