Merge branch 'master' into eocd_rewrite

CHANGELOG.md

@@ -1,5 +1,18 @@
 # Changelog
 
+## [2.2.1](https://github.com/zip-rs/zip2/compare/v2.2.0...v2.2.1) - 2024-11-20
+
+### <!-- 1 -->🐛 Bug Fixes
+
+- remove executable bit ([#238](https://github.com/zip-rs/zip2/pull/238))
+- *(lzma)* fixed panic in case of invalid lzma stream ([#259](https://github.com/zip-rs/zip2/pull/259))
+- resolve new clippy warnings on nightly ([#262](https://github.com/zip-rs/zip2/pull/262))
+- resolve clippy warning in nightly ([#252](https://github.com/zip-rs/zip2/pull/252))
+
+### <!-- 4 -->⚡ Performance
+
+- Faster cde rejection ([#255](https://github.com/zip-rs/zip2/pull/255))
+
 ## [2.2.0](https://github.com/zip-rs/zip2/compare/v2.1.6...v2.2.0) - 2024-08-11
 
 ### <!-- 0 -->🚀 Features

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "zip"
-version = "2.2.0"
+version = "2.2.1"
 authors = [
     "Mathijs van de Nes <[email protected]>",
     "Marli Frost <[email protected]>",
@@ -39,7 +39,7 @@ memchr = "2.7.4"
 pbkdf2 = { version = "0.12.2", optional = true }
 rand = { version = "0.8.5", optional = true }
 sha1 = { version = "0.10.6", optional = true }
-thiserror = "1.0.63"
+thiserror = "2.0.3"
 time = { workspace = true, optional = true, features = [
     "std",
 ] }
@@ -62,7 +62,7 @@ walkdir = "2.5.0"
 time = { workspace = true, features = ["formatting", "macros"] }
 anyhow = "1"
 clap = { version = "=4.4.18", features = ["derive"] }
-tempdir = "0.3.7"
+tempfile = "3"
 
 [features]
 aes-crypto = ["aes", "constant_time_eq", "hmac", "pbkdf2", "sha1", "rand", "zeroize"]

benches/read_metadata.rs

@@ -5,7 +5,7 @@ use std::io::{self, prelude::*, Cursor};
 
 use bencher::Bencher;
 use getrandom::getrandom;
-use tempdir::TempDir;
+use tempfile::TempDir;
 use zip::write::SimpleFileOptions;
 use zip::{result::ZipResult, CompressionMethod, ZipArchive, ZipWriter};
 
@@ -102,7 +102,7 @@ fn parse_stream_archive(bench: &mut Bencher) {
     let bytes = generate_random_archive(STREAM_ZIP_ENTRIES, STREAM_FILE_SIZE).unwrap();
 
     /* Write to a temporary file path to incur some filesystem overhead from repeated reads */
-    let dir = TempDir::new("stream-bench").unwrap();
+    let dir = TempDir::with_prefix("stream-bench").unwrap();
     let out = dir.path().join("bench-out.zip");
     fs::write(&out, &bytes).unwrap();
 
@@ -116,11 +116,27 @@ fn parse_stream_archive(bench: &mut Bencher) {
     bench.bytes = bytes.len() as u64;
 }
 
+fn parse_large_non_zip(bench: &mut Bencher) {
+    const FILE_SIZE: usize = 17_000_000;
+
+    // Create a large file that doesn't have a zip header (generating random data _might_ make a zip magic
+    // number somewhere which is _not_ what we're trying to test).
+    let dir = TempDir::with_prefix("large-non-zip-bench").unwrap();
+    let file = dir.path().join("zeros");
+    let buf = vec![0u8; FILE_SIZE];
+    fs::write(&file, &buf).unwrap();
+
+    bench.iter(|| {
+        assert!(zip::ZipArchive::new(std::fs::File::open(&file).unwrap()).is_err());
+    })
+}
+
 benchmark_group!(
     benches,
     read_metadata,
     parse_archive_with_comment,
     parse_zip64_archive_with_comment,
     parse_stream_archive,
+    parse_large_non_zip,
 );
 benchmark_main!(benches);

src/read.rs

@@ -1603,7 +1603,7 @@ mod test {
     use crate::CompressionMethod::Stored;
     use crate::{ZipArchive, ZipWriter};
     use std::io::{Cursor, Read, Write};
-    use tempdir::TempDir;
+    use tempfile::TempDir;
 
     #[test]
     fn invalid_offset() {
@@ -1805,7 +1805,7 @@ mod test {
         v.extend_from_slice(include_bytes!("../tests/data/symlink.zip"));
         let mut reader = ZipArchive::new(Cursor::new(v)).unwrap();
         assert!(reader.by_index(0).unwrap().is_symlink());
-        let tempdir = TempDir::new("test_is_symlink")?;
+        let tempdir = TempDir::with_prefix("test_is_symlink")?;
         reader.extract(&tempdir).unwrap();
         assert!(tempdir.path().join("bar").is_symlink());
         Ok(())

src/read/lzma.rs

@@ -1,6 +1,6 @@
 use lzma_rs::decompress::{Options, Stream, UnpackedSize};
 use std::collections::VecDeque;
-use std::io::{BufRead, Read, Result, Write};
+use std::io::{BufRead, Error, ErrorKind, Read, Result, Write};
 
 const OPTIONS: Options = Options {
     unpacked_size: UnpackedSize::ReadFromHeader,
@@ -29,7 +29,11 @@ impl<R: Read> LzmaDecoder<R> {
 
 impl<R: BufRead> Read for LzmaDecoder<R> {
     fn read(&mut self, buf: &mut [u8]) -> Result<usize> {
-        let mut bytes_read = self.stream.get_output_mut().unwrap().read(buf)?;
+        let mut bytes_read = self
+            .stream
+            .get_output_mut()
+            .ok_or(Error::new(ErrorKind::InvalidData, "Invalid LZMA stream"))?
+            .read(buf)?;
         while bytes_read < buf.len() {
             let compressed_bytes = self.compressed_reader.fill_buf()?;
             if compressed_bytes.is_empty() {

src/write.rs

@@ -1292,11 +1292,7 @@ impl<W: Write + Seek> ZipWriter<W> {
     ///     Ok(())
     /// }
     /// ```
-    pub fn raw_copy_file_rename<S: ToString>(
-        &mut self,
-        mut file: ZipFile,
-        name: S,
-    ) -> ZipResult<()> {
+    pub fn raw_copy_file_rename<S: ToString>(&mut self, file: ZipFile, name: S) -> ZipResult<()> {
         let mut options = SimpleFileOptions::default()
             .large_file(file.compressed_size().max(file.size()) > spec::ZIP64_BYTES_THR)
             .last_modified_time(
@@ -1308,7 +1304,15 @@ impl<W: Write + Seek> ZipWriter<W> {
             options = options.unix_permissions(perms);
         }
         Self::normalize_options(&mut options);
+        self.raw_copy_file_rename_internal(file, name, options)
+    }
 
+    fn raw_copy_file_rename_internal<S: ToString>(
+        &mut self,
+        mut file: ZipFile,
+        name: S,
+        options: SimpleFileOptions,
+    ) -> ZipResult<()> {
         let raw_values = ZipRawValues {
             crc32: file.crc32(),
             compressed_size: file.compressed_size(),
@@ -1364,6 +1368,47 @@ impl<W: Write + Seek> ZipWriter<W> {
         self.raw_copy_file_rename(file, name)
     }
 
+    /// Add a new file using the already compressed data from a ZIP file being read and set the last
+    /// modified date and unix mode. This allows faster copies of the `ZipFile` since there is no need
+    /// to decompress and compress it again. Any `ZipFile` metadata other than the last modified date
+    /// and the unix mode is copied and not checked, for example the file CRC.
+    ///
+    /// ```no_run
+    /// use std::io::{Read, Seek, Write};
+    /// use zip::{DateTime, ZipArchive, ZipWriter};
+    ///
+    /// fn copy<R, W>(src: &mut ZipArchive<R>, dst: &mut ZipWriter<W>) -> zip::result::ZipResult<()>
+    /// where
+    ///     R: Read + Seek,
+    ///     W: Write + Seek,
+    /// {
+    ///     // Retrieve file entry by name
+    ///     let file = src.by_name("src_file.txt")?;
+    ///
+    ///     // Copy the previously obtained file entry to the destination zip archive
+    ///     dst.raw_copy_file_touch(file, DateTime::default(), Some(0o644))?;
+    ///
+    ///     Ok(())
+    /// }
+    /// ```
+    pub fn raw_copy_file_touch(
+        &mut self,
+        file: ZipFile,
+        last_modified_time: DateTime,
+        unix_mode: Option<u32>,
+    ) -> ZipResult<()> {
+        let name = file.name().to_owned();
+        let mut options = SimpleFileOptions::default()
+            .large_file(file.compressed_size().max(file.size()) > spec::ZIP64_BYTES_THR)
+            .last_modified_time(last_modified_time)
+            .compression_method(file.compression());
+        if let Some(perms) = unix_mode {
+            options = options.unix_permissions(perms);
+        }
+        Self::normalize_options(&mut options);
+        self.raw_copy_file_rename_internal(file, name, options)
+    }
+
     /// Add a directory entry.
     ///
     /// As directories have no content, you must not call [`ZipWriter::write`] before adding a new file.

tests/extract_symlink.rs

@@ -2,13 +2,13 @@
 #[cfg(all(unix, feature = "_deflate-any"))]
 fn extract_should_respect_links() {
     use std::{fs, io, path::PathBuf, str::FromStr};
-    use tempdir::TempDir;
+    use tempfile::TempDir;
     use zip::ZipArchive;
 
     let mut v = Vec::new();
     v.extend_from_slice(include_bytes!("data/pandoc_soft_links.zip"));
     let mut archive = ZipArchive::new(io::Cursor::new(v)).expect("couldn't open test zip file");
-    let temp_dir = TempDir::new("pandoc_soft_links").unwrap();
+    let temp_dir = TempDir::with_prefix("pandoc_soft_links").unwrap();
     archive.extract(&temp_dir).unwrap();
 
     let symlink_path = temp_dir.path().join("pandoc-3.2-arm64/bin/pandoc-lua");

tests/repro_old423.rs

@@ -2,11 +2,11 @@
 #[test]
 fn repro_old423() -> zip::result::ZipResult<()> {
     use std::io;
-    use tempdir::TempDir;
+    use tempfile::TempDir;
     use zip::ZipArchive;
 
     let mut v = Vec::new();
     v.extend_from_slice(include_bytes!("data/lin-ub_iwd-v11.zip"));
     let mut archive = ZipArchive::new(io::Cursor::new(v)).expect("couldn't open test zip file");
-    archive.extract(TempDir::new("repro_old423")?)
+    archive.extract(TempDir::with_prefix("repro_old423")?)
 }